Joachim von Braun

Franz W. Gatzweiler Editors

Marginality
Addressing the Nexus of Poverty,
Exclusion and Ecology
Marginality
Joachim von Braun • Franz W. Gatzweiler
Editors

Marginality
Addressing the Nexus of Poverty,
Exclusion and Ecology
Editors
Joachim von Braun Franz W. Gatzweiler
Center for Development Research Center for Development Research
University of Bonn University of Bonn
Bonn, Germany Bonn, Germany

ISBN 978-94-007-7060-7 ISBN 978-94-007-7061-4 (eBook)

DOI 10.1007/978-94-007-7061-4
Springer Dordrecht Heidelberg New York London

Library of Congress Control Number: 2013946367

© The Editor(s) (if applicable) and the Author(s) 2014. The book is published with open access at
SpringerLink.com.
Open Access This book is distributed under the terms of the Creative Commons Attribution
Noncommercial License which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.
All commercial rights are reserved by the Publisher, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, re-use of illustrations, recitation, broadcasting,
reproduction on microfilms or in any other way, and storage in data banks. Duplication of this publication
or parts thereof is permitted only under the provisions of the Copyright Law of the Publisher’s location,
in its current version, and permission for commercial use must always be obtained from Springer.
Permissions for commercial use may be obtained through RightsLink at the Copyright Clearance
Center. Violations are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for
any errors or omissions that may be made. The publisher makes no warranty, express or implied, with
respect to the material contained herein.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

Foreword on “Marginality and Development”

Development is a process through which an individual, a community or a nation

can progress economically, socially, culturally, psychologically and spiritually.
The centre of any development is the human being, and human wellbeing is the
ultimate goal of development.
Major development concerns worldwide are eradication of poverty; improvement
of health, nutrition and educational status of the people; removal of discrimination
along gender, racial, and ethnic lines; and protection of environment for ensuring
inter-generational equity. These are concerns that have figured prominently in setting
the millennium development goals. Significant progress has been achieved in the
reduction of income poverty in the developing world through economic progress,
targeted safety nets and social protection programs. But economic growth and
targeted transfers may not be sufficient to achieve development. We must address
structural factors such as ecological vulnerability, discrimination and deprivation of
rights caused by poor governance and other socio-political and cultural drivers, and
forces of ecological diversity and climatic change that cause vulnerability and erode
the resource base for livelihoods of the people, particularly for those who are at the
margin of subsistence. All this calls for a broader perspective on development
policies and programs through which the nexus of poverty, discrimination and
environmental degradation can be addressed. It points to the need of understanding
the increasingly complex drivers of persistent poverty at societal and ecological
margins. This is where the marginality concept and framework as introduced by
this book become extremely relevant to understand the complexities of under-
development and poverty.
Amartya Sen has influenced recent development thinking by introducing the
concept of social exclusion, a subset of the concept of marginality, the theme of this
book. Social exclusion widens the concept of poverty and points to the need of
formulating development policies and programs beyond the narrow goals of
eradicating income poverty that many nation states have been pursuing. The UNDP
has taken the concept forward by articulating Human Development Index, a measure
of multi-dimensional poverty that incorporates deprivations in the dimensions of
health, education and living standards.

v
vi Foreword on “Marginality and Development”

The governments and the civil society organizations have already started designing
and implementing programs to address the issues of marginality. Realizing that the
pockets of extreme poverty remain in the ecologically marginal areas throughout
the world, the international agricultural research centers under the Consultative
Group of International Agricultural Research (CGIAR) system have recently
diverted their attention to developing improved agricultural technologies for these
areas. Notable progress has been made in developing innovations for marginal
ecologies under climate stress, such as submergence-tolerant and salt-tolerant
rice and heat-tolerant wheat varieties, and efforts are underway to incorporate
drought-tolerance and cold-tolerance in many crops. Many NGOs, both international
and national, have been addressing the issues of gender inequality and injustices,
insecurity and poor governance, and education, health and water and sanitation in
hardest to reach geographies and communities (such as indigenous and tribal
people) that the governments have found difficult to reach or to address through
mainstream development interventions. Some of them have extended their operations
by establishing social enterprises or businesses to provide services to people at the
bottom of the pyramid that the corporate sector has not been able to reach adequately,
and to overcome problems of missing markets or market failure.
This book uses the marginality lens to understand and explore the links between
poverty, exclusion and ecology. It provides a conceptual framework of marginality
and explains how it links with the concepts of poverty and underdevelopment.
It looks at marginality from the ecological perspectives and revisits basic economics
to understand social exclusion. It maps the physical locations and societal positions
of marginalized people and communities, and traces formal and informal rules,
regulations and institutions that govern access to land, water and biological resources
fundamental to operations of the livelihoods of their choice. A section of the
book addresses the environmental drivers of marginality, understanding the interface
between land degradation and poverty, taking into account the role of nature in
providing livelihoods opportunity to the extreme poor. The authors emphasize that
resource endowments and land degradation are critical determinants of poverty, and
identify agricultural development strategies in a framework of different opportunity
costs of land and labor. They point to the need for accelerating investment in targeted
R&D focusing on crops and traits that are important for the marginal ecologies to
address the environmental limitations that the poor face in these regions. With recent
advancement in biological sciences and invention of tools and techniques in the field
of biotechnology, the probability of research success in these difficult domains of
technological advancement has improved greatly in recent years.
The book then reviews recent attempts to address marginality through development
interventions in selected countries in Asia and Africa through a number of case
studies. This part of the book throws light on the complexities of marginality and social
exclusion, and difficulties to address them. The degree of pro-poor income growth is
compared between various agricultural and non-agricultural livelihoods, rural and
urban areas, and among ethnic caste and religious groups. The development actors of
other countries can substantially gain from the experiences and lessons learnt from
these cases of development interventions already undertaken to address marginality.
Foreword on “Marginality and Development” vii

Marginality is not a matter exclusive to public policy and rights, but also an issue
that can be addressed by business and civil society organizations. The role of the
corporate sector in combating marginality and addressing the demand for service of
the bottom of the pyramid are reviewed and discussed in the last part of the book.
This part also reviews the experience and effectiveness of social protection measures
taken by the government and civil society organizations, particularly conditional
cash transfer programs that are increasingly becoming part of the national poverty
reduction and development strategies to address market failure for the poorest.
This book provides new evidence and constructive insights in support of a vision
of a world free from exploitation and discrimination where everyone has the opportu-
nity to realize their potential, as we pursue it at BRAC, and to empower people and
communities in situations of poverty, illiteracy, disease and social injustice. Our
interventions at BRAC aim to achieve large scale, positive changes through economic
and social programs that enable men and women to realize their potential.
In short, the book breaks new grounds in holistic thinking and understanding of
underdevelopment and persistent poverty. It extends the concept of social protection
to include the impact of adverse ecological environments and climate on the
wellbeing of poor people. Thereby it broadens the concept of poverty and shows the
complexity of the issues in need to be addressed for designing development programs
beyond the narrow objective of eradication of income poverty, and the complementary
roles of government, corporate sector and civil society organizations for tackling the
intractable problems in the last miles of our path to poverty reduction. The book
definitely extends the horizon on existing knowledge of underdevelopment. It will
be very useful to students, policy makers and development actors.

Executive Director Mahabub Hossain

BRAC Dhaka, Bangladesh
Acknowledgements

This volume was made possible through the cooperation among researchers and
scientists from a variety of disciplinary backgrounds, who kindly agreed to come
together to contemplate the problem of “Marginality.” Ecologists, economists, and
other social scientists have rather different ideas and concepts of what marginality
is and of how the processes of marginalization take place. We express our sincere
gratitude to all of the contributors to this volume, not just because they shared their
insights from work on poverty and development—based on decades of work—but
also because all of the contributors were kind enough to join in this effort in order
to reflect in new ways on the nexus of poverty, exclusion, and ecology that is at the
foundation of marginality.
Several sets of discussions lead us towards this volume, including: discussions at
ZEF along with our colleague Paul Vlek who kept emphasizing the societal risks of
land degradation, exchanges with a large network of researchers who work to
address extreme poverty, and intensive exchanges on this topic with Prabhu Pingali
and Monika Zurek at the Bill and Melinda Gates Foundation. All of these were
helpful in shedding light on and clarifying the multifaceted marginality issues, and
for identifying windows of opportunity for actions to address marginalization.
An exciting international conference on marginality held at the Center for
Development Research (ZEF) at the University of Bonn in 2011 was made possible
due to generous financial support provided by the Bill and Melinda Gates Foundation
and the Volkswagen Foundation. Workshop participants—being researchers and
practitioners—critically reviewed the presented papers at that stage and identified
gaps, which we subsequently tried to fill in the preparation of this volume. Moreover,
helpful critiques from three external anonymous reviewers on an earlier draft of this
volume are gratefully acknowledged.
We are grateful for the superb editing assistance of Erick H. Baur and acknowledge
his important contributions to this volume.

Joachim von Braun

Franz W. Gatzweiler

ix
Contents

1 Marginality—An Overview and Implications for Policy..................... 1

Joachim von Braun and Franz W. Gatzweiler

Part I Concepts and Theory

2 Marginality—A Framework for Analyzing Causal

Complexities of Poverty .......................................................................... 27
Franz W. Gatzweiler and Heike Baumüller
3 Exclusion and Initiatives to ‘Include’: Revisiting
Basic Economics to Guide Development Practice ................................ 41
Sajjad Zohir
4 Marginality from a Socio-ecological Perspective ................................. 57
Daniel Callo-Concha, Jan Henning Sommer, Janina Kleemann,
Franz W. Gatzweiler, and Manfred Denich

Part II Dimensions and Prevalence of Marginality

5 Mapping Marginality Hotspots ............................................................. 69

Valerie Graw and Christine Husmann
6 The Poorest: Who and Where They Are? ............................................. 85
Akhter U. Ahmed, Ruth Vargas Hill, and Farria Naeem
7 Targeting the Poorest and Most Vulnerable: Examples
from Bangladesh ..................................................................................... 101
Nusha Yamina Choudhury and Christa Räder
8 Correlates of Extreme Poverty in Rural Ethiopia................................ 117
Degnet Abebaw and Assefa Admassie

xi
xii Contents

9 Examining the Circle of Attachment Trauma, Shame,

and Marginalization: The Unheard Voices
of Young Kutchi Girls ............................................................................. 133
Manasi Kumar

Part III Environmental Drivers of Marginality

10 Poverty, Agriculture and the Environment: The Case

of Sub-Saharan Africa ............................................................................ 151
Prabhu Pingali, Kate Schneider, and Monika Zurek
11 The Marginal Poor and Their Dependence
on Ecosystem Services: Evidence from South Asia
and Sub-Saharan Africa ......................................................................... 169
Pushpam Kumar and Makiko Yashiro
12 Land Degradation, Poverty and Marginality ....................................... 181
Nicolas Gerber, Ephraim Nkonya, and Joachim von Braun

Part IV Experiencing Marginality in Africa and Asia

13 Tackling Social Exclusion and Marginality for Poverty

Reduction: Indian Experiences.............................................................. 205
Sukhadeo Thorat
14 Consumption Behavior of the Poorest and Policy
Implications in Indonesia ....................................................................... 221
Evita Hanie Pangaribowo
15 Addressing Extreme Poverty and Marginality: Experiences
in Rural China ......................................................................................... 239
Ling Zhu
16 Experiences in Targeting the Poorest: A Case Study
from Bangladesh ..................................................................................... 257
Syed Masud Ahmed
17 Rural Poverty and Marginalization in Ethiopia: A Review
of Development Interventions ................................................................ 269
Assefa Admassie and Degnet Abebaw

Part V Responses to Marginality at Different Levels:

State, Business, and Community

18 Macro, Fiscal and Decentralization Options to Address

Marginality and Reach the Extremely Poor ......................................... 303
S. Ehtisham Ahmad
Contents xiii

19 Social Protection, Marginality, and Extreme Poverty:

Just Give Money to the Poor? ................................................................ 315
David Hulme, Joseph Hanlon, and Armando Barrientos
20 Innovative Business Approaches for the Reduction
of Extreme Poverty and Marginality?................................................... 331
Heike Baumüller, Christine Husmann, and Joachim von Braun
21 Business Initiatives That Overcome Rural Poverty
and Marginality Through Creating Shared Value ............................... 353
Niels Christiansen
22 The Marginalized and Poorest in Different Communities
and Settings of Ethiopia.......................................................................... 365
Tadesse Woldemariam Gole, Fite Getaneh Ilfata,
Motuma Tafa, and Aleka Aregachew

Index ................................................................................................................. 383

Contributors

Degnet Abebaw Ethiopian Economic Policy Research Institute, Ethiopian

Economics Association, Addis Ababa, Ethiopia
Assefa Admassie Ethiopian Economic Policy Research Institute, Ethiopian
Economics Association, Addis Ababa, Ethiopia
S. Ehtisham Ahmad Center for Development Research (ZEF), University of Bonn,
Bonn, Germany
Asia Research Center, London School of Economics, London, UK
Akhter U. Ahmed International Food Policy Research Institute (IFPRI), Dhaka,
Bangladesh
Syed Masud Ahmed BRAC Research and Evaluation Division, Dhaka, Bangladesh
Aleka Aregachew Environment and Coffee Forest Forum (ECFF), Addis Ababa,
Ethiopia
Armando Barrientos Brooks World Poverty Institute, University of Manchester,
Manchester, UK
Heike Baumüller Center for Development Research (ZEF), University of Bonn,
Bonn, Germany
Daniel Callo-Concha Center for Development Research (ZEF), University of
Bonn, Bonn, Germany
Nusha Yamina Choudhury World Food Programme, Dhaka, Bangladesh
Niels Christiansen Creating Shared Value Advisors, Lutry, Switzerland
Manfred Denich Center for Development Research (ZEF), University of Bonn,
Bonn, Germany
Franz W. Gatzweiler Center for Development Research (ZEF), University of
Bonn, Bonn, Germany

xv
xvi Contributors

Nicolas Gerber Center for Development Research (ZEF), University of Bonn,

Bonn, Germany
Tadesse Woldemariam Gole Environment and Coffee Forest Forum (ECFF),
Addis Ababa, Ethiopia
Valerie Graw Center for Development Research (ZEF), University of Bonn, Bonn,
Germany
Joseph Hanlon Visiting Senior Fellow, Department of International Development,
London School of Economics, UK
Honorary Research Fellow, School of Environment and Development, University of
Manchester, UK
Ruth Vargas Hill International Food Policy Research Institute (IFPRI), Washington,
DC, USA
Mahabub Hossain BRAC, Dhaka, Bangladesh
David Hulme Brooks World Poverty Institute, University of Manchester,
Manchester, UK
Christine Husmann Center for Development Research (ZEF), University of Bonn,
Bonn, Germany
Fite Getaneh Ilfata Environment and Coffee Forest Forum (ECFF), Addis Ababa,
Ethiopia
Janina Kleemann Center for Development Research (ZEF), University of Bonn,
Bonn, Germany
Manasi Kumar Department of Psychiatry, College of Health Sciences, University
of Nairobi, Nairobi, Kenya
Pushpam Kumar Ecosystem Services Economics Unit, Division of Environmental
Policy Implementation, United Nations Environment Programme, Nairobi, Kenya
Farria Naeem International Food Policy Research Institute (IFPRI), Dhaka,
Bangladesh
Ephraim Nkonya International Food Policy Research Institute (IFPRI), Washington,
DC, USA
Evita Hanie Pangaribowo Center for Development Research (ZEF), University
of Bonn, Bonn, Germany
Department of Environmental Geography, Gadjah Mada University, Yogyakarta,
Indonesia
Prabhu Pingali Bill and Melinda Gates Foundation, Seattle, WA, USA
Christa Räder World Food Programme, Dhaka, Bangladesh
Kate Schneider Bill and Melinda Gates Foundation, Seattle, WA, USA
Contributors xvii

Jan Henning Sommer Center for Development Research (ZEF), University of

Bonn, Bonn, Germany
Motuma Tafa Environment and Coffee Forest Forum (ECFF), Addis Ababa,
Ethiopia
Sukhadeo Thorat Centre for the Study of Regional Development, School of Social
Sciences, Jawaharlal Nehru University, New Delhi, India
Joachim von Braun Center for Development Research (ZEF), University of Bonn,
Bonn, Germany
Makiko Yashiro Ecosystem Services Economics Unit, Division of Environmental
Policy Implementation, United Nations Environment Programme, Nairobi, Kenya
Ling Zhu Institute of Economics, Chinese Academy of Social Sciences, Beijing,
China
Sajjad Zohir Economic Research Group, Dhaka, Bangladesh
Monika Zurek Bill and Melinda Gates Foundation, Seattle, WA, USA
Chapter 1
Marginality—An Overview and Implications
for Policy

Joachim von Braun and Franz W. Gatzweiler

Abstract The marginality concept calls for the integration of poverty concepts
with those of social exclusion, geography, and ecology. The difficulties in reaching
people at the margins of systems are explained by a set of distances, (i.e., physical
distances such as being located in remote or harsh environments), social distances
(being excluded, discriminated against, or not having rights or access to services or
opportunities), but may also be related to technological and institutional infrastruc-
ture deficiencies. This chapter provides an overview of the concept of marginality
and offers a synthesis of the findings of all the chapters in this volume. A review of
policies intended to reduce marginality suggests that none of the marginality deter-
minants need to be accepted as long term. Coherent policies and actions, however,
need to match the systemic causality of marginality in order to be effective.

Keywords Marginality • Poverty • Economic policy • Exclusion • Ecology •

Environment

1.1 Why Focus on Marginality?

There has been significant progress in the reduction of poverty in the developing
world over the past few decades. The prevalence of income poverty defined at
US$1.25/day per capita declined from 43 to 22 % from 1990 to 2008 according to
information presented in Ahmed et al. (Chap. 6 this volume). This progress is the
result of various factors, including economic growth reaching the poor and in
many countries there has also been increased attention to social protection

J. von Braun (*) • F.W. Gatzweiler

Center for Development Research (ZEF), University of Bonn,
Walter Flex Strasse 3, 53113 Bonn, Germany
e-mail: [email protected]; [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 1
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_1,
© The Author(s) 2014
2 J. von Braun and F.W. Gatzweiler

policies. Most of this progress was homegrown, but both factors—growth and
social protection—were somewhat assisted by development aid (Sachs 2006).
A simplistic extrapolation of the declining trend in overall poverty prevalence by
about one percentage point per annum over the past 20 years would automatically
misguide us with the expectation that absolute poverty could end within two decades.
However, it would be more realistic to assume that any further reduction of the
remaining poverty will be more protracted. The diversity of people far below that
income poverty line (i.e., the extremely poor) is high. Economic growth alone may
contribute less to poverty reduction at societal margins, both at the bottom end of
the income distribution and in geographically remote areas. Examples of the latter
are revealed already by the persistence of poverty in marginal areas of China and
Indonesia (see Zhu Chap. 15 and Pangaribowo Chap. 14 this volume). More than
half of the world’s poor now live in large, emerging economies that happen to be
members of the economically leading G20 nations. The bottom billion has shifted
and no longer only lives in the poorest and often fragile states (Collier 2007).
On the other hand, the capacities to design and implement social protection
policies have become more widespread over the past two decades and will assist
with effective poverty reduction, even among the extremely poor (von Braun
et al. 2009). Understanding the constraints and behavior of the marginalized
poor is essential for effective program designs, and there has been research
progress in that domain too, partly through randomized control trials of large
and small development investments (Adato and Hoddinott 2010; Banerjee and
Duflo 2011). Addressing extreme poverty effectively is, however, not just a mat-
ter of growth and targeted transfer policies, but also a matter of addressing
structural forces such as exclusion, discrimination, and the deprivation of rights;
constrained access to services and technology; governance deficiencies and cor-
ruption; and the forces of ecological change that are increasing the vulnerability
and eroding the resilience of the poor, many of whom depend on natural
resources at the margins in rural areas or live in high risk margins of urban
areas. Some of these issues can be further illuminated by experimental research,
but many are not open to that possibility due to the rigidities and scale of some
of these structural forces.
A broader perspective regarding poverty reduction and development policies and
programs is called for, where the nexus of poverty, environmental change, and
development is addressed. This entails understanding the increasingly context-
specific determinants of the serious poverty problems at societal and environmental
margins. This is where the marginality concept and framework come into play. The
set of actors with the potential for engaging in overcoming marginality has become
more diverse and promising in strength. Besides government (central and increas-
ingly local) and civil society actors (which in addition to charities, are often more
focused on developing capacity, skills, and rights), the private sector is much more
deeply involved along all elements of value chains and in service delivery (not just
with corporate social responsibility approaches, but also with inclusive business and
shared value approaches).
1 Marginality—An Overview and Implications for Policy 3

1.2 The Concept of Marginality

“Marginality” is the position of people on the edges, preventing their access to

resources and opportunities, freedom of choices, and the development of personal
capabilities. Being excluded, not only from growth but also from other dimensions
of developmental and societal progress, is an indication of the extremely poor being
at the margins of society and in many cases marginality is a root cause of poverty
(von Braun et al. 2009). Depending on which aspects of marginality are being con-
sidered to explain extreme poverty, millions of people, especially in Sub-Saharan
Africa (SSA) and South Asia, belong to those who are marginalized and living in or
coming from marginal areas.

1.2.1 Definition of Marginality

We define marginality as “an involuntary position and condition of an individual

or group at the margins of social, political, economic, ecological, and biophysi-
cal systems, that prevent them from access to resources, assets, services, restrain-
ing freedom of choice, preventing the development of capabilities, and eventually
causing extreme poverty” (Gatzweiler et al. 2011, 3; also see Gatzweiler and
Baumüller Chap. 2 this volume). The marginality perspective encompasses those
poor who are below certain thresholds and outside mainstream socio-economic
and human geographical systems, where improved access to rights, resources,
and services would help enable decent standards of living. With reference to
biological systems, marginality describes the state of organisms outside optimal
homeostatic ranges which are necessary for living systems (Damasio 2011; also
see Callo-Concha et al., Chap. 4 this volume).
As the meaning of the word indicates, marginality is a relative concept that refers
to where people are and to what they have. “Where people are” not only refers to
physical locations, but also to societal positions. “What they have” refers to capital
assets—using a broad definition of capital—and the rules and regulations (both for-
mal and informal) that enable access to these assets and their use. For example,
ethnic minorities in rural Africa or Asia may live in geographically remote areas and
be perceived as socially marginal by their respective national governments.
Nevertheless, according to traditional institutions they may have user rights and
access to land, water, and biological resources (e.g., forests) that enable them to
function in the livelihoods of their choice. By disabling them of traditional access
and rights to use these resources (e.g., by imposing rules of the nation state, prohib-
iting access to resources, and measuring their well-being in terms of material own-
ership or monetary savings), they become poor and marginalized. The marginalized
poor are those who are affected by both marginalization and poverty. Causalities of
the two concepts are complex (Dasgupta 2009) and the linkages between them need
to be further explored, and that is a task of this book.
4 J. von Braun and F.W. Gatzweiler

1.2.2 Marginality and Poverty

The concept of marginality should not be construed as an alternative to the concept

of poverty; rather these two concepts overlap and are complementary. Marginality
encompasses broad approaches like relative deprivation, social exclusion, or the
capabilities approach. It entails an interdisciplinary and systemic perspective on the
lives of the poor with the aim of revealing the underlying contributors to poverty,
which have their roots in the functioning (or malfunctioning) of economic, socio-
cultural, or ecological systems. This includes taking spatial dimensions such as
geography and location into account. In the words of Dasgupta (2003, 2) “policies
matter, as do institutions, but the local ecology matters too.” Influenced by Sen’s
“capability approach” (1987a, 1993, 1999), which focuses on functions or living
conditions defined as “what we can do or cannot do, can or cannot be” (1987b, 16),
and the ability to achieve them (capabilities), the marginality concept goes beyond
a measurement of well-being in terms of goods or commodities. This concept seeks
to reveal real opportunities or barriers that exist as a result of what people have
(goods, rights, knowledge, and opportunities) and where they are (understood as
their geographical location or their positions within socio-political and economic
systems). Both define their access to resources, services, or decision making. As in
Sen’s capabilities approach, what matters for achieving desired well-being is not
only the amount of what people have and can do, but also how these assets are trans-
formed (by real opportunities) into fully functioning lives that people choose to live
(Nussbaum and Sen 1993; Nussbaum 2004; Schlosberg 2007).
Where these opportunities do not exist or where they are actively constrained,
barriers hinder the empowerment of people and they are constrained from devel-
oping the full potential of their lives. These barriers are constraints or capability
deprivations that can originate from multiple causes rooted in economic, social,
political, or ecological systems, which can also be considered enabling conditions
or freedoms (see Sen 1999).
Within the concept of marginality, the nexus of poverty, environmental change,
and development may be addressed. Marginality is a multidimensional and inter-
disciplinary concept integrating poverty, discrimination, and social exclusion; the
degradation of ecosystem function; and access to services, markets, and technol-
ogy. Marginality incorporates the understanding that subjective perceptions of
poverty, values, and aspirations matter, and that all of these are part of calibrating
the tools of poverty measurement. Apart from being inclusive and interdisciplin-
ary, the concept of marginality offers an integrated and systemic basis for under-
standing the interactions between social and ecological systems. In a world with
fewer natural resources for all, the linkages and changing patterns between both
kinds of systems become more apparent, and the role of access to services and
technologies ever more critical.
“De-marginalizing” the marginalized requires the creation of the physical
infrastructure and institutional arrangements that can help to overcome the
1 Marginality—An Overview and Implications for Policy 5

barriers to access, exchange, and communication, and facilitate a shift away from
the margins of development through building accessible assets beyond natural
capital (i.e., access to services that foster human capital and technology), while
including the marginalized in the process. The ability to live a life to its full poten-
tial, or to achieve the living conditions of one’s choice, can be constrained by
factors stemming from social or ecological systems. If access is denied to land,
forest, water or the benefits they provide (e.g., habitat, nutrient and hydrological
cycling, soil fertility, carbon sequestration), then we speak of constraints or barri-
ers to the delivery of ecosystem goods and services.
The concept of marginality facilitates an understanding of the underlying sys-
temic contributors of poverty and exclusion that can overlap with the lack of
resources and opportunities needed to achieve the desired conditions in life.
Although the income poverty approach is easy to conceptualize and measure (in
its simplest form the headcount ratio counts the proportion of people living under
a particular poverty line), it has been criticized because it does not relate to mul-
titudinous concepts of individual need “or to any agreed definition of what it is
to be poor” (Gordon et al. 2000, 8). Relative deprivation (RD) measures build on
the idea that the value of objective circumstances depends on subjective compari-
sons (Townsend 1979; Stark and Bloom 1985; Walker and Smith 2002; Wilkinson
and Pickett 2007; Stark et al. 2011). The RD concept adds a whole new dimen-
sion to absolute poverty concepts, because it refers to the circumstances under
which comparisons are made and does not define a general threshold. The RD
approach is helpful, as it differs from traditional poverty concepts in that it
defines the relative level of deprivation of material or social assets and condi-
tions, whereas poverty is understood as “lacking or being denied the resources to
obtain these conditions of life” (Townsend et al. 1987, 85; as cited in Sauders
1994, 235–236).
The conceptual foundations of marginality are elaborated further in Chap. 2
(this volume) by Gatzweiler and Baumüller. The authors review the distinctions
between frameworks, theories, and models, and highlight that marginality is
best perceived as a framework within which different sets of theories and mod-
els can be usefully applied and tested. Reducing the diversity of poverty to a few
indicators entails the risk of overlooking critical features and causalities under-
lying poverty as analyzed from a systems perspective. Gatzweiler and Baumüller
review the evolution of the marginality concept, as well as classes of poverty
theories. In practical ways, they illustrate situations of the marginality of an
actor (a person or group) by a specific position within the multiple dimensions
(e.g., social, economic, political, nutritional, educational, geographical) of peo-
ples’ lives in which they are more or less marginalized. They further refine the
conceptual framework for investigating marginality by distinguishing “biophys-
ical causality clusters” from “societal causality clusters” as enabling or con-
straining factors that lead to outcomes of well-being or poverty, and changes in
positions and conditions being shaped by feedback to the factors and causality
clusters.
6 J. von Braun and F.W. Gatzweiler

The social exclusion approach overlaps with the concept of marginality when it is
seen as a condition and process of “becoming detached from the organizations and
communities of which the society is composed and from the rights and obligations
that they embody” (Room 1995, 243). In Chap. 3 (this volume) Zohir emphasizes
that in relation to exclusion:
In all development endeavors over the last two decades or more, either in the guise of
Poverty Reduction Strategy (PRS) and Millennium Development Goals (MDGs), or to
achieve greater equity and foster inclusive growth, the general concern was reaching out to
those who are “left out”…[where]…“left out’s” could be children not being sent to schools
by their parents, poor rural women unable to avail themselves of health services or market
opportunities, people from minority or socially outcast groups having little or no access to
jobs and public amenities, and the physically or mentally disabled. We refer to these people
or social groups as “excluded” and to the broader subject of study as “exclusion.”

Both belonging to a group or community and being embedded within an insti-

tutional environment that secures rights and duties are necessary for escaping
marginality, improving function, and achieving freedoms. Basic but crisp eco-
nomic theory is invoked by Zohir to guide development practice toward under-
standing exclusion and initiatives to “include” the marginalized. The relationships
between contracts, goods and services, and exclusion are highlighted, and types
of exclusion are identified with the help of a supply and demand analysis of ser-
vices: voluntary exclusion, exclusion caused by a lack of awareness, exclusion for
survival, exclusion due to a lack of demand, and exclusion caused by “distance”
(social exclusion, poor connectivity). Useful practical insights are gained when
this theoretical analytical framework is applied by Zohir to identify the scope of
different interventions for addressing the different types of exclusion, such as
innovations in production, price subsidies, or transfers.
A socio-ecological perspective of marginality is pursued by Callo-Concha et al.
in Chap. 4 (this volume). The authors highlight that marginality is the norm rather
than an exception in ecological systems. They underline that many complex phe-
nomena are shaped by both social and ecological systems simultaneously and insep-
arably, and therefore the complex issue of marginality can usefully be addressed by
integrated socio-ecological systems approaches. The three conceptual and theoreti-
cal chapters mentioned above provide guidance to many of the more applied chap-
ters of this book. They underline the complementary roles of alternative disciplinary
approaches for addressing marginality and the opportunities for further conceptual
and theoretical research.
Changing perceptions of development and global change call for such a new
approach for addressing poverty through a marginality lens. Throughout the last two
decades the global development landscape has changed dramatically. In many
developing countries high economic growth rates have enabled millions to move out
of income poverty. However; in South Asia and SSA we find countries with high
economic growth rates, but also with a high prevalence of poverty. This is the case
where in many rural settings land is becoming less available; ecosystems often turn
out to be more degraded, while access to drinking water, sanitation, and energy
resources remains constrained for many poor communities.
1 Marginality—An Overview and Implications for Policy 7

1.3 Dimensions and Prevalence of Marginality

1.3.1 Disaggregated Global Views

A book on marginality obviously can be expected to include descriptions of the

magnitude and dimensions of the global marginality problem. In Part II of the
book, two global views on marginality in the context of the developing world
are presented first, one based on geographical information systems (GIS) and
another taking a poverty statistics approach. In Chap. 5 (this volume) Graw and
Husmann offer a comprehensive, multidimensional geographical mapping of mar-
ginality and identify marginality “hotspots” based on carefully selected sets of
indicators at sub-national levels. Thereby the authors provide an original mapping
effort of the global marginality landscape. The developing world is certainly not
one uniform marginality domain. The diverse patterns, especially in Africa and
Asia, are illuminating. Also interesting are the different degrees of overlap between
income poverty and marginality patterns revealed in the marginality hotspot maps
by the authors. The fast moving GIS based research at national and local levels
offers tremendous opportunities for identifying the geographic distribution of mar-
ginality and thereby guide actions towards priority “hotspots.”
In Chap. 6 (this volume) Ahmed et al. also take a disaggregated global view and
identify where the world’s poor and particularly the poorest live, and describe the
extent of recent progress made in poverty reduction and the characteristics of the
poor in selected countries. They found that poverty reduction was quite similar
between the “subjacent” poor relatively closer to the poverty line of US$1.25/day
per capita and the ultra-poor farther below—in fact the ultra-poor were slightly
favored and the authors highlight that it is encouraging to see that the theory of
“poverty traps” may not be holding generally true for those in ultra-poverty in recent
years. Moving from global to specific country levels, marginality dimensions and
patterns were analyzed for Bangladesh and Ethiopia, two countries that are further
explored in the book.
The first MDG called for halving the proportion of the world’s population that
suffers from extreme poverty and hunger by 2015. The hunger objective of this goal
addresses extreme deprivation. Global progress towards this poverty goal has been
significant, but progress on the hunger goal has been much less impressive. As Ahmed
et al. point out in Chap. 6, around 1.3 billion people in the developing world
subsist on less than US$1.25/day and 234 million live on less than US$0.63/day.
Compared to the poor with incomes closer to the former poverty line, the very
poorest are typically also marginalized: they belong to socially excluded groups,
live in remote rural areas, or have less education, fewer assets, and less access to
markets. Ahmed et al. also found that the proportion of the developing world’s
population living on less than US$1.25/day fell from 43.1 % in 1990 (the base
year for the MDGs) to 22.4 % in 2008 (see Chap. 6). This suggests that the poverty
component of the first MDG has virtually been met at the global level before the
prescribed timeline.
8 J. von Braun and F.W. Gatzweiler

Regional progress, however, has been uneven. Whereas the numbers and
proportions of the extremely poor fell in East Asia/the Pacific and in South Asia,
progress in SSA has been very limited. Worldwide the absolute number of under-
nourished people has been growing since the mid-1990s. In 2009 an estimated
870 million people were undernourished (FAO 2012). The majority of the world’s
undernourished people live in developing countries. Most of them are found in
Asia, the Pacific, and SSA, and live in rural areas and often on degraded lands
(Nkonya et al. 2011). Whereas in Africa most of the poor live in low-income
countries, in Asia they are often found in middle-income countries. The bottom
billion no longer lives in the poorest countries. Statistically speaking, the majority
of the poor has shifted within the last 20 years from low to middle income countries
such as India, Indonesia, Pakistan, and China, and often live in marginal regions
of these countries. This has far-reaching implications for cooperative development
policies, as a singular focus on poor countries will miss out on the majority of the
world’s poor people.
Although there has been progress in reducing the number of poor, especially
those just below the US$1.25/day per capita poverty line, the poorest and hungry
have often been left behind. They do not have an effective political or societal voice,
they are often decoupled from market economies (Grant and Shepherd 2008), and
conventional poverty reduction and development programs are poorly designed to
reach them or respond to their needs. Such a situation is extremely inefficient in
macroeconomic terms and unjust from a human rights perspective. Economy-wide
growth often does not reach this group, as they are not linked to (or are excluded
from) the economic processes that otherwise generate and share the benefits of
growth. Policies and practices that redistribute wealth from economic growth are
often not in place. According to Grant and Shepherd the proximate causes for this
pattern can be that the extremely poor:
• live in unfavorable areas (with poor agricultural asset bases, poor or a lack of
transportation infrastructure)
• can (for various reasons) make only minimal use of their labor and lack opportu-
nities to acquire skills
• direct most of their efforts towards achieving caloric and nutritional minimums
• are socially excluded or are of low status in society (e.g., because of the loss of
the income earning household members, disease, childlessness, or being born
into a disenfranchised social class)
• are excluded from public services or poverty reduction efforts.
To date research findings have shown that there is a correlation between extreme
poverty and both geographical remoteness and social exclusion, and that the inci-
dences of extreme poverty and food insecurity are concentrated in remote rural
areas farthest from transportation, public service (health, education), and market
infrastructure. In addition, the poorest often belong to ethnic minorities and other
groups that are socially (Ahmed et al. 2009; Ahmed et al. Chap. 6 this volume).
Marginality research can complement the large body of literature on poverty that is
1 Marginality—An Overview and Implications for Policy 9

based on indisputable facts about those who belong to the poorest and hungry today
(von Braun et al. 2009):
• The poorest are becoming increasingly concentrated in SSA and South Asia.
More than three-quarters of those living on less than US$0.50/day live in SSA
and the proportion of the global poor in this region is increasing.
• Poverty and widespread hunger remain in regions that have experienced rapid
economic growth and substantial reductions in poverty.
• Whereas the number of urban poor is increasing rapidly, the poor are still pre-
dominantly rural. Poverty reduction remains strongly connected to agricultural
development in many countries.
• Poverty and hunger reduction has been slower among the poorest and excluded
groups—ethnic minorities, disadvantaged people, and the disabled—causing
poverty and hunger to be increasingly concentrated in these groups. In addition,
poor women and children are particularly vulnerable to the long-term effects of
poverty and hunger on individual health and education status.
Tracking progress (or the lack thereof) on poverty reduction, including move-
ments below the poverty line at global, national, and sub-national levels with up-to-
date and detailed measurements, remains of great importance. Although the total
number of people in poverty may change little, this stability masks substantial
movements in and out of poverty and population growth. Some that are living above
a poverty line are vulnerable to becoming poorer, and some living below that line
may rise above it. Those living far below a poverty line (i.e., the extremely poor),
will likely be there for longer, perhaps for generations (see Ahmed et al. Chap. 6 this
volume). Tracking poverty as such, however, needs to be integrated with the identi-
fication and tracking of many other aspects of societal, structural, ecological, and
technological changes in order to identify causal mechanisms of poverty and prom-
ising options for remedial action. Such broader attempts are a promise of the mar-
ginality approach pursued in this volume.

1.3.2 Detailed Looks at Marginality: Locations

and Socio-cultural Contexts

In Chap. 7 (this volume) Choudhury and Räder present a comprehensive mapping

effort of the poorest and vulnerable residents of Bangladesh that was prepared for
targeting interventions of the World Food Programme. The programs for social
safety, disaster risk reduction, nutritional improvement, and education were priori-
tized and guided by this mapping effort, bringing mapping and action together in
innovative ways. This offers an approach that could be adapted in other low-income
countries that are under persistent budget constraints.
In Chap. 8 (this volume) Abebaw and Admassie used a host of household demo-
graphic and socioeconomic variables from rural surveys to examine the determi-
nants of extreme poverty in rural Ethiopia, employing econometric estimation to
10 J. von Braun and F.W. Gatzweiler

characterize extreme poverty. The estimation results reveal strong associations

between location and ultra-poverty. The authors confirmed that ultra-poverty in
Ethiopia is positively associated with distance from educational and health facili-
ties, roads, and other socioeconomic infrastructure.
Kumar (Chap. 9 this volume) adds a dimension of insight into marginalization by
offering an interesting psychological perspective of the circle of attachment trauma
leading to shame and self-perpetuated marginalization. The experiences of social
marginality in the lives of young girls from the Indian state of Gujarat highlights
how secure interpersonal attachments help form the primary “capabilities” that have
direct bearing on an individual’s sense of identity and freedom. Insecure interper-
sonal and emotional attachments, particularly dismissing kinds, lead to severe inhi-
bitions in personality development and the accumulation of layers of shame and
self-doubt. Kumar examines how the psyche is tormented by repeated experiences
of social marginalization in the form of dismissal at the hands of family, and how
shame becomes an abiding emotion—creating further doubts, disenfranchisement,
and alienation that permeate the lives of the girls interviewed by the author. Breaking
this cycle of marginality apparently needs a broader and deeper approach, and calls
for further expansion of interdisciplinary efforts in research beyond economics and
ecology.

1.4 Environmental Drivers of Marginality

Ecological and agricultural dimensions and determinants of marginality are the

focus of the chapters in Part III of the book. Pingali et al. (Chap. 10 this volume)
explore the linkages between poverty, agriculture, and the environment in the case
of SSA. They found that natural resource endowments and land degradation are
critical determinants of poverty and marginality, and identify agricultural develop-
ment strategies (distinguishing farming systems, productivity, and environmental
strategy components) in a framework of different opportunity costs of land and labor.
Efforts to reduce the significant agricultural yield gaps with technological and policy
measures in support of sustainable intensification are identified by the authors as
opportunities, even in many marginal areas. They conclude that investing in targeted
Research and Development (R&D), especially focused on the crops and traits that are
important to the poor and the environmental limitations they face, have the potential
to dramatically increase agricultural productivity and lessen marginality in SSA.
Kumar and Yashiro (Chap. 11 this volume) underline the dependence of the mar-
ginal poor upon ecosystem services. In their analyses of conditions in Asia and
Africa they found that the poor are often more vulnerable than others to the loss of
ecosystem functions that restrict the availability of natural goods and the perfor-
mance of services. They also found that those who benefit most from environmental
conservation efforts are not poor, while the poor typically suffer most from environ-
mental degradation. Pointing at further research needs, the authors emphasize that
in all of the reviewed cases of poverty-ecosystem services linkages, indicators of the
intensity and directionality of linkages were either diffuse or incoherent. Information
1 Marginality—An Overview and Implications for Policy 11

is particularly limited on how changes in regulatory ecosystem services, many of

which are critical for supporting the lives of the poor, affect human well-being.
Gerber et al. (Chap. 12 this volume) investigated the relationships between
poverty and land degradation (LD) at the global level, and also in a set of develop-
ing country cases. They suggest a broader context for the assessment of LD, its
causes, consequences, and costs, where the proximate causes of LD include: bio-
physical factors (climatic conditions, topography, etc.) and anthropogenic factors
(unsustainable land management practices such as forest clearing, over grazing,
etc.). Underlying causes of LD include the policies and institutional or other
socioeconomic factors that determine land management practices. For example,
poverty contributes to the failure to invest in sustainable land management prac-
tices. Similarly, policies that enhance investment in land management, such as
payments for ecosystem services that result in reforestation efforts on steep
slopes, can mitigate the proximate causes of LD in the form of soil erosion. There
are feedback effects between poverty and LD that make it harder to determine the
direction of the causality, and the institutional frameworks within which land
users operate and make their land management decisions can supersede the
impacts of poverty on LD. This was exemplified in some of the case studies exam-
ined by the authors, who propose a systematic and science-based assessment of
the extent, severity, and costs of LD worldwide (Nkonya et al. 2011) as a first step
towards the inclusion of LD into general measures of human well-being and of
LD remediation efforts into strategies for the reduction of marginality.

1.5 Experiencing Marginality in Africa and Asia

In Part IV of this book the causes and experiences of marginality are explored in
countries of Asia and Africa. Entire country, sector, and spatial dimensions, and
also household behaviors are explored, and experiences with policies and interven-
tion programs are evaluated.
Thorat (Chap. 13 this volume) addresses marginality in the context of social
exclusion and poverty in India. Values, the lack of rights, and discrimination
are discussed as important determinants of marginality. The experiences of
change in poverty in rural India were found to have distinct patterns according
to ethnic, caste, or religious group. The relative degree of “pro-poor” income
growth was compared between various agricultural and nonagricultural liveli-
hoods, rural and urban areas, and among ethnic, caste and religious groups.
Rural self-employed households experienced greater income increases than
wage-laborers across all ethnic, caste, and religious groups. The analyses also
showed that poverty was generally reduced at a lower rate for minorities such as
the “scheduled tribes,” “scheduled castes,” and Muslims. Members of these
groups typically own less agricultural land, have less access to private non-
farming economic activities, are more dependent on wage employment, and
have historically higher levels of poverty. Thorat also identifies policy opportu-
nities: self-employed households may benefit from opportunities for engaging
12 J. von Braun and F.W. Gatzweiler

in micro-scale production and business, whereas wage laborers may benefit

from increases in employment or wages. The author proposes poverty reducing
agricultural growth strategies specifically tailored for rural agricultural and non-
agricultural livelihoods. Off-farm income opportunities, incentives for increas-
ing the profitability of small producers and businesses, and investments in the
education and skill development of minority groups are among the policy port-
folio options suggested by the author.
Consumption behavior of the poorest households in Indonesia was analyzed by
Pangaribowo in Chap. 14 (this volume). The author’s approach is based on the
premise that designing effective food policies and intervention programs for the
poor requires accurate knowledge of their consumption patterns and behavior. For
example, an important consumption behavior pattern observed was an overall
decreasing share of household expenditures on food items over the last decade
except for increasing expenditures on goods from the “dried and snack foods” and
“meat and fish” categories. The author also found that the poor consistently spent
from their meager incomes on so called “adult goods” (i.e., tobacco and alcoholic
products). These consumption patterns may reflect gender biases, with male
household heads deciding to spend more at the margin on these goods. Estimates
of expenditure-price and cross-price elasticity point to the potential effectiveness of
policy interventions for improving food security of the poor. Pangaribowo
found that for the poorest households over the last decade, “dairy products”
remained expenditure elastic, “staple foods” became less expenditure elastic, and
that expenditure elasticity for “alcohol and tobacco products” increased. Similar to
expenditure elasticity, price elasticity for categories such as “dairy products,”
“meat and fish,” as well as “alcohol and tobacco products” predicted the most sub-
stantial decreases in consumption in response to price increases. Complementary
and substitutive relationships among consumed goods categories imply that
“staple foods” consumption among the poor will increase when prices for “meat
and fish” and other nutrient rich products increase. The author concludes that in
order to more effectively address extreme poverty and food security, policies
should integrate information and nutrition education campaigns, and also be targeted
toward women.
China’s recent experiences in addressing extreme poverty and marginality are
presented and analyzed by Zhu in Chap. 15 (this volume). The evolution of effective
and efficient policies, especially in rural areas, was found to be closely connected to
China’s economic growth in the past three decades as well as to regional inequality.
In order to alleviate and eradicate extreme poverty, the Chinese government
extended the Di Bao (minimum livelihood guarantee system) program to the entire
rural sector, enhanced agricultural policies, and stepped up rural socioeconomic
development efforts. In developed regions of the country the Di Bao program, social
insurance, and public services have effectively mitigated extreme poverty. Zhu
reviewed policies that have been introduced and changes that might still be needed,
especially in the rural parts of western China. The author describes the transition of
Di Bao from an emergency relief and basic needs program to a mature social
1 Marginality—An Overview and Implications for Policy 13

protection system, aligning the poverty program for the poor and extremely poor
with efforts in urban areas.
Zhu found that the successes of past anti-poverty programs have mainly benefitted
the poor closest to the national poverty line, whereas the extremely poor generally
remain poor because of their marginalized geographic locations and positions in
society. For them the Di Bao provides a social safety-net. The author identifies
requirements for improving benefits from poverty reduction programs and for
linking other such programs with the Di Bao system. The requirements include
improved targeting mechanisms, data availability, and management capacities,
and also increased funding for intervention programs. One lesson from experiences
in more developed regions in China is to target the specific difficulties of extremely
poor households and individuals, for instance through increased coordination of
the Di Bao social assistance program with other social insurance programs and
public services.
Experiences in targeting the poorest in Bangladesh were analyzed by Ahmed
(Chap. 16 this volume), who in particular looked into the impressive case of the
program “Challenging the Frontiers of Poverty Reduction” (CFPR), designed
and administered by the Bangladeshi nongovernmental organization BRAC
(Bangladesh Rural Advancement Committee). CFPR was designed to enable the
ultra-poor to enter into mainstream economic activities that would generate sus-
tainable income sources. BRAC recognized that existing microfinance efforts
had failed to address the needs of the ultra-poor, either because they lacked
access to microfinance opportunities, or because they lacked the needed human
capital and/or other preconditions for productive engagement. To achieve the
program goal of lifting participants out of extreme poverty within 2 years and
facilitate their entry into mainstream development programs, this program com-
bines interventions specifically tailored for the ultra-poor with interventions to
create an enabling environment for them. Ahmed describes the ideas behind
the approach, the specific steps taken for implementing the program, and the
key success criteria, which include the active participation of local villagers in
identifying the poorest households, tailored training of motivated community
workers, and the verification of collected data.
Policy measures and program interventions to alleviate rural poverty and
marginality, along with trends in general economic indicators in Ethiopia, were
reviewed by Admassie and Adebaw in Chap. 17 (this volume). The authors
provide a comprehensive overview of Ethiopian policy measures and program
interventions that covers food security, social safety nets, resettlement, environ-
mental, health, and education sector policies. Despite overall increasing economic
and particularly agricultural sector growth rates over the last decade, their review
revealed that growth divergence and inequalities in income, health, and education
remain high, and that poverty remains persistent, especially among the most
vulnerable and socially excluded groups of rural Ethiopia. Thus, significant efforts
to increase the equitable distribution of economic opportunities and services are
still needed.
14 J. von Braun and F.W. Gatzweiler

1.6 Responses to Marginality at Different Levels:

State, Business, and Community

1.6.1 Key Roles of Government and Civil Society

Addressing marginality is not only a matter for central and local governments,
but also a task for civil society organizations (see the review of the BRAC experi-
ence in Chap. 16 this volume), business, and local communities themselves. In
Part V of this book, policies and programs, as well as actors addressing marginal-
ity are discussed.
Public policy assessments begin with Ahmad’s work (Chap. 18 this volume) on
macroeconomic, fiscal, and decentralization options to address marginality. How
does one reach the extremely poor and marginal groups in countries where local
politicians and officials may have little incentive to provide for them? The instru-
ments that might be involved include a range of options: transfers or assistance from
higher levels of government (donors), cash support, the provision of public services
(particularly health care and education), forms of employment support, and assis-
tance for small-scale enterprises. The author outlines some of the difficulties
attached to making government sourced cash transfers more effective in reaching
the marginalized poor, such as the lack of political will, leakage of funds and cap-
ture at the local level by vested interests, weak public financial management sys-
tems, the lack of information about target groups, and incomplete knowledge about
beneficiary needs. Ahmad offers a typology of options that is useful for improving
the reach of efforts to provide for the poorest of society and concludes that simple
categorical targeting mechanisms, such as through clinics or schools, can improve
the effectiveness of reaching the marginalized. Good governance efforts to track and
account for funds and their outcomes are among the essential ingredients for poli-
cies to successfully provide for the poorest. In the absence of genuine local interest
in providing support for the marginalized and extremely poor, direct funding to
beneficiaries by central governments and aid agencies appears to be the only feasi-
ble option.
Social protection programs, especially cash transfer programs, have spread
across low-income and middle-income countries over the past two decades and are
increasingly part of national poverty reduction and development strategies. Hulme
et al. (Chap. 19 this volume) pose the question, “Just Give Money to the Poor?” and
discuss insights from long-term research on social protection policies, including
both conditional and unconditional cash transfer programs, which have become
quite popular around the developing world over the past 20 years. The authors pro-
vide a brief history of the development of social protection and cash transfer pro-
grams, which can fill gaps where markets fail, especially with regards to the labor
market and the environment—two crucial sources of livelihood for the marginalized
poor. The authors give examples of the specific goals, targets, and conditions of
various social protection and particularly cash transfer programs from a broad
cross-section of countries, and how they have been perceived by both their
1 Marginality—An Overview and Implications for Policy 15

detractors and supporters. Hulme et al. argue why and how cash transfers can
succeed by trusting the poorest and enabling them to escape poverty, as the poor are
quite capable of identifying what is best for their own well-being. The authors iden-
tify the following five overriding features for social protection programs and cash
transfers to be successful, they should be: fair, assured, practical, large enough to
impact household income, and popular. They stress that these principles need inter-
pretation at the national level, because no “model” can be automatically transferred
from one country to another. The growing body of research on the limitations of
these programs is noteworthy, especially when they constrain the mobility of labor,
as highlighted by Levy (2008) in the case of Mexico.

1.6.2 Potential for Business

Marginality is not a matter exclusive to public policy and rights, but also an issue
that can be addressed by business. The “bottom of the pyramid” approach (i.e., the
latent power of demand by the poor as a business case) has been very much rec-
ognized by now (Prahalad 2004). Innovative business approaches offer new
opportunities to reduce extreme poverty and marginality, and there are promising
new initiatives for overcoming marginality through inclusive business models.
The actual and potential roles of the corporate sector in combatting marginality
are reviewed and discussed by Baumüller et al. (Chap. 20 this volume) and
Christiansen (Chap. 21 this volume).
In Chap. 20 Baumüller et al. outline the rapid evolution of related thinking in the
business world and explore in detail some of the relatively new business approaches
that have emerged for addressing societal problems. The potential for innovative
business approaches to target the poor appears to be receiving increasing attention.
The authors also examine whether and how these approaches can support not only
those living close to a poverty line, but also help engage the marginalized at the low-
est end of the income scale.
A promising new initiative for overcoming marginality through inclusive busi-
ness approaches is the “Creating Shared Value” (CSV) approach, which is described
by Christiansen in Chap. 21. CSV means that when making business decisions on
future products and allocations of investments, companies simultaneously consider
what long-term values can be created both for society and for shareholders. The
author defines criteria for successful shared value approaches and gives a detailed
example of how CSV worked in Nestle’s dairy farmer development approach at the
Moga milk factory in the Punjab region of India and Pakistan. Apart from develop-
ing local diary businesses and providing incomes, a larger section of local society
has benefited from improved infrastructure, schools, sanitary and medical facilities,
and nutrition education programs.
Despite the successful impacts of inclusive business approaches, rural poverty
cannot be sustainably reduced by private business initiatives alone. These initiatives
can achieve greater results when accompanied by public investments in agriculture
16 J. von Braun and F.W. Gatzweiler

and rural development, and other forms of improved cooperation between the public
and private sectors. While the potential for government support to facilitate business
solutions that address extreme poverty and marginality is clear, the forms that this
support should take and the actual benefits of various support measures remain
poorly understood. At times governments provide support measures that reduce
social welfare overall. The example of fiscal incentives is a case in point: while tax
breaks are a popular tool used by governments to try to attract direct foreign invest-
ments into certain areas or sectors, studies from several countries have shown that
these incentives often have had little influence on investors’ decisions, while gov-
ernments lost revenues (Baumüller et al. 2011). Thus research should explore
options for government measures in support of business activities for social devel-
opment and assess their potential impacts in the different contexts in which they are
applied. Indirect approaches for inclusive business and shared value generation that
reach the extremely poor may be of particular significance, such as infrastructure
investments in marginal areas, or access to improved seeds or livestock, health ser-
vices, and nutritionally enhanced foods that have elements of comprehensive cover-
age and do not exclude the poor. Tracing the results and impacts of such investments
on the extremely poor should, however, be integral to such initiatives.
It seems that the boundaries of innovative business operations can be pushed
much further to include a far greater number of the marginalized and extremely
poor. The corporate sector could examine these opportunities as low risk/high return
ventures and continue experimenting. Development partner communities may best
serve these initiatives by providing local insight and co-funding. The research com-
munity may best serve these efforts by considering innovative ideas that foster insti-
tutional arrangements that bring together unusual alliances, by accompanying
efforts with solid impact studies, and through insights from comparative assess-
ments of cases of successful efforts that were scaled up.

1.6.3 Attention to Diverse Local Contexts

As many of the above mentioned chapters of this book underline, involvement

of the marginalized communities themselves is often a critical component of
effectively mitigating marginalization. Gole et al. (Chap. 22 this volume) take a
holistic perspective from the bottom up at the community level. Their insights
are based on a series of interviews in carefully selected communities in mar-
ginal areas of Ethiopia.
In the Jida district in the Central Highlands of Ethiopia, extreme land degrada-
tion on slopes, erratic rainfall, and unsustainable logging practices on the valley
plains have contributed to the marginal conditions in the area. Marginalization here
is predominantly caused by anthropogenic environmental factors and human behav-
ior. The Yayu region in contrast, is endowed with plentiful natural resources. The
lack of formal credit opportunities and the resulting reliance on high-interest per-
sonal loans, shortages of family labor, and the limited size of coffee plantations all
1 Marginality—An Overview and Implications for Policy 17

contributed to the poverty of some community members. Marginalization here is

mainly rooted in institutional factors and resource constraints, especially of land
availability. In the Adami Tullu Jiddo Kombolcha district the potential for income
generation depends on the possession of land with potential for irrigated agriculture
or livestock production. The proximity of the district to major domestic markets has
attracted investment in commercial agriculture, displacing local residents and leav-
ing them with only marginal production areas. Here economic and socially influen-
tial forces related to investments and access to irrigation water are the main drivers
of marginality. In the pastoralist community of Borana, degraded rangeland and
water resources, cultural values attached to livestock, poor range management and
livestock production practices, and marginal environmental conditions were identi-
fied as factors contributing to marginality. Marginality is caused by factors that vary
across regions and livelihoods, but is always a complex interplay of various aspects
that exclude some people from economic growth opportunities, and similar drivers
of marginality are found to be root cause of poverty in all sites. The need for dif-
ferentiated participatory approaches, including much attention to local leadership
and governance, strengthening human capital, and access to productivity enhancing
technology in agriculture, services, and infrastructure are highlighted.

1.7 Conclusions

1.7.1 Focus on Marginalized People

Of course the marginality framework developed and explored here should ulti-
mately help reduce marginality and enhance the opportunities of marginalized peo-
ple, be they women, ethnic minorities, the disabled, or some otherwise excluded
group. At intermediate stages it should help frame action and assist pertinent actors
(i.e., marginalized communities to gain more recognition and voice, or policy advi-
sors, program designers and researchers to facilitate focused initiatives). This book
not only provides evidence and insights of promising and innovative concepts and
approaches, but also provides examples for revitalizing the poverty discourse and
triggering action for change. Focusing merely on the income dimensions of poverty
will not address its root causes shaped by marginality.
Marginalization appears to be very difficult to reverse once complex systems
have produced such a state, but the “poverty trap” concept offers limited insights for
action, compared to human-environmental interrelationships as conceptualized in
socio-ecological systems (SESs). One way to overcome marginalization is to
increase the resilience and adaptability of SESs, however, multiple needs must be
considered simultaneously, mainly: food security, income generation, and ecosys-
tem services. Creating incentives for economic growth works well mostly for those
who already have access to assets or have been reached by employment creation and
service expansion efforts, but not for those excluded from them, whether for reasons
18 J. von Braun and F.W. Gatzweiler

of geography or active discrimination. For many of the extremely poor to benefit

from economic growth it may be necessary to establish certain preconditions first
(i.e., addressing factors that contribute to marginality), in order to make economic
growth an effective means for poverty reduction.

1.7.2 Changing Marginal and Marginality Creating

Environments

The difficulties in reaching the poor at the margins are frequently explained by
physical (being located in remote or harsh environments) and social (being
excluded, discriminated against, or not having access to resources and opportuni-
ties) distances, but may also be related to technological and institutional infrastruc-
ture deficiencies. In the long run, none of these deficiencies need to be taken for
granted. Marginal areas, for instance, can be transformed by investments in tech-
nology and infrastructure, if this pays off in the long run. Demographic shifts in
SSA for instance, are making productivity enhancing investments more attractive
in some areas where demand is rising and the opportunity costs of land and labor
are increasing. As Pingali et al. (Chap. 10 this volume) point out, a strategy pursu-
ing intensification makes sense where it pays off, and releasing marginal lands
from agriculture to provide other ecosystem services where returns on productivity
growth investments will continue to remain low will make sense in some cases.
Investing in targeted R & D to address SSA’s particular constraints—especially
focusing on the crops and traits that are important to the poor and the particular
environmental limitations they face—can lessen marginality and contribute to
widespread reduction of poverty.
More clarity on indicators is still needed, as well-being is partially attributable to
the state of the environment, which can be captured to some extent by global indica-
tors such as genuine savings (Aglietta 2011), which adjusts the savings rate for the
depletion of natural resources. However, as noted in Nkonya et al. (2011), land
degradation is a form of squandering that is not addressed by any current frame-
works and indicators of human well-being, nor is it addressed in measures of genu-
ine savings. This needs to change, as productive land is not only a private good, but
also contributes to the global public well-being, delivering ecosystem benefits to
support a growing world population that creates increasing competition over land
ownership (Deininger and Byerlee 2011).
Institutional infrastructure deficiencies are a major barrier to overcoming mar-
ginality. Institutions consist of property rights that define the flow of cost and ben-
efit streams from resource use, they grant access to the use of resources and
guarantee long-term user rights, and thereby incentives for investing in land and
other forms of capital become more attractive. Institutions of marginality, in con-
trast, are characterized by mixed systems of informal and formal rules, which cre-
ate conditions of limited access to rights and resources. So-called “limited access
orders” (North et al. 2009) maintain themselves by means of extracting others
1 Marginality—An Overview and Implications for Policy 19

(Acemoglu and Robinson 2012) and the environment. Thereby human and natural
capital are wasted and benefits from creativity, competition, innovation, and entre-
preneurial spirit remain untapped.
Creating institutional environments that are more inclusive will also create new
ways for markets, governments, and collective action to reduce poverty. In this vol-
ume, Zhu (Chap. 15) shows that collective action happens among advanced and
marginalized villages in rural China. Although not explicitly referred to, Ahmed
(Chap. 16) points to the potential of collective action among governmental, nongov-
ernmental organizations, and the extremely poor in rural Bangladesh. Ahmad
(Chap. 18) hints to the potential of decentralization and the benefits of a system in
which the marginalized, the government, and actors of the fiscal apparatus work
closely together to overcome marginality, and Christiansen (Chap. 21) mentions
collective action opportunities between business and government. Mwangi et al.
(2011) provide further examples of opportunities of collective action for overcom-
ing marginality.

1.7.3 Need for Multi-dimensional Policies and Programs

Because of the interlinked nature of multiple variables that constitute marginality in

socio-ecological systems, coherent policies (Bromley 2012) and comprehensive
strategies are needed to address them. Point source “optimal” solutions in a narrow
economic accounting framework and panaceas often create problems rather than
solutions for disentangling marginality patterns (Higgs 1996; Scott 1998; Wilson
2002; Ostrom 2007).
While implementing comprehensive interventions to mitigate economic risks
that the extremely poor have encountered, actions to eliminate social exclusion
should be given greater consideration. This may not bring any significant poverty
incidence changes in the short-term, but is an essential step towards eradicating
poverty and promoting inclusive development, as pointed out by Thorat (Chap. 13
this volume), who calls for a shift from attention on changes in the incidence of
poverty to an increasing consideration of how much the poor are benefited from
multi-dimensional intervention policies.
In the absence of genuine local interest in providing for the marginalized, direct
provision of funding by central governments appears to be the only feasible option,
which may be reflected by the increasing popularity of central-source conditional
cash transfers, such as the Oportunidades program in Mexico or the Bolsa project
in Brazil. In such cases, using simple categorical mechanisms as vehicles for target-
ing may be more effective (Ahmad Chap. 18 this volume). Public works with lim-
ited financial incentives can be an alternative, especially if the respective community
and marginalized groups are able to identify the areas where public works may be
most useful (von Braun 1995). Above all, the basic building blocks for good gover-
nance, the ability to track the funds, and the outcomes of public spending remain as
important as individual interventions.
20 J. von Braun and F.W. Gatzweiler

One of the challenges is to strategize the appropriate division of tasks regarding

appropriate measures and actions from the side of policy, civil society, and business.
New ways of doing business to include the marginalized poor have been on the
development agenda of governmental and nongovernmental organizations for some
time now. Especially in rural areas and small farming communities, ecosystem
goods and services provide an important proportion of overall income and founda-
tion of livelihoods. Recognizing the strong dependence of the marginalized poor on
natural capital underlines the importance of policies intended to preserve the envi-
ronment and natural resource bases for agriculture.

1.7.4 What to Expect from the Marginality Framework

In conclusion we ask, what can be expected for whom, from the insights gained
under the marginality framework developed here and the analyses done in that
context? The answers to this question with the following four assertions are a call
for accompanying policy actions with applied and interdisciplinary research for
marginality reduction:
First, recognizing marginality and marginalized people as complex phenomena
leads to equally multi-facetted solutions. At first glance, this appears not to be
helpful for focused development policy. However, narrowly focused policies often
simply aim to treat the most prominent problem areas independent of social,
ecological, and economic contexts, and may not deal with systemic problems. When
marginality patterns and causes are identified and assessed in a multi-level (local,
national, and global) context, appropriate actions become visible. Once the domains
in which marginality may be rooted are identified they can be better addressed, such
as: the lack of rights (e.g., ethnic, gender, disabilities), or the entrenched human
behavior of people who have been discriminated against for a long time in a certain
setting, the lack of access to services or technology, or ill designed macroeconomic
policies that adversely impact certain segments of the labor market. This is an added
value to income-poverty assessment, which are also most valuable as part of margin-
ality assessment, but are further removed from cause and effect concepts.
Second, the marginality framework clarifies the actors’ and agents’ poten-
tial roles, including the marginalized themselves. Beginning with asking questions
like: “who are the marginalized?”; “where do they live?” and “how and why are they
marginalized?” leads to a people and community focus and the institutional setting
under which they act. The marginality framework sees social and ecological sys-
tems as coupled, and facilitates the identification of the relationships between actors
and public infrastructure (physical and social), and how the institutional environ-
ment creates structures which lead to processes of marginalization. By implication
the framework draws attention to the marginality problem and thereby can facilitate
targeting action, not in the sense of targeting a set of poor people, but targeting a set
of root causes of marginality, say social exclusion, ecological risks, or deficient
institutions.
1 Marginality—An Overview and Implications for Policy 21

Third, under the marginality framework, available tool boxes and analytical
approaches can be put to task more appropriately, in order to assess success or
failure of programs or policies. The framework does not guide toward narrowly
focused and reductionist applications of tools that may be internally consistent, cost
effective, and “elegant,” but that lack external and contextual relevance. Participatory
data monitoring, innovative micro-modeling, collective action assessments, experi-
mental designs, or macro–micro simulations may all play a role. The litmus test
here is “did marginality get reduced?”
Fourth, the marginality framework also may be conducive to realism in terms of
temporal dimensions, which is often lacking in development policy. Such realism
may result from assessment of the persistence of marginality in a particular setting.
The framework, however, besides primarily drawing attention to structural issues of
gross inequality—such as exclusion—also calls attention to the temporal subscripts
of causes that shape marginality, say for instance vulnerability to acute shocks or
long-term exclusion. This attention to time can facilitate better identification of pri-
ority actions, as the marginalized poor have no time to lose.
One message of this book is that marginality needs to move more into the center
of public policy in order to achieve inclusion and poverty reduction. Another mes-
sage is that economic and ecological approaches need to come together to compre-
hensively identify and address marginality patterns. A third message is that on the
road towards overcoming marginality, institutional and technological innovations
often need to be more integrated.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Acemoglu D, Robinson J (2012) Why nations fail: the origins of power, prosperity, and poverty.
Crown Business, New York
Adato M, Hoddinott J (2010) Conditional cash transfers in Latin America. Johns Hopkins and the
International Food Policy Research Institute, Baltimore
Aglietta M (2011) Sustainable growth: do we really measure the challenge? In: Measure for
measure-how well do we measure development? Proceedings of the 8th AFD-EUDN confer-
ence, Paris, AFD (Agence Francais de Développement), Paris, 1 Dec 2010, pp 177–230
Ahmed AU, Hill RV, Smith LC, Frankenberger T (2009) The poorest and hungry: characteristics and
causes. In: von Braun J, Hill RV, Pandya-Lorch R (eds) The poorest and hungry: assessments,
analyses, and actions. International Food Policy Research Institute, Washington, DC, pp 107–128
Banerjee A, Duflo E (2011) Poor economics: a radical rethinking of the way to fight global poverty.
Public Affairs, New York
Baumüller H, Ladenburger C, von Braun J (2011) Innovative business approaches for the reduction
of extreme poverty and marginality? ZEF working paper 80, Center for Development Research,
University of Bonn, Bonn
Bromley DW, Anderson GD (2012) Vulnerable people, vulnerable states. Routledge, New York
22 J. von Braun and F.W. Gatzweiler

Collier P (2007) The bottom billion: why the poorest countries are failing and what can be done
about it. Oxford University Press, Oxford
Damasio A (2011) Neural basis of emotions. Scholarpedia 6(3):1804
Dasgupta P (2003) World poverty: causes and pathways. Plenary lecture presented at the World
Bank’s Annual Bank Conference on Development Economics (ABCDE), Bangalore, 21–22
May 2003. http://www.econ.cam.ac.uk/emeritus/dasgupta/worldpov.pdf. Accessed 23 Feb 2013
Dasgupta P (2009) Poverty traps: exploring the complexity of causation. In: von Braun J, Hill RV,
Pandya-Lorch R (eds) The poorest and hungry: assessments, analyses, and actions. International
Food Policy Research Institute, Washington, DC, pp 129–146
Deininger K, Byerlee D (2011) Rising global interest in farmland: can it yield sustainable and
equitable benefits? World Bank, Washington, DC
FAO (2012) The state of food insecurity in the world. Food and Agriculture Organization of the
United Nations, Rome
Gatzweiler F, Baumüller H, von Braun J, Ladeburger C (2011) Marginality: addressing the root
causes of extreme poverty. ZEF working paper 77, Center for Development Research,
University of Bonn, Bonn
Gordon D, Levitas R, Pantazis C, Patsios D, Payne S, Townsend P (2000) Poverty and social exclu-
sion in Britain. Joseph Rowntree Foundation, York
Grant U, Shepherd A (2008) Should and can development policy focus on the poorest too? Working
paper 125, Chronic Poverty Research Center, Overseas Development Institute, London
Higgs R (1996) Legally induced technical regress in the Washington Salmon Fishery. In: Alston
LJ, Eggertsson T, North DC (eds) Empirical studies in institutional change. Cambridge
University Press, New York, pp 247–279
Levy S (2008) Good intentions, bad outcomes: social policy, informality, and economic growth in
Mexico. Brookings Institute, Washington, DC
Mwangi E, Markelova H, Meinzen-Dick R (2011) Collective action and property rights for poverty
reduction. Insights from Africa and Asia. University of Pennsylvania Press, Philadelphia
Nkonya I, Gerber N, Baumgärtner P, von Braun J, De Pinto A, Graw V, Kato E, Kloos J, Walter T
(2011) The economics of land degradation: toward an integrated global assessment. Peter
Lang, Essen
North DC, Wallis JJ, Weingast BR (2009) Violence and social orders: a conceptual framework for
interpreting recorded human history. Cambridge University Press, New York
Nussbaum MC (2004) Beyond “Compassion and Humanity”: justice for nonhuman animals. In:
Sunstein CR, Nussbaum MC (eds) Animal rights: current debates and new directions. Oxford
University Press, Oxford, pp 299–320
Nussbaum MC, Sen A (1993) The quality of life. Oxford University Press, Oxford
Ostrom E (2007) Sustainable Social-ecological systems: an impossibility? Paper presented at the
annual meetings of the American association for the advancement of science, San Francisco,
15–19 Feb 2007
Prahalad CK (2004) The fortune at the bottom of the pyramid. Wharton School Publishing,
Philadelphia
Room G (1995) Beyond the threshold: the measurement and analysis of social exclusion. Policy
Press, Bristol
Sachs J (2006) The end of poverty: economic possibilities for our time. Penguin Press, London
Sauders P (1994) Welfare and inequality: national and international perspectives on the Australian
welfare state. Cambridge University Press, Cambridge
Schlosberg D (2007) Defining environmental justice: theories, movements, and nature. Oxford
University Press, Oxford
Scott JC (1998) Seeing like a state: how certain schemes to improve the human condition have
failed. Yale University Press, New Haven
Sen A (1987a) Commodities and capabilities. Oxford University Press, New Delhi
Sen A (1987b) The standard of living. Cambridge University Press, Cambridge
Sen A (1999) Development as freedom. Oxford University Press, Oxford
1 Marginality—An Overview and Implications for Policy 23

Stark O, Bloom DE (1985) The new economics of labor migration. Am Econ Rev 75(2):173–178
Stark O, Kobus M, Jakubek M (2011) A concern about low relative income, and the alignment of
utilitarianism with egalitarianism. ZEF discussion papers on development policy 156, Center
for Development Research, University of Bonn, Bonn
Townsend P (1979) Poverty in the United Kingdom: a survey of household resources and standards
of living. Penguin Books, Harmondsworth
Townsend P, Corrigan P, Kovarzik U (1987) Poverty & labour in London: interim report of a
centenary survey. Low Pay Unit, London
von Braun J (ed) (1995) Employment for poverty reduction and food security. International Food
Policy Research Institute, Washington, DC
von Braun J, Hill RV, Pandya-Lorch R (eds) (2009) The poorest and hungry: assessments, analy-
ses, and actions. International Food Policy Research Institute, Washington, DC
Walker I, Smith HJ (2002) Relative deprivation: specification, development and integration.
Cambridge University Press, Cambridge
Wilkinson RG, Pickett KE (2007) The problems of relative deprivation: why some societies do
better than others. Soc Sci Med 65(9):1965–1978
Wilson JA (2002) Scientific uncertainty, complex systems, and the design of Common-pool
institutions. In: Committee on the Human Dimensions of Global Change, National
Research Council (ed) The drama of the commons. National Academies Press, Washington,
DC, pp 327–359
 Part I
Concepts and Theory
Chapter 2
Marginality—A Framework for Analyzing
Causal Complexities of Poverty

Franz W. Gatzweiler and Heike Baumüller

Abstract This chapter presents an interdisciplinary framework for the investigation

of marginality which is inclusive of the diversity of existing poverty research
approaches. Marginality is presented as a systemic and evolutionary concept with
particular reference to the role of institutions that constrain or motivate actions as
measured against a performance indicator such as productivity growth. Based on a
brief review of marginality research in social, economic, and development fields,
this chapter presents a definition of marginality and explains the differences between
this conceptual framework and those of poverty. Finally, the components of the
framework and its interrelationships are described and awareness for the need for
further research on marginality is raised.

Keywords Causal complexities • Conceptual framework • Capability approach

• Systems • Institutions

2.1 Towards the Development of a Concept

The persistence of poverty has motivated research to shift from looking at single
dimension explanations towards recognizing that the causes of poverty are “complex,
multifaceted, and difficult to isolate” (Haveman and Smeeding 2007, 2). As a result,
poverty research has contributed to a more comprehensive understanding of
the causes of poverty, and it has been recognized that the traditional methods
of disciplinary science fall short of seeing and explaining the “big picture” of causal

F.W. Gatzweiler (*) • H. Baumüller

Center for Development Research (ZEF), University of Bonn,
Walter Flex Strasse 3, 53113 Bonn, Germany
e-mail: [email protected]; [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 27
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_2,
© The Author(s) 2014
28 F.W. Gatzweiler and H. Baumüller

factors underlying poverty (Meinzen-Dick et al. 2004; von Braun et al. 2009, 44).
These shortcomings led us to the development of a more inclusive and interdisci-
plinary research framework, that of marginality.
Kant (1819) noted that a concept is a general representation that is common to
several specific objects. Accordingly, the concept of marginality is an abstraction of
the idea that the causal complexities underlying people’s living conditions interact
in ways that are at systemic margins. These conditions are far from what would be
considered optimal, in balance, just, equal, sufficient, good, or fair—attributes that
describe conditions and positions in human life that are enabling and supportive,
and that are used to define poverty.
Despite the critiques of marginality in social science (Cullen and Pretes 2000;
Del Pilar and Udasco 2004), the persistence of marginality as a concept (Dickie-Clark
1966) should be regarded as an indication of the demand to express observations of
a similar kind across different epistemic cultures (Knorr Cetina 1999), and to find
solutions, here, to the phenomena of poverty. The problem of measuring the degree
of marginalization is that the reference is not fixed or unknown, and therefore when
used as a theory marginality has been criticized for the lack of construct validity
(Del Pilar and Udasco 2004, 11). In their critique, however, those authors reviewed
the use of marginality as a theory and came to the conclusion that “marginality
cannot work [as a theory] if it has multiple levels of meaning.” This critique of
the marginality concept as a theory also rests on the belief that a concept must be
a “uniform kind of mental representation” (Weiskopf 2009, 145). Weiskopf
rejects this assumption in psychology and outlines a pluralist theory of concepts in
which they are constituted by multiple representational kinds. In the following
we present a framework for the investigation of marginality—not a theory of
marginality.
Although poverty can be observed in many different forms and is caused by
many different factors, all forms of poverty can be described through the concept of
marginality. Someone who is poor will always be marginalized in one or more
dimensions, whereas the socio-cultural context and individual perception will define
in which and in how many dimensions someone needs to be marginalized in order
to be considered poor. The aim of establishing a concept of marginality is therefore
to better understand the various causal complexities of poverty by deepening and
broadening the scope of scientific investigation through:
1. identifying common causalities of poverty across scientific disciplines, and
2. including phenomena that are not typically considered as poverty or contributing
to poverty alone (e.g., living in harsh or resource scarce environments).
Deepening and broadening the scope of investigation thereby also includes
incorporating theories and models from other (non-social science) epistemic cultures
and scientific disciplines. In that sense marginality is not only a concept, but also a
conceptual framework. It is a framework for different theories of poverty within
which various models can be tested.
Frameworks. Theories and models are understood as “nested set[s] of theoretical
concepts, which range from the most general to the most detailed types of
2 Marginality—A Framework for Analyzing Causal Complexities of Poverty 29

Table 2.1 Classes of poverty theories and selected references

Classes of poverty theory Selected references
Individual deficiencies Rainwater (1970), Ryan (1976), Gwartney and
Poverty is caused largely as a result McCaleb (1985), Herrnstein and Murray (1994),
of the attributes of individuals Weber (2001)
and the choices they make
Socio-cultural and belief systems Kapp (1963), Moynahan (1965), Valentine (1968),
Poverty is culturally manifested Murray (1984), Putnam (1993), Putnam and
and transmitted by values, Helliwell (1995), Bhalla and Lapeyre (1997),
beliefs, and norms (e.g., ghetto Lewis (1998), Mbakogu (2004), Roland (2004),
or slum subcultures) Small et al. (2010)
Political-economic (structural) barriers Smith (1776, Marx (1867), Polanyi (1944), Hamilton
The political, social, and economic systems (1967), Sen (1982), Nussbaum and Sen (1993),
provide insufficient incentives and Tobin (1994), Jencks (1996), Dasgupta (2003),
opportunities: shift in focus from the Desai (2007), North et al. (2009), von Braun
actors of the game to the game itself et al. (2009), Acemoglu and Robinson (2012)
Geographical and environmental Von Thünen (1826), Goldsmith and Blakely (1992),
disparities Dasgupta and Mäler (1994), Shaw (1996), UNDP
Causes of poverty are spatial factors (1998), Pingali (2003), Bradshaw and Muller
(e.g., distance to growth centers, (2004), Duraiappah (2004), Weber and Jensen
marginal land and climatic (2004), Pingali et al. (2005), Dellink and Ruijs
conditions) (2008), Diamond (2011)
Modified from Bradshaw and Muller (2004)

assumptions made by the analyst” (Ostrom 2005, 27). Frameworks organize, form
boundaries around the inquiry, set up general relationships among categories or
dimensions, as well as define the scope and levels of the inquiry. They do not
explain or predict, rather they organize the diagnostic inquiry. (Ostrom et al. 1994;
Schlager 1999).
Theories. Theories explain particular parts of a framework, and therefore need to
make assumptions about the patterns of relationships within frameworks. Several
theories may be accommodated within a single framework. Table 2.1 presents a
brief overview of classes of poverty theories and selected references. As Bradshaw
(2005) argues, the choice of a theory and the definition of the problem are thereby
not only scientifically motivated, but also politically influenced. O’Connor (2001, 12)
mentions that poverty research is also a political act, which is influenced by biases
and values of an educated elite who aims to “categorize, stigmatize, but above all to
neutralize the poor and disadvantaged through analysis that obscures the political
nature of social and economic inequality.”
Defining and choosing a theory means defining what is to be explained and
therefore the choice of theories is also political. For example the application
of neoclassical economic theory makes use of methodological individualism,
which can be seen as reinforcing individualistic sources of poverty. Theories in
this tradition attempt to redress the problem that “[p]overty researchers have in
effect focused on who loses out at the economic game, rather than addressing the
fact that the game produces losers in the first place” (Rank et al. 2003, 3). Models
are then applied to further specify analyses by defining concrete assumptions and
30 F.W. Gatzweiler and H. Baumüller

specific variables. “Logic, mathematics, game theory models, experimentation and

simulation, and other means are used to explore the consequences of these assump-
tions systematically” (Ostrom 2005, 28).
Eventually the challenge of drawing the bigger picture of marginality includes
taking a critical look at the science applied to understand the complexities of
marginality itself. Here science at the margins needs to investigate which knowledge
about marginality is adequate, how it influences the outcomes of the investigation,
and how this knowledge should be obtained. Despite the fact that the marginality
framework is open to different disciplinary approaches, different epistemic cultures
(Knorr Cetina 1999), with their specific rules and norms for acquiring knowledge,
do not always easily lend themselves to a synergy of findings.

2.2 Definition and Delineation of Marginality

Here we adopt the definition of marginality provided by Gatzweiler et al. (2011, 3):
an involuntary position and condition of an individual or group at the margins of social,
political, economic, ecological or biophysical systems, preventing them from access to
resources, assets, services, restraining freedom of choice, preventing the development of
capabilities, and eventually causing (extreme) poverty.

This definition is anthropocentric and describes the position and condition of an

individual, actor, or group within social, economic, and ecological systems. The
position of an actor describes their place and function within social and geographical
spaces. For example, actors can be the head of a cooperative or peasant association
and in that position are authorized or required to make particular decisions. Their
geographical position refers to where they are actually located in physical space.
The condition of an actor refers to their decision-making and information processing
capabilities, and the assets and resources they can make effective use of. Being
marginalized means being positioned at the margin of one or more societal or spatial
systems, and having few assets and/or capabilities that would allow the actor to
move away from or change that marginal situation. It then depends on the theory
and model applied in the investigation of marginality, and which evaluative criteria
are used to measure the degree of marginality.
While it is generally agreed upon that marginality is always relative to a particular
point, perspectives differ on how and by whom this point (or center) should be
defined. In some cases marginality may be regarded as fixed, such as different
regions, groups, or individuals that are part of a hierarchy centered on an (immobile)
reference point (Cullen and Pretes 2000). This perspective is reflected in the common
discourse on ‘development,’ when countries are described as either ‘developing’ or
‘developed’ along a predefined linear trajectory. Others dispute the idea of a single center,
but rather define marginality depending on which one of multiple reference points is
used. Thus as Dunne (2005, 15) notes, marginality can be seen as “a multidimensional
phenomenon in that a given person may be simultaneously integrated with one or
more centers while being marginal from one or more other centers.”
2 Marginality—A Framework for Analyzing Causal Complexities of Poverty 31

Fig. 2.1 Changing position

and condition to reduce
marginality within a
multidimensional system: the
large circles represent
systemic dimensions (social,
economic, political, etc.), the
two smaller shaded circles
represent marginalized
individuals or groups, the
arrow represents the
directionality of change
toward the center of more
than one system (reduction
of marginality) (Modified
from Gatzweiler et al. 2011, 15)

Figure 2.1 illustrates marginality in the context of a person or group by a specific

position within multiple dimensions. Their position can be closer to, or further away
from, the desired center. Distance needs to be understood in terms of transaction and
interdependence costs (i.e., the efforts required to change and maintain a position).
The condition refers to the well-being of the individual or group and is depicted by
the size of the small, solid grey circles, whereby the larger of both depicts improved
well-being. The circles represent different systemic (social, economic, political,
nutritional, educational) dimensions of people’s lives in which they are more or less
marginalized. The size of the circles indicates the importance of these dimensions
to the respective actor or group.
Our definition of marginality draws on different disciplines, including: econom-
ics, development theory, sociology, ecology, and anthropology. Marginality is fre-
quently defined by two conceptual frameworks—spatial and social marginality
(Gurung and Kollmair 2005)—that determine the manifestations and drivers of
marginality (a distinction that is not always clear, however). We add ecological
marginality which, in addition to the other two, draws attention to extreme values
outside homeostatic ranges of living systems (see Callo-Concha et al., Chap. 4
this volume). Spatial marginality tends to focus on the distance or connectivity of
geographical areas in relation to centers of economic activity at different geographi-
cal scales (e.g., globally or regions within a country). In this context economic
determinants tend to be seen as the main drivers of marginality based on center-
periphery or core-periphery models, where one region is the center and the others
are marginal (Cullen and Pretes 2000).
Social marginality is concerned with “human dimensions such as demography,
religion, culture, social structure (e.g., caste, hierarchy, class, ethnicity, gender),
economics, and politics in connection with access to resources by individuals
and groups” (Gurung and Kollmair 2005, 10). Research into social marginality
examines the underlying reasons for exclusion, inequality, social injustice, and the
spatial segregation of people. Marginalization is seen here as a social construction,
with socio-political power as the central determinant (Cullen and Pretes 2000).
32 F.W. Gatzweiler and H. Baumüller

Social marginality can also lead to spatial marginality, for instance in cases where
ethnic groups are displaced to disadvantaged geophysical regions.
In the study of spatial marginality, the investigation of factors driving the emergence
of spatially divergent growth paths within countries has emerged as an important
research area in economics, including the fields of location theory, urban economics,
and economic geography (Fujita and Thisse 2009). This line of research originated
from the work of von Thünen (1826) and his attempts to model the location of
agricultural production in relation to markets. Since the early 1990s, drivers and
dynamics of economic concentration have attracted renewed interest among econo-
mists, in particular through the work of Krugman in the field of ‘new economic
geography’ (e.g., Krugman 1991, 1999; Krugman and Venables 1995).
Two sets of geographical factors—first- and second-nature geography—are
commonly identified as determinants shaping the spatial distribution of economic
activity (World Bank 2008). First-nature geography refers to the geographical
endowments of certain regions such as proximity to coasts, rivers, borders, or ports,
which may underlie economic success (e.g., of Chinese coastal zones or Mexican
border regions close to the USA). Gallup et al. (1999) highlight such geographic
factors as some of the key determinants of industrial location, such as the extent
of a country’s land area located in the tropics, concentration of populations in
relationship to the interior versus coasts, access to maritime transport routes, and
distance to core markets.
Second-nature geography relates to the interactions between economic agents,
and in particular economies of scale that can be achieved through agglomeration
and economic concentration (World Bank 2008). These factors have been the focus
of work by Krugman (1999), who argues that the emergence of an industrial core
is largely (though not exclusively) driven by centripetal forces in the form of backward
and forward linkages in the growth centers (e.g., a large pool of suppliers and/or
consumers), thick labor markets, and information spillovers. Forces counteracting
these drivers—what Krugman refers to as centrifugal forces—include the availability
of immobile factors in certain regions, increased demand and prices for land in
areas characterized by economic concentration, and external diseconomies (such
as congestion).
In this context the center tends to be defined through certain indicators and perfor-
mance may be judged in relation to a particular average or the performance of leading
regions. Some categorizations of regional development focus on economic indicators,
such as income, consumption, or GDP per capita (e.g., World Bank 2008; Ghani 2010).
Others compare regions on the basis of more diverse socio-economic indicators. For
instance, in Mexico the federal government has developed a marginality index that
ranks regions according to their performance in terms of education, housing, monetary
income, and distribution of the population (CONAPO 2005). What these approaches
have in common is that they tend to evaluate regions in relation to a particular reference
point and performance indicator for economic or human development.
Spatial marginality is also reflected in the field of development theory, notably
through Latin American structuralisms, dependency theory, and world systems the-
ory, which generally focus on nation states (and the world) as the unit of analysis.
2 Marginality—A Framework for Analyzing Causal Complexities of Poverty 33

Starting with Prebisch in the 1940s and developed by leading economists and
sociologists such as Furtado, Gunder Frank, and Wallerstein, this line of thinking
contends that the global market is divided into powerful and technologically
advanced economies (the center), and relatively weak peripheral economies (the
periphery) that supply raw material and low-tech manufacturing (Preston 2002).
The underdevelopment of periphery economies is thus a result of their position
in the global (capitalist) system. To understand and address problems of underde-
velopment Wallerstein argued that it is necessary to look at the global system as a
whole rather than at individual nation states.
In the social sciences the concept of marginality can be traced back to Park’s
essay “Human Migration and The Marginal Man” published in 1928 and Stonequist’s
(1937) elaboration of the concept a decade later. Park saw the ‘marginal man’ as a
personality type that emerges as a consequence of migration. Thus a ‘marginal man’
is “a cultural hybrid, a man living and sharing intimately in the cultural life and
traditions of two distinct peoples” (Park 1928, 892). This form of marginalization
can lead to social disorganization, but also social reconstruction where the margin-
alized become intermediaries between cultures.
Park’s and Stonequist’s thinking influenced North American sociologists in
particular, who developed the idea over time. Marginality is generally seen from the
perspective of the individual or group. While debates initially focused on issues of
race and ethnic relations, the concept was later extended to the study of occupations,
gender, and scientific innovation (Goldberg 2012). Others have also sought to
broaden the concept to include anyone who in one way or another is marginalized
from one or more social groups (e.g., Hughes 1949; Deegan 2002). For instance,
Deegan defines a “marginal person” as anyone whose “perception of self, experience
of the world, and access to material resources” do not fit the prevailing society or
culture (Deegan 2002 cited in Goldberg 2012, 208).
Yet others have turned their focus to groups of individuals in similar situations that
may share a common marginal culture or identity. Goldberg (1941, 53) was the first to
argue that marginal individuals may give rise to a marginal culture over time, that is
“every bit as real and complete to him as is the nonmarginal culture to the nonmarginal
man.” Thus a subculture may be marginalized vis-à-vis the prevailing culture, but the
individual may not feel marginal within this new subculture. In some cases a group
may in fact choose to remain outside of the dominant culture in order to preserve their
own identities and independence (Scott 2009). Using the case of the Haitian diaspora
in Guadeloupe, Brodwin (2003, 403) illustrates how a group’s self-definition (or sub-
jectivity) is indeed shaped by the “experience of marginalization in a specific time and
place.” Wright and Wright (1972) distinguish between three types of marginality that
characterize groups: cultural marginality (shared behavior patterns), psychological
marginality (shared attitudes), and social marginality (patterned interrelationships).
Our units of analysis are the marginalized poor (i.e., individuals or groups that
are affected both by marginality and poverty). To define the ‘center,’ we draw
on Sen’s capability approach (e.g., Sen 1979, 1992, 1999). A person’s pursuit of
well-being is shaped by what someone has chosen or been able to achieve their
“functionings” (Crocker 1992, 585). The extent of potential “functionings” is in
34 F.W. Gatzweiler and H. Baumüller

turn determined by the person’s capabilities (i.e., what they can feasibly achieve).
What people choose to do may differ even if they have the same capabilities as one
another. Resources or commodities then simply become the means for achieving
well-being. Thus we define the center as the place where individuals or groups are
able to realize the desired “functionings” within their capabilities that lift them out
of poverty. The ability to do so will be determined by individual characteristics,
social institutions, and the geophysical context (Dissart et al. 2008).
Marginality can influence this process at different stages. It may restrict people’s
access to resources, for instance where people live in geo-physically disadvantaged
areas characterized by poor soils or limited water resources. Marginality may also
influence what a person may be able to achieve, for instance, where people or groups
are excluded from certain opportunities due to their gender (e.g., reduced school
attendance among girls in some patriarchal cultures) or ethnicity (e.g., preference
given to certain ethnic groups over others with respect to assuming positions of
authority). Finally, marginality can influence a person’s ability to take advantage of
the opportunities that are open to them. Here it is important to note that the potential
“functionings” that someone may regard as achievable can also be shaped by margin-
ality, given that the experience of marginality influences how individuals and groups
define themselves, their opportunities, and their abilities (Brodwin 2003).
We regard this as a dynamic and circular process where marginality can function
both as a cause and a consequence of poverty. Also the different determinants,
types, and outcomes of marginality are often interrelated. For instance, marginality
in education is influenced by factors such as poverty, language, stigmatization
of certain groups (e.g., on ethnic, cultural, or gender grounds), and legal rights
(UNESCO 2010). As already noted, social marginality (e.g., due to ethnicity or
income), can also lead to spatial marginality, as witnessed in urban ghettos or the
relocation of certain groups to remote or less productive land, which in turn leads to
marginalization from jobs, services, education, or infrastructure.

2.3 Components of the Conceptual Framework

The marginality framework takes into account the diversity of causes of people living
in poverty. As this diversity of factors causing poverty is frequently clustered, it has
been referred to as “causal complexes” or “marginality patterns” by Gatzweiler et al.
(2011, 9). Examples of such patterns are low agricultural productivity caused by an
inability to irrigate as a result of water shortages due to low water tables and the lack
of collective action resulting from central planning that in turn cause critical human
health conditions under which the elderly, woman, and children suffer most and therefore
cannot contribute their labor or receive education, which limits their ability to change
or escape the systems they are part of. Causal complexes that are self-referential
systemic feedback loops are also referred to as ‘poverty traps’ (Dasgupta 2009).
Marginality is caused by complexes of interrelating factors that are seldom
directly observable. Conceptualizing marginality in terms of causal complexes or
networks requires shifting the focus from isolated causal entities to the relationships
2 Marginality—A Framework for Analyzing Causal Complexities of Poverty 35

Table 2.2 Components of the framework for the analysis of marginality

Actors Decide and act on, and in particular positions and conditions
Follow strategies, interact and coordinate in particular ways at different
levels of organization
Engage in rule-making themselves
Biophysical Composed of the enabling or constraining social and biophysical
and societal environments (biophysical, climatic, material, environmental
variables conditions)
Societal conditions Attributes and assets of the community and the individual
Institutions (rules in use and in form at various levels of decision making)
Governance (coordination mechanisms and strategies)
Outcomes Well-being, poverty, degree of marginalization
Evaluative criteria (re/evaluate position and condition of an actor
by criteria selected according to the theory and model applied,
the criteria measure particular, not all aspects of marginality)
Feedback and corrective action (actions based on outcomes and aimed at
changing the enabling or constraining variables as well as the way
actors have made decisions and behave)

among them. Understanding these causal complexes of marginality better—instead of

merely understanding correlations between a few variables and poverty—contributes
to a better understanding of the behavior of socio-ecological systems, and to our
ability to design policies and programs that are more responsive to the needs of the
marginalized poor.
The general question behind the conceptual framework concerns the causal
complexities and how they can be changed in order to improve livelihoods and reduce
poverty. This framework is inspired by the Institutional Analysis and Development
framework of Ostrom et al. (1994) and the Institutions of Sustainability (IoS)
framework of Hagedorn (2008), and consists of the components shown in Table 2.2.
The marginality framework groups the causal complexes into societal and bio-
physical clusters. Societal causes refer to ‘software’ variables and include the capa-
bilities of actors and communities, their social and human capital, the rules and
regulations they have set up for their societies to function, and the ways in which
rules are made and executed. Biophysical causes include ‘hardware’ variables such
as geographic location, soil quality, vegetation, and climate, but also man-made
capital (infrastructure) such as roads and buildings. Both groups have enabling or
constraining impacts on how actors make decisions and act.
Actors behave according to certain types of rationality in order to improve their
societal position or geographical location and their condition. Depending on the
theories and models applied, the actors’ rationality is a function of the institutional
setting in which they act, and their condition includes the assets and opportunities
they have for establishing and improving their level of well-being and degree of
marginalization. Vatn (2005, 113) explains that what is considered rational depends
on the institutional context, and as this context can vary, so can the meaning of
rationality and the assumed objective of rational behavior. Maximizing individual
utility by means of rational choice is assumed in neoclassical economics and with
36 F.W. Gatzweiler and H. Baumüller

Fig. 2.2 Conceptual framework for the investigation of marginality: actors are in particular positions
and conditions, and make decisions that are enabled or constrained by biophysical or societal
factors, as part of the causality clusters (actors’ decisions result in outcomes that change their
well-being, measured by evaluative criteria, and also change the enabling and constraining factors
for decision making in the next round)

this type of rationality come the assumptions of a particular institutional setting in

which the actors make decisions. Recognizing that information for making choices
comes at a cost leads to bounded rationality and satisficing rationality (Simon 1957,
1959), and recognizing the fact that people behave socially, and that mutually
responsive implies different types of social rationality (Etzioni 1988; Gintis 2000;
Ostrom 2000). Which rationality is applied depends on the theories and models
applied in the investigation of marginality (Fig. 2.2).
Ideally the outcomes of actors’ decisions and actions improve their well-being and
reduce poverty. Alternatively the outcomes can be measured as the relative change in
the positions and conditions of the actors. Frequently, however, the poor are caught in
a constraining environment that prevents them from improving their positions and
conditions. Ideally feedback mechanisms would send signals to actors at higher
decision-making levels and allow them to change critical variables in the causality
clusters, changing the environment from constraining to enabling. For example, improv-
ing land tenure security in the Ethiopia would be expected to motivate owners to make
long-term investments into land and productivity gains (Deininger et al. 2003).

2.4 Conclusions

In this chapter we have explained the opportunities and the need for research on
marginality that include the views from multiple disciplines on multiple dimensions
of poverty. Reducing the diversity of poverty to a few indicators in order to facilitate
its measurement entails the risk of overlooking critical features and causalities
2 Marginality—A Framework for Analyzing Causal Complexities of Poverty 37

underlying poverty that may be detected when poverty is evaluated from a systems
perspective. In recognition of the diversity of poverty dimensions and their irreduc-
ible and partly incommensurable nature, we have proposed a conceptual framework
and described its components and relationships. Institutions play a particular role in
explaining causal relationships among different types of marginality.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Acemoglu D, Robinson J (2012) Why nations fail: the origins of power, prosperity, and poverty.
Crown Business, New York
Bhalla A, Lapeyre F (1997) Social exclusion: towards an analytical and operational framework.
Dev Chang 28(3):413–433
Bradshaw TK (2005) Theories of poverty and anti-poverty programs in community development.
Rural Poverty Research Center, University of Missouri and Oregon State University,
Columbia
Bradshaw T, Muller B (2004) Shaping policy decisions with spatial analysis. In: Goodchild MF,
Janelle DG (eds) Spatially integrated social science: examples in best practice. Oxford
University Press, New York, pp 300–322
Brodwin P (2003) Marginality and subjectivity in the Haitian Diaspora. Anthropol Q 76(3):383–410
CONAPO (2005) Índice De Marginación a Nivel Localidad 2005. Consejo Nacional de Población,
Mexico DF. http://www.conapo.gob.mx/index.php? option= com_content&view=article&id=4
6&Itemid=194. Accessed 18 Feb 2012
Crocker DA (1992) Functioning and capability: the foundations of Sen’s and Nussbaum’s develop-
ment ethic. Polit Theory 20(4):584–612
Cullen BT, Pretes M (2000) The meaning of marginality: interpretations and perceptions in social
science. Soc Sci J 37(2):215–229
Dasgupta P (2003) World poverty: causes and pathways. Plenary lecture delivered at the world
bank’s Annual Bank Conference on Development Economics (ABCDE), Faculty of Economics,
University of Cambridge, Bangalore, 21–22 May 2003
Dasgupta P (2009) Poverty traps: exploring the complexity of causation. In: von Braun J, Hill RV,
Pandya-Lorch R (eds) The poorest and hungry: assessments, analyses, and actions. International
Food Policy Research Institute, Washington, DC, pp 129–154
Dasgupta P, Mäler K-G (1994) Poverty, institutions and the environment-resource base. World
Bank Environment paper no 9, World Bank, Washington, DC
Deegan MJ (2002) Race, hull-house, and the University of Chicago: a New conscience against
ancient evils. Praeger, Westport
Deininger K, Jin S, Adenew B, Gebre-Selassie S, Nega B (2003) Tenure security and land-
related investment: evidence from Ethiopia. Research working paper 2991, World Bank,
Washington, DC
Del Pilar JA, Udasco JO (2004) Marginality theory: the lack of construct validity. Hisp J Behav Sci
26(1):3–15
Dellink RB, Ruijs A (2008) Economics of poverty, environment and natural resource use. Springer,
Dortdrecht
Desai RM (2007) The political economy of poverty reduction. Wolfensohn Center for Development
working papers no. 2, Brookings Institution, Washington, DC
Diamond J (2011) Collapse: how societies choose to fail or succeed. Penguin Press, New York
38 F.W. Gatzweiler and H. Baumüller

Dickie-Clark HF (1966) The marginal situation: a sociological study of a coloured group.

Routledge and Kegan Paul, London
Dissart J, Bresson F, Lallau. B (2008) Assessing rural marginality: contribution of the capability
approach. Paper presented at the annual conference of the Human Development and Capability
Association (HDCA), New Delhi, 10–13 September 2012. http://www.capabilityapproach.
com/pubs/Dissart%5B1%5D.Bresson.Lallau-HDCA.pdf. Accessed 19 Feb 2012
Dunne RJ (2005) Marginality: a conceptual extension. In: Dennis RM (ed) Marginality, power, and
social structure, vol 12, Issues in race, class, and gender analysis. Elsevier, Oxford, pp 11–28
Duraiappah AK (2004) Exploring the links: human well-being, poverty and ecosystem services.
International Institute for Sustainable Development and United Nations Development Program,
Winnipeg
Etzioni A (1988) The moral dimension: toward a new economics. Free Press, New York
Fujita M, Thisse J-F (2009) New economic geography: an appraisal on the occasion of Paul
Krugman’s 2008 Nobel prize in economic sciences. Reg Sci Urban Econ 39(2):109–119
Gallup JL, Sachs JD, Mellinger AD (1999) Geography and economic development. Int Reg Sci
Rev 22(2):179–232
Gatzweiler F, Baumüller H, von Braun J, Ladenburger C (2011) Marginality: addressing the root
causes of extreme poverty. ZEF working paper no 77, Center for Development Research,
University of Bonn, Bonn
Ghani E (2010) Development disparities and peculiarities. In: Ghani E (ed) The poor half billion in
south Asia: what is holding back lagging regions? Oxford University Press, Oxford, pp 29–63
Gintis H (2000) Beyond homo oeconomicus: evidence from experimental economics. Ecol Econ
35:311–322
Goldberg M (1941) A qualification of the marginal man theory. Am Sociol Rev 6(1):52–58
Goldberg CA (2012) Robert Park’s marginal man: the career of a concept in American sociology.
Laboratorium 4(2):199–217
Goldsmith WW, Blakely EJ (1992) Separate societies: poverty and inequality in American cities.
Temple University Press, Philadelphia
Gurung GS, Kollmair M (2005) Marginality: concepts and their limitations. IP6 working paper,
Swiss National Centre of Competence in Research North South, Bern
Gwartney J, McCaleb TS (1985) Have antipoverty programs increased poverty. Cato J
5(5):1–16
Hagedorn K (2008) Segregating and integrating institutions—a dichotomy for nature related
institutional analysis. In: Schäfer C, Rupschus C, Nagel UJ (eds) Enhancing the capacities of
agricultural systems and producers, proceedings of the second green week scientific conference,
16–17 January 2008. Weikersheim, Margraf
Hamilton D (1967) The political economy of poverty: institutional and technological dimensions.
J Econ Issues 1(4):309–320
Haveman R, Smeeding T (2007) Poverty policy and poverty research over four decades. Focus 25(1):1
Herrnstein RJ, Murray C (1994) The bell curve. Free Press, New York
Hughes EC (1949) Social change and status protest: an essay on the marginal man. Phylon
10(1):58–65
Jencks C (1996) Can we replace welfare with work? In: Darby MR (ed) Reducing poverty in
America. Sage, Thousand Oaks, pp 69–81
Kant E (1819) Logic. First part general doctrine of elements, first section: conceptions, translated
from German by Richardson J, printed for W Simpkin and R Marshall, Stationers’ Court,
London
Kapp KW (1963) The social costs of business enterprise. Spokesman, Nottingham
Knorr Cetina K (1999) How the sciences make knowledge. Harvard University Press, Cambridge/
London
Krugman P (1991) Increasing returns and economic geography. J Polit Econ 99(3):483–499
Krugman P (1999) The role of geography in development. Int Reg Sci Rev 22(2):142–161
Krugman P, Venables AJ (1995) Globalization and the inequality of nations. Q J Econ 110(4):
857–880
2 Marginality—A Framework for Analyzing Causal Complexities of Poverty 39

Lewis O (1998) The culture of poverty. Society 35(2):7

Marx K (1867) Das Kapital. Kritik der Politischen Ökonomie. Verlag von Otto Meissner, Hamburg
Mbakogu IA (2004) Is there really a relationship between culture and development? Anthropologist
6(1):37–43
Meinzen-Dick R, Adato M, Haddad L, Hazell P (2004) Science and poverty. An interdisciplinary
assessment of the impact of agricultural research. Food policy report, International Food Policy
Research Institute, Washington, DC
Moynahan D (1965) The Negro family. US Department of Labor, Office of Policy Planning and
Research, Washington, DC
Murray C (1984) Losing ground. Basic, New York
North DC, Wallis JJ, Weingast BR (2009) Violence and social orders: a conceptual framework for
interpreting recorded human history. Cambridge University Press, New York
Nussbaum M, Sen A (1993) The quality of life. Clarendon Press, Oxford
O’Connor A (2001) Poverty knowledge. Princeton University Press, Princeton
Ostrom E (2000) Collective action and the evolution of social norms. J Econ Perspect 14(3):137–158
Ostrom E (2005) Understanding institutional diversity. Princeton University Press, Princeton/Oxford
Ostrom E, Gardner R, Walker J (1994) Rules, games, and common pool resources. University of
Michigan Press, Ann Arbor
Park RE (1928) Human migration and the marginal man. Am J Sociol 33(6):881–893
Pingali P (2003) Sustaining food security in the developing world: the top five policy challenges.
Q J Int Agric 42(3):261–272
Pingali PL, Alinovi L, Sutton J (2005) Food security in complex emergencies: enhancing food
system resilience. J Disaster Stud Policy Manag 29(1):5–24
Polanyi K (1944) The great transformation. The political and economic origins of our time. Beacon
Press, Boston
Preston PW (2002) Development theory: an introduction. Blackwell Publishers, Oxford
Putnam R (1993) Making democracy work: civic tradition in modern Italy. Princeton University
Press, Princeton
Putnam R, Helliwell JF (1995) Social capital and economic growth in Italy. Eastern Econ
J 21:295–307
Rainwater L (1970) Neutralizing the disinherited: some psychological aspects of understanding
the poor. In: Allen VL (ed) Psychological factors in poverty. Markham, Chicago, pp 9–28
Rank MR, Yoon HS, Hirschl TA (2003) American poverty as a structural failing: evidence and
arguments. J Sociol Soc Welfare 30(4):3–29
Roland G (2004) Fast-moving and slow-moving institutions. CESifo DICE Report 2. http://www.
cesifo-group.de/portal/pls/portal/docs/1/1193608.pdf. Accessed 12 Feb 2012
Ryan W (1976) Blaming the victim. Vintage, New York
Schlager E (1999) A comparison of frameworks, theories and models of policy processes. In:
Sabatier PA (ed) Theories of the policy process. Westview, Boulder, pp 233–260
Scott JC (2009) The Art of Not being governed: an anarchist history of upland Southeast Asia. Yale
University Press, New Haven
Sen A (1979) Equality of what? In: McMurrin SM (ed) The tanner lectures on human value.
University of Utah Press, Salt Lake City, pp 195–220
Sen A (1982) Poverty and famines: an essay on entitlements and deprivation. Oxford University
Press, Oxford
Sen A (1992) Inequality reexamined. Harvard University Press, Cambridge
Sen A (1999) Development as freedom. Harvard University Press, Cambridge
Shaw W (1996) The geography of United States Poverty. Garland Publishing, New York
Simon HA (1957) Models of man. Wiley, New York
Simon HA (1959) Theories of decision making in economics and behavioral science. Am Econ
Rev 49:253–258
Small ML, Harding DJ, Lamont M (2010) Reconsidering culture and poverty. Ann Am Acad
PolSoc Sci 629(1):6–27
Smith A (1776) The wealth of nations, reprinted 1997. Penguin Books, London
40 F.W. Gatzweiler and H. Baumüller

Stonequist EV (1937) The marginal man: a study in personality and culture conflict. Charles
Scribner’s Sons, New York
Tobin J (1994) Poverty in relation to macroeconomic trends, cycles, policies. In: Danzinder SH,
Sandefur GD, Weinberg DH (eds) Confronting poverty: prescriptions for change. Harvard
University Press, Cambridge, pp 147–167
UNDP (1998) Human development report 1998. United Nations Development Programme, Oxford
University Press, Oxford/New York
UNESCO (2010) Education for All global monitoring report—reaching the marginalized. United
Nations Educational, Scientific and Cultural Organization, Paris
Valentine CA (1968) Culture and poverty. University of Chicago Press, Chicago
Vatn A (2005) Institutions and the environment. Edward Elgar, Cheltenham
von Braun J, Hill RV, Pandya-Lorch R (eds) (2009) The poorest and hungry: assessments, analyses,
and actions. International Food Policy Research Institute, Washington, DC
von Thünen JH (1826) Der Isolierte Staat in Beziehung Auf Landwirtschaft Und Nationalökonomie.
Perthes, Hamburg
Weber M (2001) Protestant ethic and the spirit of capitalism. Routelege, New York
Weber B, Jensen L (2004) Poverty and place: a critical reivew of ruarl poverty literature. Working
paper 04–03, Rural Poverty Research Center, University of Missouri and Oregon State
University, Columbia
Weiskopf DA (2009) The plurality of concepts. Synthese 169(1):145–173
World Bank (2008) World development report 2009. World Bank, Washington, DC
Wright RD, Wright SN (1972) A plea for a further refinement of the marginal man theory. Phylon
33(4):361–368
Chapter 3
Exclusion and Initiatives to ‘Include’:
Revisiting Basic Economics to Guide
Development Practice

Sajjad Zohir

Abstract  This chapter focuses on exclusion. It reconstructs the definition of exclusion

and its various facets by applying a set of principles that underlie economic analyses.
The analysis reveals that many initiatives to reduce exclusion under the umbrella of
safety nets often lead to the introduction of differentiated products in segmented
markets that may actually contribute to the perpetuation of differentiations within a
population. The relations between contracts, goods and services, and exclusions are
highlighted. A typology of exclusion is described by the author with the help of a
supply and demand analysis of services, including: voluntary exclusion, exclusion
due to a lack of awareness, exclusion for survival, exclusion due to a lack of demand,
and exclusion caused by “distance” such as social exclusion or poor connectivity.

Keywords  Exclusion • Segmentation • Poverty • Safety nets • Development

• Bangladesh

3.1  Introduction1

Literature on the ‘economics of exclusion’ typically addresses competition and the

anti-trust laws of developed economies, not the ‘exclusion’ that development prac-
titioners are concerned about. When economists do engage in the discourse on
exclusion in development, they typically deal with poverty and deprivation without

This paper draws upon ideas put forward in an earlier work by the author and several colleagues
1 

at the Economic Research Group and the BRAC—RED (Bangladesh Rural Advancement
Committee—Research and Evaluation Division) (Zohir et al. 2008) and a presentation by the author
at UN-ESCAP, Bangkok (Zohir 2010).
S. Zohir (*)
Economic Research Group, Dhaka, Bangladesh
e-mail: [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 41
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_3,
© The Author(s) 2014
42 S. Zohir

always applying the basic analytical tools available in economics (Sen 2000;
Osmani 2003). In this chapter I make an effort to reconstruct the definition of exclu-
sion and its various facets by applying a set of principles that underlie economic
analyses. Though poverty is not directly dealt with in this paper, it is hoped that a
concept of exclusion defined within the allegedly ‘narrow’ construct of economics
can be ‘broad’ enough to encompass real deprivation, and provide additional
insight into our understanding of the marginality that underlies the poverty observed
in society.
Concern with exclusion surfaces in several domains of knowledge and related
discourse appears fragmented, with little meaningful exchange across disciplines.
A more comprehensive understanding of exclusion can be achieved by applying the
basic principles of investigation common to various disciplines within economics
and the social sciences. In the first part of this paper I take a cue from a critique of
approaches in economics by orienting the discussion on ‘goods and services’ within
the broader canvass of ‘contracts.’2 In such a setting, I define exclusion as the ‘exclusion
of an agent from the set of all viable contracts in a given sphere.’
There has been an accumulation of rich experiences from development endeavors
that have dealt with targeted programs, where the target populations were often
‘excluded’ groups. Some experiences suggest that there is a plethora of programs
that suffer from inadequate planning; and there are programs that work for one
segment of a target group but fail to reach others. Attempts to identify target groups
and transfer resources or services to them via ‘out-of-system’ approaches have often
fostered local divisiveness instead of leading to the desired reconfiguration of
the social groups and reduced exclusion. In order to take the discourse ahead for
formulating meaningful applicable measures there is a need to find common
language, to reconstruct our ideas around ‘exclusion’ and to reconstruct the observed
phenomena and draw inferences from a ‘model-dependent reality’ (in a comparative
static setting) to guide possible actions.
In the first part of the paper I focus on the reconstruction of basic concepts, and in
the second part I extend the analysis to classify exclusion, illustrated in the context
of a service market and conceptualized with supply and demand tools. The exercise
provides an opportunity to identify various options for reducing exclusion that have
obvious linkages to familiar development programs and policies. The latter reveals
that many initiatives to reduce exclusion under the umbrella of safety nets often lead
to the introduction of differentiated products in segmented markets that may actually
contribute towards perpetuating differentiations within a population. It is hoped that
the proposed approach will improve the ability to design programs that effectively
reduce exclusion (and poverty) as well as to evaluate program performance.3

2 
An earlier attempt to conceptualize exclusion (Zohir et al. 2008) used demand and supply tools
without invoking an underlying ‘system’ with agents, their preferences, and potential exchanges
amongst agents.
3 
The literature recognizes that there are instances of exclusion that exist in the absence of poverty
and instances of poverty that are not rooted in social exclusion. This discussion on exclusion is
general and considers exclusion associated with some form of poverty.
3  Exclusion and Initiatives to ‘Include’: Revisiting Basic Economics… 43

3.2  Conceptual Issues

3.2.1  E
 xclusion is Not Insurmountable: Rationale
for the Undertaking

In all development endeavors over the last two decades or more, either in the guise
of Poverty Reduction Strategy (PRS) and Millennium Development Goals (MDGs)
or in initiatives to achieve greater equity and foster inclusive growth, the general
concern was one of reaching out to those who are ‘left out.’ Earlier concern with
those who are ‘left-out’ receiving benefits from economic growth or those adversely
affected by growth processes, gradually shaped pre-defined social and economic
groups.4 Thus ‘left outs’ could be children not being sent to schools by their parents,
poor rural women unable to avail themselves of health services or market opportunities,
people from minority or socially outcast groups who have little or no access to jobs
and public amenities, and the physically or mentally disabled. I refer to these people
or social groups as ‘excluded,’ and to the broader subject of study as ‘exclusion.’
There are numerous programs under governments, the private sector, or NGOs,
to reduce exclusion, some of which are associated with poverty reduction. The fact
that these programs are often designed based on the assumption that the target
groups were ‘excluded’ from regular service delivery networks and may become
‘included’ through specially designed programs, implies that ‘exclusion’ may be
overcome. There are ample claims of program success, implying the achievement of
‘inclusion’ of individuals and groups that were formerly ‘excluded.’ Acceptance of
such rationale renders exclusion less rigid than the perception that exclusion is
rooted in the static (or slowly changing) framework of culture. This inference from
development practices provides the key entry point to the reformulation of the
analytics within a ‘model-dependent realism.’5 Given an initial equilibrium setting,
exclusion is defined and programs to reduce exclusion are identified by applying
comparative static economic analyses.

3.2.2  F
 ragmented Perspective: Reflections
on Trends in Knowledge Domain

Some question the relevance of abstract economic theory and consider this realm as
too narrow to apply to broader issues. Some advocate a pluralist pedagogy to expose
students “to the competing currents of economic thought, e.g., classical, neoclassical,

4 
Social exclusion long preceded current concerns about economic exclusion. However, social
exclusion may be historically rooted in some economic rationale and concerns about social exclusion
were boosted because such exclusion is believed to have led to economic exclusion when the pace
of economic growth broke away from its historical trend.
5 
The term ‘model-dependent reality’ is borrowed from Hawking and Mlodinow (2010), although
modelling has long been in practice among economists. Long imbued in ‘laws,’ scientists are
increasingly recognizing the need to rely upon ‘model-dependent realism.’
44 S. Zohir

Keynesian, Marxian and institutionalist” (Stilwell 2006). Development practices

have also found inroads into pluralist curricula, with gender, the environment, and
poverty analysis being some of the important ones. All of these are considered
means of making students aware of the different ways of looking at economic
processes, problems, and policies. While diversity opens doors to fresh ideas, our
failure to enrich the core elements of unity often leads to fragmented views, and this
may often lead to divisiveness.6 In a small way the current exercise on exclusion
returns to the basics of economics rooted in early political philosophy and recon-
structs the analytics of an overriding concern of development practitioners with the
hope that some guidelines for action may be derived. The search for common unity
is justified because ‘exclusion’ is being articulated among diverse disciplines that
lack adequate means of communication for broader discussion.

3.2.3  T
 aking a Step Back—A Note on Core Economic
Principles

Engaging in a detailed assessment of the various perspectives on economics involves

the challenging tasks of addressing multiple fields of epistemology and the history
of economic thought, and their roots in early (post European Renaissance) political
philosophy and economic theory of various ideologies. I use a set of principles
below that constitute the core of economics.7 These principles are similar to those
embedded in methods that are deemed necessary for knowledge building in other
fields of social science.
1. A basic analytical category or entity called an ‘agent’ is used to represent entities
such as a social class, households, individuals, communities, governments, or
political parties, etc. Each agent is assumed to have a pre-defined objective. The
choice of agents and their objective ‘functions’ (mappings) are related.8

6 
Interestingly the earlier broad pluralist ‘political economy’ approach became sub-divided into the
‘narrower’ subjects of political science, sociology, economics and many more, and did not re-emerge.
Instead various perspectives articulated in independent model-based abstractions are portrayed as
split views and are often alleged to be ideologically biased. Such a state may unduly perpetuate
divisiveness and obstruct the search for unity.
7 
Often corollaries, conditional upon a set of ‘if conditions’ and drawn from a core set of principles,
are perceived as synonymous with the core itself. Such a perception arises due to our failure to
distinguish between the method of analysis that a discipline applies and the inferences that are
drawn from the results of applying the method.
8 
An agent acts in multiple spaces with multiple objectives and outcomes that are not measurable
in the same units. Thus ‘mapping’ involves relating outcomes with several objectives and choices
across those objectives involve trade-offs, which are often reduced into a single ‘objective
function.’ The choice of agents depends on the purpose of an analysis, however, as in cluster
analysis and ANOVA (analysis of variance), one would expect the least variation in objectives
amongst the elements chosen in a given (set of) agents and greater variation (in objectives) across
different (sets of) agents.
3  Exclusion and Initiatives to ‘Include’: Revisiting Basic Economics… 45

2. Agents are considered ‘rational’ only to the extent that their actions are guided
by the pursuit of (optimizing) some transformed value of their objectives under
a set of constraints. Actions or decisions made by agents may only be defined
over the space in which the objective functions are defined.
3. Exchanges between agents can take place under various structures of power
relations, but are sustainable only if all parties in the exchange ‘gain’ (strictly
speaking) from such exchanges.
These economic principles underlie most, if not all, other disciplines in the social
sciences. Thus in the context of present-day discourse, ‘political economy’ (instead
of political philosophy) may more appropriately capture the discipline under con-
sideration. These principles are also relevant to the field of chemistry, where there
are agents in the form of molecules, atoms or sub-atomic particles and chemical
bondage may be construed as some form of contract between these agents. Scientists
recognize the need to move beyond ‘scientific determinism’ and the ‘system of laws’
to a model-dependent realism that may generate ‘multiple histories’ depending on
the assumptions made.9

3.2.4  Exchanges, Markets, Contracts, and Goods and Services

Exchanges involve transactions in two or more directions that involve two or more
agents. The time and space contingent flows of goods and services that define
exchanges, and the terms and conditions of such exchanges, together constitute
‘contracts,’ and such contracts are negotiated between various agents at different
spheres of life. Within the proposed perspective it is obvious that goods and services
are one aspect of a contract and they are real when they are elements of viable
contracts. Our constructs of time and space (and often quality) contingent
commodities accommodate additional dimensions of contracts, and yet all feasible
elements of a contract are not exhausted.
Contemporary analysis of markets also recognizes the specifics of non-market
exchanges, blurring the distinctions between market and non-market exchanges.
Rhee (2010) distinguishes relational exchanges from market exchanges. Yet the “trust,
friendship, fraternity or solidarity” that defines relational exchanges is also found
in modern-day market exchanges, making it difficult to distinguish non-­market
from market based exchanges. What may constitute an exchange and what kinds
of exchanges define markets, are less formally (and less explicitly) articulated in eco-
nomics textbooks. It is no wonder that ‘classical’ economists vying to break away
to a simpler ground of analytics pondered extensively on such terms as ‘exchange’

What I am reverting to in this paper is the set of core principles underlying such model-building.
9 

It is not akin to physicists’ search for a single theory, rather it outlines the basic elements in building
a theory or method that may generate numerous theories depending on the choice of agents, their
objective functions, and the power relations within which they negotiate contracts.
46 S. Zohir

and ‘value’ before settling with ‘goods and services,’ an ex post manifestation of a
contract, and with a concept of price (financial) that is one of many dimensions of
that contract. While discussing markets, which is often the starting point for an
intermediate textbook, the necessity of identifying a priori, the goods and services
that are being exchanged is realized. However, it is also recognized that all exchanges
are not market-based and thus the definition that the ‘market is a place where
exchanges take place’ appears too hollow.
For our purpose ‘market’ is relevant in an ex post analysis only when there are
predefined goods and services for which there are predefined agents on both sides
(providers and consumers). While such a category is resorted to in the latter part of
the paper, the primary focus is on contracts that embody exchanges, be those via
‘markets’ or not. A more pertinent question relates to the motive behind engaging in
contracts. What drives agents to negotiate contracts with other agents is embedded
in their ‘objectives’ and the ‘constraints’ within which those objectives are realized.
For the purpose of this analysis, I presume that ‘contracts’ and exchanges among
various agents are driven by self-interest.10 Thus there is no transaction other than as
exchanges—that is, there is no unidirectional transfer from one agent to another
without a matching counter-flow.
Historically, discourses on social contracts moved to market-based analyses in
economics, often at the cost of a broader understanding of the subject. Subsequently
the shift in focus from market to contracts in the economic discourse provided new
insights, but largely ran parallel to the market-based analyses with no observable
attempt to reconcile the two.11 A contract defines the characteristics of the goods or
services being exchanged between two or more agents. Thus the search for a viable
contract between two or more agents requires identifying the goods and services
that may be offered. This approach to address a problem is ex ante in nature and
much more akin to traditional ‘business models.’12 In contrast, what emerged due to
the switch to market-based analyses was the dominance of an ex post perspective:
goods and services are considered predefined and exchanges between providers
or producers and consumers take place at terms (prices) that ensure net benefits

Consider Marcel Mauss’s (1967) observations on gifts: “…prestations which are in theory voluntary,
10 

disinterested and spontaneous, but are in fact obligatory and interested. The form usually taken is
that of the gift generously offered; but the accompanying behaviour is formal pretence and social
deception, while the transaction itself is based on obligation and economic self-interest.”
11 
Brousseau and Glachant (2002) note that “the notion of contract is simultaneously broader in
scope and more general than the notion of the market.” Quite contrary to this assertion, discussion
on contracts continues to be empirically pitched within the context of a given market. Clearly the
two concepts of market are different, but what is more general than a market under one perspective
may be a specific aspect of a market otherwise.
12 
Existing textbook definitions often use the ex post perspective on business models. Investopedia.
com (2011), as well as several other websites on business, define business model as “[t]he plan
implemented by a company to generate revenue and make a profit from operations. The model
includes the components and functions of the business, as well as the revenues it generates and
the expenses it incurs.” Thus the framework used to identify a product or service whose market is
yet to be established is also an important component of a business model.
3  Exclusion and Initiatives to ‘Include’: Revisiting Basic Economics… 47

to both parties. Both approaches are considered valid and useful as long as
their corresponding aspects are recognized and the limitations of singular focus on
either is acknowledged.
The observed arrangements of ‘contracts’ among multiple agents in multiple
spheres may be considered as equilibrium outcomes in accordance with the concept
of (abstract) market equilibrium. Underlying an equilibrium are: the agents included
in a model, the assumptions related to their objectives, the constraints they face in
the forms of resource endowments or technology available for production and dis-
tribution (delivery), and the assumptions about power relations amongst the various
agents.13 More importantly equilibrium may include the exclusion of one or more
agents from viable contracts, making exclusion an equilibrium outcome. A new
equilibrium may be established if any of the parameters mentioned above change
(e.g., introduction of a new agent, innovative mechanisms that reduce cost, changes
in objective function, etc.). The resulting new equilibrium may reduce exclusion in
one sphere and increase exclusion in some other sphere. Alternatively viewed, one
may design programs to change the setting in ways that reduce exclusion of an agent
in a given sphere.

3.3  Contracts, Goods and Services, and Exclusion

Suppose there are ‘i’ categories of agents in a society, the i-th being identified as Λi
(i = 1, 2, … i). The simplest case of exclusion arises when there is a provider of a ser-
vice and all but one type of agents avail themselves of the service. That is, contracts
could be negotiated with all of those agents, but no feasible contract could be defined
for (or negotiated with) the ‘excluded’ group. Social exclusion may be interpreted
as exclusion in one space as a result of social contracts in multiple spaces within a
given power relationship. Thus a ‘socially excluded’ group may have contracts in
other spheres of the economy. Only in extreme cases of exclusion, where a group is
alienated by all others, would the excluded group face extinction, or go into complete
isolation and establish an alternative society, or create conflict for the mainstream
(causing negative welfare) so that the latter is forced to renegotiate. In all these cases
there is a set of predefined agents and a power relationship (jointly referred to as a
‘system’), and exclusion may be defined as the presence of an ‘agent’ who is unable
to negotiate a viable contract in a given sphere involving other agents in that ‘system.’
Below we use an applied research effort to provide a skill development program for
urban domestic workers who were ‘excluded’ from regular education to illustrate
the relationships between contracts, goods and services, and exclusion.
The author and several colleagues assessed the prospect of offering a tutoring
and skill development program for domestic workers in the city of Dhaka. In most

It is quite possible that multiple equilibria exist because the set of viable contracts may be large.
13 

The options, however, get reduced as the power relationships get more skewed. We consider the
observed configuration to be a single equilibrium.
48 S. Zohir

cases the domestic workers are school-age females whose parents were financially
compelled to send them to work. Normally the work arrangement does not include
any obligation on the employer to ensure the education of these child workers. It was
noted, however, that there are employers who are keen to support the education of
the children working at their homes. However, these desires did not materialize in
most cases due to the absence of a favorable community environment and the per-
ception among employers that a suitable arrangement could be reached at little or no
cost arising from frequent travels by the children outside of their workplace. In other
words, these workers were ‘excluded’ from traditional education means due to
prevailing contracts among relevant stakeholders. Based on anecdotal evidence it
was postulated that there were employers who would be willing to pay for tutoring
their employees and that a contract (a program with a service package) could be
defined that would reduce exclusion among the working children. The basic framework
for identifying the feasible contract to reduce exclusion is described below.
There were three important stakeholders relevant for the purpose of the exercise:
domestic workers, employers, and the committees that manage the affairs of each
apartment building.14 The variables assigned to the three were ‘w,’ ‘e,’ and ‘c’ respec-
tively. Any resulting program would have to be introduced by a service provider
(represented by ‘s’). Additional terms can be added to describe other variables, such
as: the regulatory regime, other actors (drivers, security guards, etc.) in an apartment
building, and the general environment (law and order situation) in the neighborhood.
While an optimization problem could be used to solve for a possible comparative
static analysis, for our purposes only the incentive compatibility issues were
addressed to assess whether the set of viable contracts was null or non-null. This
allows for remediating the exclusion of domestic workers from some form of educa-
tion. We let ‘Z’ be the vector of the current contract that describes the influence
of domestic worker variables (work hours, types of activities to be performed, pay-
ment received, holidays available, and mobility), employers’ preferences and costs,
the state of community variables (facilities available in the apartment building, etc.),
and service providers (training or class schedule, payments, etc.), which were
initially set to zeros.15

Z = (w1 , w2 , wm , e1 , e 2 , e n , c1 , c 2 , c o , s1 , s2 , sr ) (3.1)

We let Z*0 be the optimal contract prior to the program, and let {Z1} be the set of
all possible contracts involving a service provider to offer education for the domestic
workers.16 If Vij = Vi(Zj) is the perceived net benefits of the i-th agent, (i = w, e, c) in

14 
In order to ensure that the number of domestic workers interested in participating in the program
would be adequate for viable program operations, the focus of the exercise was confined to apart-
ment-based residences only.
15 
Note that there are ‘m’ number of variables for domestic workers (w), ‘n’ number of variables for
employers (e), ‘o’ number of community (c) variables, and ‘t’ number of variables to account for
service providers (s).
16 
Relevant elements in {s}, a subset of {Z1}, will be strictly positive. For example, the number of
hours spent by domestic workers in tutoring classes will be positive.
3  Exclusion and Initiatives to ‘Include’: Revisiting Basic Economics… 49

state j (0 and 1), the search for a viable contract with a service provider amounts to
finding a vector Z*1 such that, Vi(Z*1) ≥ Vi(Z*0), for all i = w, e, and c; and
Vs(Z*1) = the net returns to the service provider that are ≥ normal profit.17 Note that
in the vector {s1, s2, sr} (in Z*1) defines the new goods and services whose introduc-
tion has implications for the values of the elements in the Z-vector.
Initial consultations for the purpose of preparing a questionnaire suggested that
certain aspects of skill development training and education could be offered in ways
that encourage the employer and community stakeholders to enroll their domestic
workers. Specifically, stakeholder willingness was responsive to: variation in the
hours offered for tutoring or training, the age of domestic workers, the venue of
classes, and the characteristics of the service provider. Proponents of the particular
experiment believed that once an initial entry point was utilized and the positive
benefits became apparent to relevant stakeholders, that the exclusion of urban
domestic workers from education would be reduced at a broader level.

3.4  T
 ypes of Exclusion: Illustration with Service
Demand and Supply

3.4.1  General Setting

The presence of agents with whom no viable contract may be negotiated in a given
sphere by any other agent is meant to establish the existence of exclusion. The
search for viable contracts is meant to reduce exclusion. It is however important to
recognize that exclusion may arise for different reasons, necessitating different
remedial measures. It is therefore necessary to categorize exclusion accordingly, so
that programs or actions to reduce exclusion may be appropriately designed. In this
section I use basic demand supply tools to illustrate the various types and options.
Consider a service ‘x.’ Each individual in a population may consume only one
unit of x, or none at all.18 Those who consume x are considered ‘included’ and those
who do not consume x, either voluntarily or involuntarily, are considered ‘excluded.’
Without delving into individual optimization leading to a threshold price below
which one unit of x is availed of by an individual, I directly define a market demand
function as:

D = f (p, c, t, M, ϕ, z); (3.2)

17 
There may be several Z*1 vectors fulfilling the requirement (equivalent to the Ricardian wage-
profit frontier), and the final solution will depend on power relationships and relative access to
information.
18 
Unit demand is assumed to only illustrate the number excluded from a given service and is not
essential to the central argument of the exercise.
50 S. Zohir

where:
p = the price received by provider of the service
c = the additional direct costs to the consumers to use the service
t = the indirect cost to consumers (including opportunity cost) for using the service19
M = the income of the household whose members use the service
φ = the information available to clients, which is less than or equal to Ω (symbolizing
full information)20
z = all other variables (including preference)
The demand function in Eq. 3.2 has the following properties where fi is the first
derivative of D with regard to the i-th argument in the equation:
if f1 < 0, demand increases as provider reduces price
if f2 < 0, demand increases as consumers pays less additional cost
if f3 < 0, as opportunity cost increases, demand decreases
if f4 > 0, demand increases with increase in income (this may reverse beyond a point)
f5 may be positive or negative depending on whether inadequate information lessens
or increases demand. We assume it to be positive on the presumption that increases
in awareness increase demand for the program services.
To identify the various types of exclusion, we assumed z to be given in all illustra-
tions. I consider the population size to be ‘N’ and assume that each individual is eligi-
ble to buy only one unit of the service under consideration. A total of five broad cases
are illustrated, the first three deal with choice-based (voluntary) exclusion, although the
third is structurally imposed and cannot be strictly considered voluntary.

3.4.2  Types of Exclusion

Type 1: choice-based or voluntary exclusion. Prices received by suppliers in the

main market are equal to those paid by end consumers and unlimited supply is
ensured at zero price (see Fig. 3.1).21 This implies that p = c = t = 0. I also assume full
information (φ = Ω) and that M can take any positive value. Therefore D = N1 ≤ N
where N1 is the maximum number that may avail the service if all other factors are
(extremely) favorable and voluntary exclusion = E1 = N – N1.
Type 2: exclusion due to a lack of awareness. Inadequate information about the benefits
from a service (e.g., education) would imply lower effective demand, increasing the
number of people excluded. This is also a case of voluntary exclusion, but rather as a
result of inadequate awareness. Formally, if p = c = t = 0, φ < Ω and M can take any posi-
tive value, then D = N2 ≤ N1 and exclusion is described as E2 = N1 – N2 (see Fig. 3.1).

19 
Prices of all other commodities have not been included, some of which could be considered
under t. For the substitutes of the service under consideration the relation would be opposite.
20 
Misinformation could be deliberately added, which would further complicate the relationship.
The first three cases assume zero costs for ease of illustration. A positive equilibrium price could be
21 

assumed as well, however, that would call for adjustments on account of the changed values of φ and t.
3  Exclusion and Initiatives to ‘Include’: Revisiting Basic Economics… 51

Fig. 3.1  Three types of P

exclusion under zero price Note: All demand curves assume
to borrowers a given level of income (M>0).

D(φ=Ω, t=0)

D(t>0)

D(φ<Ω)

Q/N
N3 N2 N1

Type 3: exclusion for survival (compelled to remain excluded). The consumption of

certain goods and services that a society generally perceives as normal may be con-
sidered a luxury by people who are at the brink of survival. These people, engaged
in extremely low-earning jobs if employed at all, cannot afford the time or resources
to consume the services, even when offered at zero prices. I differentiate this from
voluntary exclusion, since the choice is ‘structurally imposed’ by initial ownership
of endowments. If p = c = 0; φ = φ (N2), t > 0 and M can take finite positive value,
then D = N3 < N2 and E3 = N2 – N3 (see Fig. 3.1).
Type 4: exclusion due to the lack of effective demand under positive prices. If one
could control for all the three types of exclusion discussed above, all forms of social
exclusion and those arising out of costly connectivity (discussed later), we would be
left with the possibility of exclusion as a result of a lack of effective demand. This
case is illustrated in Fig. 3.2 and assumes that prices received by main market
suppliers are equal to those paid by end consumers. Of those agents (consumers)
with positive effective demand at non-negative prices, (q1 q2) are unable to avail
themselves of the service at a market price of P0.
Type 5: exclusion caused by ‘distance’; (5A) social exclusion, and (5B) poor con-
nectivity/ delivery system (or other factors). While the previous four types of
exclusion are fairly straightforward, lumping together ‘social exclusion’ with exclu-
sion caused by physical distance (from growth centers) and/or the lack of appropri-
ate or efficient delivery mechanisms may be controversial. Reducing the concepts of
exclusion (including that of social exclusion) to tractable cost equivalence is not meant
to undermine the importance of the processes that generate exclusion. It is impor-
tant to note that the subject of exclusion is also of interest to development practitio-
ners because they believe that exclusion may be reversed and that there are costs
involved for all such efforts towards inclusion. The equivalence of the two types of
exclusion may be argued on the grounds that both arise due to ‘distance,’ one in
social space and the other in physical space. Thus from a purely technical perspec-
tive (using supply and demand tools), the cost of including the socially excluded is
52 S. Zohir

Fig. 3.2  Exclusion due to P

lack of effective demand

P0

Se

O q1 qe q2 Q

no different from the cost of the ‘delivery’ to a consumer in a distant market. In both
cases an excluded person would have to pay in addition to the ‘normal’ price and the
latter critically depends on the delivery system in place.
The horizontal supply line Se in Fig. 3.2 is meant to represent socially excluded
(SE) people under the assumptions of delivery at zero price and the effective demand
of socially excluded persons are represented by the lowest segment of the demand
curve. There is an interesting twist to the story in this particular case. When SE
people are represented by the lower end of the demand curve, market forces (supply-­
side agents) may be less keen on changing the status quo and they would even be
less keen if inclusion of the SE has negative effects on the willingness to pay at the
upper end of the curve. Clearly increases in resource endowments of cash, assets, or
intellectual capital among SEs will change the distribution of SE along the demand
curve and market forces may be more eager to remove social exclusion.22
A more general case is presented in Fig. 3.3 where I assume that c > 0 (c = P0—P),
meaning that consumers have to pay an additional price to avail themselves of a
service compared to a general (mainstream) consumer. With c = 0 (zero distance
cost), equilibrium price and quantity would be P0 and q0 respectively. If c > 0 and all
buyers have to pay the additional price (captured by a demand curve D1 to the left),
q0—q1 would get excluded and receipt by a provider (P1) would be less than
payments made by the consumer (P3). Innovations in service delivery, improved
connectivity reducing delivery cost, and/or wider acceptance of the formerly
excluded social groups would all push the S curve downward (or shift the D1 curve
to the right towards D0), thus reducing exclusion.

A case in hand is the Dalit in India, a group of people considered ‘untouchables’ or ‘outcastes.’
22 

This paper does not illustrate the case of labor markets, where market returns on investments in
skill development of an otherwise ‘excluded’ group may be shown to be higher.
3  Exclusion and Initiatives to ‘Include’: Revisiting Basic Economics… 53

Fig. 3.3  Exclusion caused P Reduction in exclusion is

by ‘distance’ associated with shifts in D and S
along the directions indicated
by the arrows

P3

P0
P1

D0

D1

0 q1 q0 q2 qn Q

3.4.3  Ways to Reduce Exclusion

Figures 3.1 and 3.2 make evident the kinds of shifts in demand and supply curves
that may reduce exclusion of the six types discussed earlier. Sources of such shifts
are rooted in various kinds of policies and changes in the market environment.
A summary of options to reduce various kinds of exclusion are summarized in
Fig. 3.4. If exclusion is voluntary the urge to reduce exclusion may be justified if the
social benefits from inclusion are perceived to be higher than the costs perceived by
private individuals opting for exclusion. In such cases individual action may be
influenced either by motivating changes in the objective function of excluded agents
or by providing incentives (if the cost of the latter is less than the perceived net
social benefits). Similarly the second and third types of exclusion may be addressed
by promoting awareness through networks and by addressing competing factors
respectively (e.g., excusing school-age children from daily household chores).
Options are rather limited in the first three cases since it is only possible to
address the demand side. In the last three cases options are open from both the
demand and supply sides, as well as through innovations that reduce supply cost—
through costs of production, delivery, or both. A comprehensive list of interventions
is provided in Fig. 3.4.

3.5  Exclusion and Segmentation

Since multiple intervention choices exist for the last three types of exclusion, it is
pertinent to discuss appropriate selection criteria. Traditionally we are introduced to
two basic interrelated criteria that are also used for assessing program performance
or efficiency: the cost of reducing exclusion and the efficacy of targeting. While these
are important, the obvious implications of targeted programs for creating differentiated
spaces or products in the market and society are often overlooked.
54 S. Zohir

Innovation in delivery reducing transaction cost (4)

Innovation introducing differentiated product (3)

Promote/avail agencies with network capital (6)

Social awareness/change objective function (7)

Various ways to
reduce exclusion

Innovation reducing production cost (1)

Address competing factors (9)

Increase income/transfer (8)
Improve connectivity (5)
Price subsidy (2)
Types of Exclusion

1. Voluntary exclusion

2. Lack of awareness

3. Exclusion for survival/competing factors

4. Market exclusion/lack of effective demand

5A. Receipt < payment—social exclusion

5B. Receipt < payment—poor connectivity, etc.

Fig. 3.4  Ways to reduce exclusion

As noted earlier the current availability of goods and services subsumes a balance
between agents with existing contracts (included) and agents with whom no
viable contract could be negotiated (excluded). Any attempt to expand service
coverage beyond a market-determined level calls for new contracts to be negotiated.
In this case the corresponding goods and services may no longer be considered iden-
tical to the old ones. Several cases are discussed below to illustrate this situation.
The case of a perfectly discriminating monopoly is well-known amongst econo-
mists. In this scenario no new product is introduced, but the market is segmented
to ensure greater profit to a monopolist, which also leads to the reduction of exclusion
(in contrast to a competitive solution). Without reductions in the cost of producing
or delivering a given good or service, coverage may be expanded (exclusion reduced)
by introducing new goods and services of lower quality at lower cost.23 Sometimes
funds may be provided to extend an ‘existing’ product or service (e.g., primary
education or sanitary toilets) to a targeted (excluded) population through a new con-
tract with an additional agent (e.g., an NGO). Using a viable contract for reference

23 
There is ample anecdotal evidence of this in markets, such as: housing services, apparel, all kinds
of food, education, and health (most basic necessities).
3  Exclusion and Initiatives to ‘Include’: Revisiting Basic Economics… 55

Table 3.1  Implications of various interventions on segmentations

Intervention type Implication
Innovation reducing production cost (1) Inclusive in nature, no segmentation
Price subsidy (2) If applied to all, inclusive: targeting requires
segmentation
Innovation introducing differentiated Segments, which may be irreversible in the short- to
product (3) medium-term and may percolate into other
spheres
Innovation in delivery that reduces cost Reduces segmentation
of delivery (4)
Improve connectivity/reduce distance (5) Reduces segmentation
Promote/avail services of agencies with If governed properly, reduces cost of delivery and
network capital (6) therefore reduces segmentation
Social awareness (7) If rightly pitched, reduces social differences
Increase income/transfers (8) If accompanied by reduced inequality, reduces
segmentation
Address competing factors (9) Reduces segmentation

and a corresponding product or service, the targeted program is more likely to give
rise to a new contract and a service or product that is not same as the one originally
meant to be delivered!
Figure 3.4 presents nine interventions with potentials to reduce exclusion. The
implications of these interventions for segmentation were described in Figs. 3.1,
3.2, and 3.3, where solutions were sought beyond ‘market-determined’ terms of
exchange. Those are recast with explicit emphasis on segmentation and summarily
presented in Table 3.1 above. At the practitioner level there is uncertainty about
whether efforts towards inclusion can ever be completed. Or stated alternatively,
does product differentiation introduced to ‘include’ subsequently lead to further
segmentation? At the level of analysts and program evaluators a program may be
fine-tuned (or have its performance assessed) on the basis of its degree of neutrality
to or biases towards product differentiation and segmentation.

3.6  Concluding Observation

The inquiry into exclusion I pursued in this paper presumed that exclusion can be
reduced, but ended by showing that there is a cost to most such attempts by way of
introducing segmentation into the economy and society. The discord between textbook
teaching of economics and development practices, further distorted by un-­scrutinized
inclusion of development issues into teaching, was one concern that led to this
paper. The search for a set of core principles underlying various disciplines within
‘political economy’ was justified on the ground that ‘exclusion’ is being dealt with
within those disciplines (breeding fragmented perspectives). Once these principles
were laid out the rest of the paper followed the tradition of deductive reasoning.
56 S. Zohir

Contracts embody exchanges that involve reciprocal transactions and are assumed
to be driven by self-interest. This paper considered contracts to arise between various
agents in a society in multiple economic and noneconomic spheres, and defined
exclusion as the presence of one or more agents with whom viable contracts cannot
be defined. ‘Contract’ as an analytical category was perceived more suitable a tool
for ex ante analysis, whereas ‘goods and services,’ identified as elements of viable
contracts, provide the basis for ex post analysis. Both routes of analysis were pursued
in this paper to gain better insights into exclusion and for the design and assessment
of programs to reduce exclusion. Although the subject was raised, this paper did not
delve into the role of agencies (formal manifestation of agents above micro-levels)
in introducing new forms of contracts. Nor did we pursue the arguments for identi-
fying the areas of unity among various sub-branches of knowledge, particularly in
economics, that are alleged to be ideologically tainted. It is hoped that the perspec-
tives proposed in this paper will open avenues for such discourse in future.

Open Access  This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Brousseau E, Glachant JM (2002) The economics of contracts and the renewal of economics. In:
Brousseau E, Glachant JM (eds) The economics of contract: theories and practices. Cambridge
University Press, Cambridge, pp 3–31
Hawking S, Mlodinow L (2010) The grand design: new answers to the ultimate questions of life.
Bantam Press, London
Investopedia (2011) Defintions. http://www.investopedia.com/#axzz28AKVwQed. Accessed 5
Jul 2012
Mauss M (1967) The gift: forms and functions of exchange in archaic societies. Norton Library,
London
Osmani SR (2003) Evolving views on poverty: concept, assessment and strategy. Poverty and
social development papers no 7, Asian Development Bank, Manila
Rhee SS (2010) Market exchange, relational exchange, and evolution of market institution.
Soongsil University, Seoul
Sen A (2000) Social exclusion: concept, application and scrutiny. Social development paper 1,
Asian Development Bank, Manila
Stilwell F (2006) Four reasons for pluralism in the teaching of economics. Australian J Econ Educ
3(1–2):42–55
Zohir S (2010) Exclusion, service delivery and segmentation: a partial framework to assess MDG
implementation. Presentation at the macroeconomic policy and development division seminar,
UN-ESCAP, Bangkok, 18 August 2010
Zohir S, Rabbani M, Mallik BA, Huq I, Afsana R (2008) Exclusion and poverty: an analytical
approach for understanding exclusion and assessing programmes targeting the very poor in
Bangladesh. BRAC Research and Evaluation Division and Economic Research Group, Dhaka
Chapter 4
Marginality from a Socio-ecological
Perspective

Daniel Callo-Concha, Jan Henning Sommer, Janina Kleemann,

Franz W. Gatzweiler, and Manfred Denich

Abstract The authors analyze the concept of marginality from an ecological perspective
and provide examples of some mechanisms of marginalization. Marginalization
cannot solely be described as an ecological phenomenon, but rather occurs via the
interplay of ecological and social aspects of complex arrangements. Hence the use
of socio-ecological systems as a conceptual unit is proposed. One way to combat
marginalization is to increase the resilience and adaptability of these systems.
However, multiple needs must be considered simultaneously, including: food
security, income generation, or ecosystem services. Research on marginality in the
context of interlinked socio-ecological, complex, and dynamic systems demands
paradigm shifts in scientific disciplines that are beginning to merge.

Keywords Socio-ecological systems • Carrying capacity • Competition • Resilience

• Adaptability

4.1 Marginality in Ecology

The term marginality has become a buzzword across various disciplines and contexts
(Cullen and Pretes 2000). Marginality can only be properly defined in a specific
reference context. In social systems, marginalized people are often defined as
subgroups that differ from the core or mainstream. The core group in this respect is
the reference group that the outlier subgroups are marginal to. In ecological systems
the designation of any of its components as marginalized is a more challenging

D. Callo-Concha (*) • J.H. Sommer • J. Kleemann • F.W. Gatzweiler • M. Denich

Center for Development Research (ZEF), University of Bonn,
Walter Flex Strasse 3, 53113 Bonn, Germany
e-mail: [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 57
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_4,
© The Author(s) 2014
58 D. Callo-Concha et al.

exercise. This is because ecosystems consist of manifold interdependent components

that are under the influence of a variety of biophysical factors.
In ecology marginality has been applied to describe phenomena that occur at
biophysical limits of any kind. These may be geophysical boundaries, environmental
thresholds, or habitats that are not well suited for particular species or populations
(Odum and Barrett 1971; Cassel-Ginz et al. 1997; Leimgruber 2004). Since the
diversity of nature provides a great variety of conditions, margins and marginal
conditions are innumerable. Marginal areas may be environments that pose extreme
challenges to the survival of certain species such as deserts, high altitude areas,
ephemeral water bodies, or sites with heavy metal soils. Stochastic dynamics or
disturbances such as volcanic eruptions, floods, or droughts also create temporary
marginal conditions in ecological terms. The ‘limits’ of ecosystems in most cases
merge into gradients rather than forming sharp boundaries, creating ‘ecotones,’
which are zones of transition between adjacent ecosystems (Holland 1988) where
distinctions are blurred (Odum and Barrett 1971).
To comprehensively describe marginality with respect to an ecosystem requires the
analysis of each of its components and their interactions. Each species has optimal
conditions under which it can best survive and reproduce. The most extreme condi-
tions would be lethal to a species, while less extreme conditions would prevent or
limit growth and development and thus be considered marginal (Begon et al. 2006).
For example, tree growth is determined by different factors such as temperature, soil
conditions, humidity, etc. To determine whether marginality exists in the context of a
forest requires an evaluation of the conditions of individual tree species. Changes in
the status of one species affect the status of other species (positive/negative feedback).
For instance, when resources are shared between competing species the superior com-
petitors profit at the cost of inferior ones (Araujo and Pearson 2005; Soberon 2007).
At the species level, Liebig’s law of the minimum states that the scarcest required
resource determines the overall suitability of living conditions in a given area. With
respect to the occurrence and abundance of plant or animal species, population
growth and size are limited by the availability of the scarcest required resource or
‘limiting factor.’ Therefore it makes sense to describe marginalization in relation to
the carrying capacity of the ecosystem, which is the maximum population size of a
species that can be sustained indefinitely in a given area. In addition, the carrying
capacity for one species is strongly influenced by the presence and relative abundance
of other species that compete for the same resources. For example, nutrient availability
limits the growth and size of populations of two species of phytoplankton in a
pond. The population size of either species will depend on both nutrient availability
and the ability of each competing species to procure these nutrients. If one species
can access the scarcest nutrient more proficiently it will out-compete the other
species in the long run if conditions and nutrient availability are constant over space
and time (provided that population growth of the better competitor does not
change with increases in population size, i.e., their increased abundance does not
increase mortality via greater incidence of disease or predation rate). Therefore
a species can be marginalized by the presence of a competing species despite the
overall availability of required resources in an ecosystem.
4 Marginality from a Socio-ecological Perspective 59

Ecosystems can also be classified by the limiting biophysical factors at different

hierarchical levels. For instance, edaphic conditions are more relevant at fine scales,
while climatic and orographic factors are more relevant at broader scales (Pearson
and Dawson 2003). A translation of this relationship directly to social systems
would describe social marginalization as a result of overexploitation of a system’s
resources, often due to either high human population density or else decreased
resource availability. Hence a relevant way of equating the concept of ecosystems to
social marginality is that they are analogous to social systems in which marginality
occurs when certain groups have only limited access to resources.
A direct link between marginality in ecological and social systems can be made
when considering the concept of biological value and life regulation in living systems
(Damasio 2010). The primitive of biological value (Damasio 2010, 2011) is the
physiological state of living tissue within a homeostatic range that is necessary for
normal function (i.e., health). Extremes of homeostatic ranges have low biological
value to them, whereas more central conditions along homeostatic ranges have higher
biological values. Thereby biological value is directly linked to need, and need is linked
to life. The productive capacity of organisms can be reached in healthy organisms
and environments, whereas it will always be underperforming in marginal environments
with low biological value.
The propensity of humans to attach values to virtually everything around them
relative to their desired living conditions can be perceived as a process which aims
at achieving ideal homeostatic conditions. Humans formulate values that regulate
the processes (economic and social) within their homeostatic range and thereby
contribute to their well-being. Staying alive requires processes of transforming
nutrients into energy, disposal of waste, and making use of energy for biological
processes under particular biophysical conditions. Departures from homeostatic
ranges are detected by the brain, which can stimulate corrective actions. The process
of cognition (in simple organisms such as earthworms) and consciousness (in more
complex organisms such as humans) is used to monitor and detect whether the body
is operating within this range or else in danger—within or outside the homeostatic
range (Parvizi and Damasio 2001). A situation where an organism is in danger due
to homeostatic imbalance can be conceived as being analogous to organisms living
in marginal conditions.

4.2 Socio-ecological Systems and Marginality

Evidently humans have decoupled themselves from purely physiological and

competition-related ecological limitations due to technological developments that
prop broad adaptive strategies. As a result humans are able to survive in areas that
are otherwise unsuitable for them from an ecological point of view. This may, however,
put populations that live close to (or beyond) the margins of ecological suitability at
risk and make them dependent on inputs from external sources. Extreme examples
are human settlements under completely hostile conditions such as polar or space
stations that depend entirely on external inputs and modern technologies.
60 D. Callo-Concha et al.

A couple of assumptions can be made regarding how ecological factors may

determine social marginality and how boundaries in ecological marginality are similar
to those in social marginality. We assume that marginalized people often lack access
to resources due to unfavorable (geographical) location or generally restrictive local
biophysical conditions (Gatzweiler et al. 2011). For instance, remoteness in the sense
of living far away from economic centers may result in social marginality by limiting
access to work, education, and health care (Leimgruber 2004). From an anthropocen-
tric point of view, ecological systems provide goods and services that their inhabitants
use to ensure their well-being. Constraints in resource procurement determine the
extent of marginalization. Examples are farming in isolated areas, in areas with
limited access to water, or where soils or other conditions are of limited suitability.
Complementarily, people may become marginalized as the result of ecological degra-
dation such as desertification, acidification, or salinization of soils, and air or water
pollution, etc. In these cases marginalization is the result of a gradual process.
Assuming that unfavorable ecological conditions contribute to the general
marginality of a human settlement, there may also be privileged individuals who are
able to increase their resilience and adaptive capabilities at the expense of less privi-
leged individuals or of common resources. In environments with poor or depleted
biophysical assets, forms of social exclusion often manifest (Winchester and White
1988; Gatzweiler et al. 2011). These considerations suggest that it is appropriate to
enlarge the scope of analysis over a broader realm in order to understand ecological
marginality and the ‘socio-ecological systems’ (SES) concept.
It is generally acknowledged that understanding complex phenomena requires
insights from multiple scientific disciplines, but not until the 1970s in the context of
greater awareness of the global environmental crisis did an appropriate conceptual
framework for considering ‘mingled phenomena’ emerge (Vayda and McCay 1975).
It was during this period that the concepts such as ‘lifescape,’ ‘livelihood,’ ‘coupled
human-environmental,’ and ‘human-natural systems’ were first proposed (Howorth
1999; Marschke and Berkes 2006). Recently SES has emerged as a concept inte-
grating human-natural interrelationships that is used to harmonize social development
and conservation goals. By asserting that many complex phenomena include both
social and ecological systems simultaneously and inseparably (Gallopin 2006), and
therefore any demarcation is artificial and arbitrary (Folke 2006), the application
of crosscutting and integrated approaches such as the concept of marginality are
advocated for the consideration of complex issues.
Based on this concept it is worth identifying situations where social marginalization
occurs due to ecological variables or ecosystem settings. People can be marginalized
due to environmental factors that inhibit their well-being. These factors generally
act on distinct spatial and temporal scales due to variation in the physical and chemical
characteristics of ecosystems (degraded soils, salinity, toxic pollutants, etc.). All of
these factors may be used to describe the degree of marginality to which the inhabitants
of an ecosystem are subject. For example, many tropical areas of Africa are affected
by sleeping sickness/animal trypanosomiasis, which is transmitted by tsetse flies,
making them marginal areas to inhabit and raise cattle: pastoralists who only have
access to these areas are marginalized by environmental factors.
4 Marginality from a Socio-ecological Perspective 61

People can be marginalized by limited access to natural resources due to inherent

variability of the amount and quality of the natural resources in a particular ecosystem
(Landres et al. 1999). Constraints on an ecosystem that limit the amount of resources
available to residents after having passed a physiological threshold (e.g., due to high
human population density), can result in the marginalization of the inhabitants.
Resource depletion can also occur due to the degradation of ecosystems by overexploi-
tation (e.g., soil erosion, deforestation, extirpation of wildlife, etc.) or other means
and lead to degradation-induced marginality.
Marginalization also occurs in SES through competition. This may occur directly,
as when one actor takes land or resources from another, or through indirect mecha-
nisms such as the appropriation of land by central governments (Cotula et al.
2009). Marginalization in a SES may occur consensually, for example the resettle-
ment of populations for the construction of hydroelectric dams that contribute to the
general well-being in a society.
The cases above reveal how marginality can be complex and often deal with SES
that are subject to risk and uncertainty (Leach et al. 2007). Moreover the drivers of
marginalization respond to changing socio-ecological factors and are often uncon-
trollable by the affected people, as the inherent causal mechanisms emerge on
spatial, temporal, or administrative scales that the affected people have limited con-
trol over. Strengthening the resilience and adaptability of SES, however, helps to
prevent the marginalization of its inhabitants.

4.3 Addressing Marginality in Socio-ecological

Systems via Resilience and Adaptability

It has been found that human communities, particularly when overpopulated, often
overuse local natural resources, thus aggravating poverty through natural resource
depletion (Dasgupta and Mäler 1994). This feedback loop contributes actively to the
marginalization of the poor by limiting their resource options (UNCCD 2003). This
has been documented in the case of West African savannah agricultural systems that
are constrained by limited soil productivity and highly variable precipitation patterns
(Sanchez 2002; Sanchez et al. 2002; Challinor et al. 2007), in the case of slash-and-burn
agriculture in rainforests at historically high population densities leading to deforesta-
tion and the exhaustion of soils (Palm et al. 2005), and in the case of the migration
of Andean farmers to less productive, higher altitude areas due to reduced local
availability of agricultural land at more favorable altitudes (Mayer 2002).
The difficulties of escaping such poverty/natural-resource-depletion feedback
loops are aggravated by global drivers such as human population growth, habitat
degradation and destruction, climate change, the global economic crisis, and their
combined impacts. In this context the sustainable performance of a SES requires
a dynamic configuration that can adjust to progressive or sudden changes (Adger
2006). Therefore considering marginality in SES requires taking into account
62 D. Callo-Concha et al.

Fig. 4.1 The adaptive cycle: successive phases of response of a complex system to changing conditions
(Gunderson and Holling 2002, 34; Reproduced by permission of Island Press, Washington, DC)

change and disturbance as the principal constraints that can be considered using the
concepts of resilience and adaptability (Dilley and Boudreau 2001; Vogel et al. 2007).
Resilience was originally coined in ecology to describe the capacity of ecosys-
tems to recover functionality after shocks (Holling 1973), and may be used to refer
to the capacity for renewal and reorganization of a SES (Gunderson and Holling
2002; Berkes et al. 2003). It is conceived of as a system’s capacity to maintain its
identity, structure and function in response to a disturbance event (Walker and
Meyers 2004). Adaptability is the capacity to adjust to change (Smit and Wandel
2006) or in other words the ability to remain in a stable configuration despite changes
to internal or external factors over time (Denevan 1983; Andresen and Gronau 2005).
Despite coming from ecology, the concepts of resilience and adaptability have
started to be used in interdisciplinary contexts (Janssen and Ostrom 2006; Vogel
2006) due to their utility for portraying the drivers of sustainability and long-term
functionality of complex systems (Callo-Concha and Ewert 2011). In this frame-
work a marginal SES is a system with limited or no resilience. Marginal systems are
vulnerable in the sense that their stability is menaced by events that push them
towards a threshold after which the system must either reach a new homeostasis
under different conditions or else become unstable.
The adaptive cycle (Fig. 4.1) is a conceptual tool that portrays the successive
stages of systems in response to changing conditions (Gunderson and Holling
2002). In the case of poverty and natural-resource-depletion feedback loops, the
adaptive cycle can help to identify ‘poverty traps’ or situations where system diversity,
connectivity, and resilience are reduced to the extent that system performance
cannot be regained or is destabilizing. Gunderson and Holling (2002) provide
examples of a productive savannah that became degraded due to drought and over-
grazing fueled by economic incentives for maintaining high stocking densities.
Other examples of system collapse are societies traumatized by civil war, where
4 Marginality from a Socio-ecological Perspective 63

social and cultural cohesion and adaptive abilities have been lost, preventing the
rebuilding of society (Volkan 2000). In the case of ‘rigidity traps,’ decision circuits
are highly interconnected, reinforcing themselves and making the system inflexible
(Gunderson and Holling 2002). A good example is an old forest where larger, well-
established trees inhibit the growth of saplings, impeding the renewal of the forest
as a whole and increasing vulnerability to disturbances such as wildfire (Carpenter
and Brock 2008). A corresponding social example would be the case of authoritar-
ian governments where vertical decision-making structures impede innovation and
adaptation, that tend to collapse when the overall settings such as democratic values
become gradually hegemonic.
Preventing the marginalization of a human community by increasing its resilience
and adaptive capacity cannot be achieved by concentrating on a single component
without considering other interlinked components of the SES. Contrarily it must
fulfill various complementary and successive goals (e.g., guaranteeing food security,
creating income generating opportunities, and maintaining ecosystem services)—
goals which aim at improving the conditions for living systems, either social or
ecological. In consequence any intervention to prevent and eventually reduce
marginality should consider the functionality of the components of the SES and the
involved processes. Moreover these efforts need to acknowledge the complexity of
SES and therefore adopt systemic principles that allow the identification and
characterization of the components, and consideration of the context, boundaries,
connectedness, feedback, inflows and outflows as proposed by the ‘Ecosystem
Approach’ (Waltner-Toews et al. 2008), and further elaborated by Gatzweiler et al.
(2011) in the context of social marginality.

4.4 Outlook

We have discussed basic biological, ecological, and social dimensions of marginality

from a systems perspective. We support the assertion that understanding human-
environmental interrelationships demands a broader conceptualization such as the
SES concept. Furthermore, marginalization appears difficult to reverse once complex
systems have produced states which are far outside of normal homeostatic ranges.
One way to combat marginalization is to increase the resilience and adaptability of
the SES. This can be done (e.g., by valuation, feedback, and monitoring), however,
multiple needs must be considered at the same time that refer to the ideal condition
for human and non-human life to prevail, mainly: food security, income generation,
energy provision, and the maintenance of diversity. Research on marginality in the
context of interlinked, complex, and dynamic socio-ecological systems demand
paradigm shifts in scientific disciplines that are beginning to merge. New research
networks and funding policies should address marginality by promoting interdis-
ciplinary, systems-based, and practical (problem-solving) approaches to provide
better decision-making arguments.
64 D. Callo-Concha et al.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Adger W (2006) Vulnerability. Glob Environ Chang 16(3):268–281

Andresen K, Gronau N (2005) An approach to increase adaptability in ERP systems. In: Khosrow-
Pour M (ed) Managing modern organizations with information technology: proceedings of the
2005 Information Resources Management Association international conference, San Diego,
Idea Group Publishing, Herschey, 15–16 May 2005, pp 883–885
Araújo MB, Pearson RG (2005) Equilibrium of species’ distributions with climate. Ecography
28:693–695
Begon M, Townsend C, Harper J (2006) Ecology: from individuals to ecosystems, 4th edn.
Blackwell Publishing, Malden
Berkes VF, Colding J, Folke C (2003) Navigating social-ecological systems. Cambridge University
Press, New York
Callo-Concha D, Ewert F (2011) Assessing multifunctional agroforestry systems applying the
Resilience/Vulnerability and Adaptability (RVA) approach. In: Becker M, Kreye C, Ripken C,
Tielkes E (eds) Tropentag 2011, book of abstracts, international research on food security, natural
resource management and rural development, Bonn, 5–7 October 2011. DITSL GmbH,
Witzenhausen. http://www.tropentag.de/2011/proceedings/ proceedings.pdf. Accessed 25 Jan 2013
Carpenter SR, Brock WA (2008) Adaptive capacity and traps. Ecol Soc 13(2):40
Cassel-Gintz MA, Lüdeke MKB, Petschel-Held G, Reusswig F, Plöchl M, Lammel G, Schellnhuber HJ
(1997) Fuzzy logic based global assessment of the marginality of agricultural land use. Clim
Res 8(2):135–150
Challinor A, Wheeler T, Garforth C, Craufurd P, Kassam A (2007) Assessing the vulnerability of
food crop systems in Africa to climate change. Clim Chang 83(3):381–399
Cotula L, Vermeulen S, Leonard R, Keeley J (2009) Land grab or development opportunities:
agricultural investment and international land deals in Africa. International Institute for
Environment and Development, Food and Agriculture Organization of the United Nations and
International Fund for Agricultural Development, London/Rome
Cullen BT, Pretes M (2000) The meaning of marginality: interpretations and perceptions in social
science. Soc Sci J 37(2):215–229
Damasio A (2010) Self comes to mind. Constructing the conscious brain. Pantheon Books, New York
Damasio A (2011) Neural basis of emotions. Scholarpedia 6(3):1804
Dasgupta P, Mäler KG (1994) Poverty, institutions and the environment-resource base. World bank
environment paper 9, World Bank, Washington, DC
Denevan WM (1983) Adaptation, variation and cultural geography. Prof Geogr 35(4):399–406
Dilley M, Budreau T (2001) Coming to terms with vulnerability: a critique of the food security
definition. Food Policy 26(3):229–247
Folke C (2006) Resilience: the emergence of a perspective for social–ecological systems analyses.
Glob Environ Chang 16(3):253–267
Gallopin GC (2006) Linkages between vulnerability, resilience, and adaptive capacity. Glob
Environ Chang 16:293–303
Gatzweiler F, Baumüller H, von Braun J, Ladenburger C (2011) Marginality: addressing the root
causes of extreme poverty. ZEF working paper no 77, Center for Development Research,
University of Bonn, Bonn
Gunderson LH, Holling CS (2002) Panarchy: understanding transformations in human and natural
systems. Island Press, Washington, DC
4 Marginality from a Socio-ecological Perspective 65

Holland MM (1988) SCOPE/MAB technical consultations on landscape boundaries report of a

SCOPE/MAB workshop on ecotones. Biol Int Spec Issue 17:47–106
Holling C (1973) Resilience and stability of ecological systems. Annu Rev Ecol Syst 4:1–23
Howorth C (1999) Rebuilding the local landscape: environmental management in Burkina Faso.
Ashgate Publishing, Farnham
Janssen MA, Ostrom E (2006) Resilience, vulnerability, and adaptation: a cross-cutting theme of
the international human dimensions programme on global environmental change. Glob Environ
Chang 16(3):237–239
Landres P, Morgan P, Swanson F (1999) Overview of the use of natural variability concepts in
managing ecological systems. Ecol Appl 9(4):1179–1188
Leach M, Scoones I, Stirling A (2007) Pathways to sustainability: an overview of the STEP centre
approach. STEPS approach paper. STEPS Centre, Brighton
Leimgruber W (2004) Between global and local: marginality and marginal regions in the context
of globalization and deregulation. Ashgate Publishing, Farnham
Marschke MJ, Berkes F (2006) Exploring strategies that build livelihood resilience: a case from
Cambodia. Ecol Soc 11(1):42–58
Mayer E (2002) The articulated peasant: household economies in the Andes. Westview Press, Boulder
Odum E, Barrett G (1971) Fundamentals of ecology. Thomson Brooks, Cole
Palm C, Vosti S, Sanchez P, Ericksen P (eds) (2005) Slash-and-burn agriculture: the search for
alternatives. Columbia University Press, New York
Parvizi J, Damasia A (2001) Consciousness and the brainstem. Cognition 79:135–159
Pearson RG, Dawson TP (2003) Predicting the impacts of climate change on the distribution of
species: are bioclimate envelope models useful? Glob Ecol Biogeogr 12:361–371
Sanchez P (2002) Soil fertility and hunger in Africa. Science 295:2019–2020
Sanchez P, Jama B, Niang A, Palm CA (2002) Soil fertility, small-farm intensification and the
environment in Africa. In: Lee DR, Barett CB (eds) Tradeoffs or synergies? agricultural inten-
sification, economic development and the environment. Commonwealth Agricultural Bureau
International, New York, pp 325–344
Smit B, Wandel J (2006) Adaptation, adaptive capacity and vulnerability. Glob Environ Chang
16(3):282–292
Soberón J (2007) Grinnellian and Eltonian niches and geographic distributions of species. Ecol
Lett 10:1115–1123
UNCCD (2003) Confronting land degradation and poverty through enhanced UNCCD implementation.
Discussion paper prepared by the global mechanism of UNCCD for the sub-regional partnership
building forum for the central Asian republics, Tashkent, Uzbekistan. United Nations
Commission to Combat Desertification, Bonn
Vayda AP, McCay BJ (1975) New directions in ecology and ecological anthropology. Annu Rev
Anthropol 4(1):293–306
Vogel C (2006) Foreword: resilience, vulnerability and adaptation: a cross-cutting theme of the
international human dimensions programme on global environmental change. Glob Environ
Chang 16(3):235–236
Vogel C, Moser SC, Kasperson RE, Dabelko GD (2007) Linking vulnerability, adaptation, and
resilience science to practice: pathways, players, and partnerships. Glob Environ Chang
17(3–4):349–364
Volkan VD (2000) Traumatized societies and psychological care: Expanding the concept of preventive
medicine. Mind Hum Interact 11:177–194
Walker B, Meyers JA (2004) Thresholds in ecological and social–ecological systems: a developing
database. Ecol Soc 9(2):3–19
Waltner-Toews D, Kay JJ, Lister NM (2008) The ecosystem approach: complexity, uncertainty and
managing for sustainability. Columbia University Press, New York
Winchester HPM, White PE (1988) The location of marginalised groups in the inner city. Environ
Plann D Soc Sp 6(1):37–54
 Part II
Dimensions and Prevalence of Marginality
Chapter 5
Mapping Marginality Hotspots

Valerie Graw and Christine Husmann

Abstract In this chapter the authors applied innovative Geographical Information

Systems mapping techniques to illustrate spatial dimensions of marginality at
continental and regional levels. They sought to make the marginalized and poor
more visible by identifying areas where many poor people live under difficult
biophysical and socio-economic conditions. A broad set of variables covering
ecological, social, and economic dimensions were described using existing datasets
to identify ‘marginality hotspots’ which were then overlaid with poverty distribution
data. Areas where a high percentage of poor people coincided with marginality
hotspots were found in Central and South East Africa, especially the northern parts
of Niger and in Chad, the Central African Republic, the Democratic Republic of the
Congo, Mozambique, Malawi, and Burundi.

Keywords Marginality • Poverty • GIS • Area identification

5.1 Why Do We Map Marginality?

Maps are a powerful tool for presenting information in a way that is easily comprehensible
by a non-specialist audience. Maps encourage visual comparison and make it easier to look
for spatial trends, clusters or other patterns. Maps are therefore useful not only to governments
and decision makers, but also to the local communities (Deichmann 1999, 3).

Historically the first and still one of the most famous examples of using geospatial
analysis for mapping causal linkages is the cholera map of London made in 1854.
By mapping information about drinking water, pumps, and the number of cholera
victims, John Snow, an English physician, identified a positive relationship between

V. Graw (*) • C. Husmann

Center for Development Research (ZEF), University of Bonn,
Walter Flex Strasse 3, 53113 Bonn, Germany
e-mail: [email protected]; [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 69
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_5,
© The Author(s) 2014
70 V. Graw and C. Husmann

drinking water and the spread of cholera (Kriz 2010). Today technologies and the
development of Geographic Information Systems (GIS) allow us to easily demon-
strate simple geographic spatial relationships and to analyze more complex ones.
Mapping and GIS were applied here to illustrate spatial dimensions of marginality
at continental and regional levels. We hereby sought to make the marginalized and
poor more visible by identifying areas where many poor people live under difficult
biophysical and socio-economic conditions. A broad set of variables covering
ecological, social and economic dimensions were identified for this purpose and
we used existing datasets to describe each variable in the first step of our marginality
mapping approach. We focused on Sub-Saharan Africa (SSA) and South Asia (SA),
where most of the world’s poor, and in particular the poorest, live (Ahmed et al.
2007; von Braun et al. 2009).
The number of extremely poor and hungry people remains unacceptably high.
Being excluded from economic growth and other dimensions of societal develop-
ment are indications of the extreme poor existing at the margins of society, which
may trigger a downward spiral into poverty (Gatzweiler et al. 2011). Marginality is
frequently cited as a root cause of poverty (von Braun et al. 2009) and is a complex
issue that is not amenable to simple solutions or answers. It may be defined as:
an involuntary position and condition of an individual or group at the margins of social,
political, economic, ecological and biophysical systems, preventing them from access to
resources, assets, services, restraining freedom of choice, preventing the development of
capabilities, and eventually causing extreme poverty (Gatzweiler et al. 2011, 3).

Marginality thus helps explain why individuals or groups are excluded from, or do
not have access to processes or resources that would otherwise free them from extreme
poverty. Single causal factors alone are not sufficient to explain marginality, which is
seen more as a network of contributing causal factors that together lead to extreme
poverty. Marginality is not only multidimensional with regard to the causation of pov-
erty, but also multi-relational with regard to the character of the network of causal
relations. We identified geographic areas where multiple dimensions of marginality
overlap and described the nature of the observed overlap. If the reasoning that
marginality contributes to poverty is true, then the areas where multiple dimensions of
marginality overlap should also be areas where many people are extremely poor.

5.2 Marginality Hotspots

5.2.1 Finding Proxies for Marginality Indicators

on a Global Scale

Given that marginality is a complex and multifaceted phenomenon, we included a

broad set of variables covering ecological, social, and economic dimensions of
human well-being in the focus regions. These “marginality dimensions” were based
on the “spheres of life” defined in Gatzweiler et al. (2011, 13), including: “Economy”;
“Quality of life”; “Landscape design and infrastructure”; “Ecosystems, natural
5 Mapping Marginality Hotspots 71

resources, and climate”; “Public domain and institutions”; and “Demography.” For
the purpose of this mapping exercise, single indicators were used to represent each
of the spheres. Here the spheres “Landscape design, land use, and location” and
“Infrastructure” are both captured by the single indicator “accessibility”, and the
sphere “Behavior and quality of life” is represented by stunting.
For each dimension a cut-off point along a range of indicator values was used to
define the threshold below which an area was considered to be marginal. Indicator
layers for each of the different dimensions of marginality were overlaid to find the
areas where multiple layers of marginality overlap. We defined a ‘marginality
hotspot’ as an area in which at least three dimensions of marginality overlapped.
The maps were based on national and sub-national data published by the World
Bank, the Food and Agriculture Organization of the United Nations (FAO), Harvest
Choice, and others. Table 5.1 provides a detailed overview of the data used, the data
sources, and the cut-off points chosen for each indicator.

5.2.1.1 Economic Dimension

Gross National Income (GNI) per capita (in current US$) was used to represent the
‘Economy’ dimension rather than Gross Domestic Product (GDP) per capita, which
is more commonly used for analyzing economic well-being (Syrquin 2011). This
approach follows the United Nations Development Programme (UNDP), 2010 Human
Development Report in which GDP was replaced by GNI as a more representative
standard of living indicator. In contrast to GDP per capita—which only gives informa-
tion on the monetary value of goods and services produced in a country but excludes
information on how much is retained within the country—GNI also includes inter-
national flows such as remittances and aid, and thereby represents a more accurate
measure of a country’s economic welfare (UNDP 2010).1 The World Bank also uses
GNI per capita as a key indicator for classifying economies into high-, middle-
(subdivided into lower middle- and upper middle-income), and low-income countries
(World Bank 2011).
The World Bank divides countries by GNI per capita into different income level
categories using the World Bank Atlas method (World Bank 2011).2 Low-income
countries are those countries with a GNI per capita of US$1,005/year or less, lower
middle-income countries are from US$1,006 to US$3,975/year, upper middle-
income countries are from US$3,976–12,275/year, and high-income countries are
at US$12,276/year or more. For the purpose of marginality mapping we set the cut-
off point for the economic dimension at US$1,005/year or less per capita, which is
the World Bank threshold for low-income countries.

1
See also: http://hdr.undp.org/en/statistics/hdi/
2
The World Bank Atlas Method includes the use of an atlas conversion factor which should reduce
the impact of exchange rate fluctuations in the cross-country comparison of national incomes
(http://go.worldbank.org/ IEH2RL06U0).
72 V. Graw and C. Husmann

Table 5.1 Proxies used for mapping marginality hotpots

Dimension I: Economy
Indicator: Gross National Income (GNI) per capita PPP (current US$)
Input: World Bank Data 2010, visualized and geo-processed in ArcGIS (GIS software
from ESRI (Environmental Systems Research Institute)
Cut-off point: US$1,005 GNI per capita World Bank definition for a low-income country
(≤US$1,005)
Source: World Bank (compiled data from 2008 to 2010)

Dimension II Behavior and quality of life

Indicator: Prevalence of stunting among children under age five, by lowest available
subnational administrative unit, varying years (FGGD)
Input: Global raster data layer with 5-arc-minutes resolution, data compilation by FAO
including the prevalence of stunting, LandScan global population database
and the percentage of stunted children under age 5
Cut-off point: Prevalence of stunting among children under 5 >50 %, which is the FGGD
definition for ‘very high’ stunting prevalence
Source: FAO 2007—the data are based on sources compiled by UNICEF, the map was
created within the FGGD Digital Atlas

Dimension III Landscape design, land use and location/Infrastructure

Indicator: Travel time to major cities—a global map of accessibility
Input: Infrastructural data (based on data on populated places, cities, road networks,
travel speeds, railway networks, navigable rivers, major water bodies,
shipping lanes, borders, urban areas, elevation, and slope)
Cut-off point: More than 10 h travel time to the nearest city of 50,000 people or more
Source: Nelson 2000 (http://bioval.jrc.ec.europa.eu/products/gam/sources.htm)
Dimension IV Ecosystems, natural resources and climate
Indicator: Global land area with soil constraints
Input: Soil depth, chemical status, fertility, drainage, texture, and miscellaneous
constraints
Cut-off point: Soils that have ‘frequent severe’ and ‘very frequent severe’ soil constraints as
well as ‘unsuitable for agriculture’ according to FAO 2007 classification
Source: FGGD, IIASA 2000 study on Global Agro-Ecological Zones (GAEZ) (van
Velthuizen et al. 2007)
Dimension V Public domain and institutions
Indicator: Political stability, governance indicator
Input: perceptions of the likelihood that the government will be destabilized or
overthrown by unconstitutional or violent means, including political violence
or terrorism (Thomas 2009, 5)
Cut-off point: Lowest third-quantile
Source: World Bank 2009 (http://info.worldbank.org/governance/wgi/index.asp)

5.2.1.2 Demography and Quality of Life Dimension

‘Stunting,’ which refers to low height for a particular age resulting from malnutrition
over a long period (de Onis et al. 2011) was used to represent the ‘Behavior and
quality of life sphere’. Children were defined as stunted if their height was below
5 Mapping Marginality Hotspots 73

the fifth percentile for their age group in their reference population (Lewit and
Kerrebrock 1997). Stunting is a result of chronic malnutrition and thus a good
overall indicator of health and hunger, as it reflects the cumulative effects of long
term nutritional deficiency (Yohannes et al. 2010; Syrquin 2011).
The ‘Prevalence of stunting among children under 5 by lowest available subnational
administrative unit, varying years’ dataset (FAO 2007) was produced by the FAO
‘Food Insecurity, Poverty, and Environment’ (FGGD) project.3 The data was compiled
from different sources such as ‘Demographic and Health Surveys,’ the ‘Multiple
Indicator Cluster Survey’ of the United Nations International Children´s Fund
(UNICEF/MICS), the World Health Organization (WHO), the Global Database
on Child Growth and Malnutrition, as well as national-level surveys. According to
the FGGD dataset stunting is rated as ‘very high’ when the prevalence of stunting
among children under age 5 is more than 50 %. We used this threshold for the
“Behavior and quality of life” dimension of marginality.

5.2.1.3 Landscape Design and Infrastructure Dimension

An interesting approach developed by Nelson (2009) that measures accessibility

via the travel time to the nearest major city was used to represent the ‘Landscape
design and infrastructure’ sphere. Accessibility was defined as “the travel time
to a location of interest using land (road or off-road) or navigable water based
travel” (JRC 2010). To calculate travel time, a friction surface had to be devel-
oped that included any geographic features of interest for the analysis. The key
indicators of this approach are sources of agglomeration economics, and include:
population size, population density, travel time, land-cover, and elevation
(Hirotsugu and Nelson 2010). An enumeration of all input variables is shown in
Table 5.1.
The cut-off point for this dimension was set at the point where more than 10 h
of travel time were required to reach the nearest city of 50,000 or more people.
The size of 50,000 was based on Uchida and Nelson (2008), who defined settle-
ments with 50,000 inhabitants as large. We chose 10 h of travel time—a rela-
tively high value—as the cut-off-point for an approximation of a travel time
required to reach a ‘distant’ city in developing countries. In some countries of
SSA people travel days to reach hospitals with specialized services or markets
for special goods and services while in other countries—such as Bangladesh—
the travel time to the nearest large city is much lower. Therefore we tried to find
an approximate travel time as it was difficult to find a precise definition of ‘distant’
as used in the literature.

3
The Food Insecurity, Poverty, and Environment Global GIS Database (FGGD) was also implemented
by FAO (as FIVIMS) as an initiative to improve the use of disaggregated spatial information on
different scales, global- and national-levels (see also: http://geonetwork3.fao.org/fggd).
74 V. Graw and C. Husmann

5.2.1.4 Ecological Dimension

We used data from the FGGD ‘Global land area with soil constraints’ dataset to
represent the ‘Ecosystems, natural resources, and climate’ sphere.4 The rural poor
are often particularly dependent on local natural resources and the land they live on
(Lee and Neves 2009). The chosen dataset includes information on soil depth, soil
chemistry, soil fertility, drainage, texture, and miscellaneous land-cover classes
(i.e., land that is unsuitable for agriculture, such as salt flats, deserts or glaciers) (van
Velthuizen et al. 2007). This information was derived from various datasets, including
several GIS layers on: soil, elevation, land-cover, climate variables, and remote
sensing images (e.g., for slope data). The ‘Global land area with soil constraints’
dataset includes a broad set of information on soils in 5-min grid-cell resolution.
Elevation data from the GTOPO 30—a global digital elevation model dataset of the
Earth Observation and Science (EROS) Data Center with a resolution of ca. 1 km
(30-arc-seconds)—also provided information on slopes. That dataset was compiled
using different raster sources and remote sensing images.5 Based on the FAO
classification we defined marginal soils as those in the categories ‘frequent severe
soil constraints’ and ‘very frequent severe soil constraints,’ as well as soils classified
as ‘unsuitable for agriculture’ (van Velthuizen et al. 2007).

5.2.1.5 Public Domain Dimension

For the sphere of ‘Public domain and institutions’ we took into account the
Worldwide Governance Indicators (WGI) developed by Kaufmann et al. (2010) for
the World Bank6 that defined governance as:
the traditions and institutions by which authority in a country is exercised. This includes (a) the
process by which governments are selected, monitored and replaced; (b) the capacity of the
government to effectively formulate and implement sound policies; and (c) the respect of citizens
and the state for the institutions that govern economic and social interactions among them.

Following this definition, the WGI were based on six indicators with two measures
of governance for each of the three areas: (a) ‘Voice and Accountability’ and ‘Political
Stability and the Absence of Violence/Terrorism,’ (b) ‘Government Effectiveness’
and ‘Regulatory Quality,’ and (c) ‘Rule of Law’ and ‘Control of Corruption.’ All
indicators were based on subjective or perception based measures of governance
gathered through surveys of households and firms, as well as expert assessments
produced by various organizations (Kaufmann et al. 2010). The indicators were
based on data from 1996 to 2010 and included 212 countries (in 2010), compiled

4
For more information visit: http://geonetwork3.fao.org/fggd
5
For more information visit: http://eros.usgs.gov/#/Find_Data/Products_and_Data_Available/
gtopo30_info
6
See also: http://info.worldbank.org/governance/wgi
5 Mapping Marginality Hotspots 75

from several hundred individual variables that measured perceptions of governance

according to 35 separate data sources conducted globally by 33 different organizations
(Kaufmann et al. 2009). We assessed the correlation between all six governance
indicators and found that they were all highly correlated.
Among the six indicators, ‘Political Stability’ was chosen to represent the
‘Public domain and institutions’ sphere. Political stability is one of the indicators
of the “perceptions of the likelihood that the government will be destabilized or
overthrown by unconstitutional or violent means, including political violence or
terrorism” (Thomas 2009, 5). According to Collier (2002), civil wars (severe politi-
cal instability) are one of the three main causes for a country remaining in a
‘developing’ state. Political stability or instability indicates the ability of govern-
ments to lead their respective populations. Political stability is also linked to
economic growth, in as much as instability is associated with reduced growth in
terms of per capita GDP, especially in low-income countries (Polachek and
Sevastianova 2010; The Economist 2011). Political stability is also an important
issue with regard to the socio-economic development of Africa, as a result of the
establishment of an institutional and legislative framework (Ong´ayo 2008). This
dimension was described by three quantiles, and we chose the cut-off point at the
lowest quantile.
Because we were interested in the number of people affected by the different
marginality dimensions we included population data, particularly data on the
number of poor by Harvest Choice based on the population database provided by
the Socioeconomic Data and Applications Center (SEDAC) and the Center for
International Earth Science Information Network (CIESIN) (CIESIN/IFPRI/World
2004). The Gridded Population of the World and the Global Rural–Urban Mapping
Project provided data with a resolution of 2.5-arc-minutes on population densities
based on a population layer that was compatible with the datasets on social, economic,
and earth science fields.

5.2.2 The Marginality Hotspots

Using classification and overlay techniques in ESRI ArcGIS, a ‘marginality hotspot

map’ was produced that reveals areas where three or more dimensions of marginality
overlapped (Fig. 5.1). In terms of marginality hotspots, we identified heavily affected
areas in SA (particularly India and Nepal) and SSA, especially Central and Eastern
Africa (Eritrea, Mozambique, Central African Republic, the Democratic Republic
of the Congo, Sudan, and large areas of Niger). Comparing the two regions, in SSA
the value of at least one dimension lies below the cut-off in nearly 70 % of the total
area compared to 20 % for SA (Fig. 5.1). Marginality hotspots included 27 % of the
area of SSA and 11 % of SA. In SA there were no areas with five overlapping
dimensions in contrast to SSA, where around 1 % of the total area is affected in this
way (Fig. 5.2).
 76

Fig. 5.1 Dimensions of marginality—where do negative values in different dimensions of marginality overlap?
V. Graw and C. Husmann
5 Mapping Marginality Hotspots 77

30
% of land area affected 25

20

15 1 Dimension
2 Dimensions
10
3 Dimensions
5 4 Dimensions
0 5 Dimensions
SSA (%) SA (%)
1 Dimension 16.56 1.38
2 Dimensions 25.09 8.26
3 Dimensions 18.98 9.51
4 Dimensions 7.48 1.05
5 Dimensions 0.73 0

Fig. 5.2 Comparison of areas affected by marginality dimensions in Sub-Saharan Africa and
South Asia (Based on data from map in Fig. 5.1)

5.3 Poverty and Marginality Hotspots—Where

Do They Overlap?

After identifying marginality hotspots based on the data describing the five dimensions
of marginality, two additional maps were generated to assess how these dimensions
overlap with data on poverty. To this end, marginality hotspots were overlaid with
sub-national poverty data provided by Harvest Choice of the proportions of popula-
tions, and the total number of people whose consumption levels were below
the poverty line of US$1.25/day. It is important to bear in mind that the poverty
data set is still under development and that data for some countries was missing, particularly
outside of SSA, making it difficult to draw final conclusions. Nevertheless a number
of areas can be identified that were marginal in several dimensions and that were
also greatly affected by poverty.
Figure 5.3 shows the overlay of the number of dimensions in which people were
marginalized and the percentage of the population living below US$1.25/day. We
identified marginality and poverty overlap in large areas of SSA as well as SA. Areas
where a high percentage of poor people coincided with marginality hotspots were
found in Central and South East Africa, especially the northern parts of Niger and in
Chad, the Central African Republic, the Democratic Republic of the Congo (espe-
cially the western part of the country), Mozambique, Malawi, and Burundi. In SA
marginality hotspots coincided with high poverty rates, particularly in Bangladesh
and Nepal. The case of Ethiopia highlights the difficulties of relying on official poverty
data. In Fig. 5.3 parts of the country appear in dark-blue color indicating marginality
hotspots, but official poverty rates for the country are astonishingly low (Ahmed et al.
2007) and no areas of extreme poverty and marginality hotspots overlapped.
Figure 5.4 shows the overlay of marginality hotspots with the number of people
living below US$1.25/day. The greatest number of marginalized poor in SA was
 78

Fig. 5.3 Overlay of the number of marginality dimensions with percentages of populations living below US$1.25/day
V. Graw and C. Husmann
 5
Mapping Marginality Hotspots

Fig. 5.4 Overlay of the number of marginality dimensions with the number of people living on less than US$1.25/day
79
80 V. Graw and C. Husmann

located in India and Bangladesh, and in SSA the most were in Ethiopia, southeastern
Africa, and some parts of western Africa. While poverty rates in India were generally
not as high as in other regions, large populations were affected, particularly in central
and western parts of the country.
Comparing the maps in Figs. 5.3 and 5.4 we clearly identified that the percentage
of poor and extremely poor was high in SSA, while in SA (particularly India and
Bangladesh), the sheer number of poor was extremely high. In contrast Africa’s
central regions no longer stand out as clearly on the map when the number of poor
people (rather than poverty rates) was taken into account. Nevertheless small areas
of the Democratic Republic of the Congo, Ethiopia, Uganda, Burundi, Mozambique,
Malawi, and the coastal areas of several West African countries appear to be
marginality hotspots where large numbers of poor people live. Figure 5.4 confirms
that in Asia the quantity of poor people was extremely high, whereas in Africa the
percentages of poor and extremely poor people were particularly high.

5.4 Limitations and Outlook

Marginality is a complex issue that often lies at the root of poverty. To geographically
reflect the different dimensions that influence marginality, we undertook this
mapping exercise in a first attempt to take into account the different ‘spheres of
life,’ instead of focusing on subsets of social, economic, or environmental aspects.
The most important limitation of this approach were the variable scales of the
different data sets. In the absence of comprehensive sub-national data, we used
GNI and political stability data at national-level scales. In contrast data on stunting
were provided on sub-national scales, and soil constraints on a pixel-level scale of
5-arc-minute grid cells. These different scales made comparisons difficult.
Therefore further work is needed to find more representative data sets at finer
levels. An additional advantage would be to refine the pixel resolution, as our map-
ping approach used a pixel size of 8 × 8 km, which is a small scale for the purpose of
global mapping, however, 64 km2 is a relatively large area in terms of the number of
people potentially living there and therefore interpretations needed to be made
carefully.
We also acknowledge that the determination of the cut-off points in each
dimension were arbitrary and therefore debatable. Further research is needed to
assess how different cut-off points influence marginality hotspot mapping. One
example was our definition of ‘distant’ using the accessibility dataset of Nelson
(2009) based on an infrastructure and cost-distance model. Most of the extremely
poor live in rural areas, particularly in remote areas (Sachs 2005). Even where
available public transport might be unaffordable to those in extreme poverty, the
distances in the map were based on the minimum time needed to get to the nearest
large settlement using vehicular transportation. The time expenditure is much higher
for people without access to vehicular transportation. Further research could apply
5 Mapping Marginality Hotspots 81

the accessibility methodology on a more local scale in order to measure walking

distances or alternative forms of travel to the nearest roads with regular transporta-
tion services.
The use of satellite images to identify marginal areas should also be explored
further. The advantage of remote sensing is relatively low cost compared to other
methods (except for very high resolution data). It is much more time consuming to
conduct a large survey than to process data (for example, land-use data from sev-
eral remote sensing images). Satellite imagery also provides pixel-level data,
allowing researchers to determine the geographic scope of their analyses at greater
detail than by relying on national or regional-level data that may generate mislead-
ing results.
It was also a challenge to map coupled human-environmental systems. For a more
precise observation of environmental aspects, data on climatic factors such as
precipitation or drought events should be integrated into analyses. More detailed
information on soils could help evaluate the potential for agricultural productivity
as well as land use. Finding good indicators of the status of complex socio-ecological
systems is therefore challenging. Several mapping approaches are working on this
issue, especially those undertaken by the FAO. Some difficulties arose from the lack
of data, such as rainfall data in developing countries, because analyses relied on
complete time series data information to get more insight in the rainfall variability
and its effects (Hughes 2006). Progress is being made on expanding rainfall data
collection stations in many remote areas, but in many cases these data are verified
only rarely, and therefore the data is not always accurate or consistent. As a result,
monitoring and the assessment of environmental variables are increasingly based on
remote sensing imagery, offering good tools for monitoring land surface changes,
whether human or natural. Nevertheless this data, even if very high resolution,
should also be ground-truthed to validate results.
The limitations of mapping approaches are manifold, especially when using data
on a global-scale. The lack of data in some areas also makes it difficult to make
comparisons between countries and regions. It is important to bear in mind that
the maps presented here were primarily meant to provide a general indication of the
distribution of particularly disadvantaged areas, which may be analyzed further
through more detailed data analysis at smaller scales.
A logical next step in mapping marginality would involve the optimization of indica-
tors for identifying marginality hotspots on a global scale, while developing more
reliable and representative approaches on the national level. In addition to identifying
hotspots at the global-level, further studies at national and sub-national levels are
necessary. Follow-up research will include scenario modeling to better understand
causal linkages between different indicators of the spheres of life and to identify pos-
sibilities for the design and application of poverty and marginality alleviation efforts.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.
82 V. Graw and C. Husmann

References

Ahmed AU, Hill RV, Smith LC, Wiesmann DM, Frankenberger T (2007) The world’s most
deprived: characteristics and causes of extreme poverty and hunger. 2020 discussion paper 43,
International Food Policy Research Institute, Washington, DC
CIESIN/IFPRI/World Bank/CIAT (2004) Global rural–urban mapping project, Version 1
(GRUMPv1): population density grid, Palisades. Center for International Earth Science
Information Network, Columbia University/International Food Policy Research Institute/
World Bank/Centro Internacional de Agricultura Tropical. Socioeconomic Data and
Applications Center, Columbia University, New York. http://sedac.ciesin.columbia.edu/gpw.
Accessed 15 Jan 2012
Collier P (2002) On the incidence of Civil War in Africa. J Confl Resolut 46(1):13–28
de Onis M, Blössner M, Borghi E (2011) Prevalence and trends of stunting among pre-school
children, 1990–2020. Public Health Nutr (Epub):1–7
Deichmann U (1999) Geographic aspects of inequality and poverty. World Bank, Washington, DC.
http://siteresources.worldbank.org/INTPGI/Resources/Pro-Poor-Growth/5319_povmap.pdf.
Accessed 15 Jan 2012
FAO (2007) Prevalence of stunting among children under five, by lowest available subnational
unit, varying years (FGGD). Raster dataset. Food and Agriculture Organization of the United
Nations, Rome. http://www.fao.org/geonetwork/srv/en/metadata.show?id=14055&currTab=si
mple. Accessed 12 Feb 2013
Gatzweiler F, Baumüller H, von Braun J, Ladenburger C (2011) Marginality: addressing the root
causes of extreme poverty. ZEF working paper no 77, Center for Development Research,
University of Bonn, Bonn
Hirotsugu U, Nelson A (2010) Agglomeration index, towards a new measure of urban concentration.
World Institute for Development Economics Research, United Nations University, Helsinki
Hughes D (2006) Comparison of satellite rainfall data with observations from gauging station
networks. J Hydrol 327(3–4):399–410
JRC (2010) Travel time to major cities. A global map of accessibility. Data source. Joint Research
Centre, European Commission, Brussels. http://bioval.jrc.ec.europa.eu/products/gam/sources.
htm. Accessed 12 Feb 2013
Kaufmann D, Kraay A, Mastruzzi M (2009) Governance matters VIII: aggregate and individual
governance indicators, 1996–2008. World Bank Policy Research working paper no 4978,
World Bank, Washington, DC
Kaufmann D, Kraay A, Mastruzzi M (2010) The worldwide governance indicators: methodology
and analytical issues. World Bank Policy Research working paper no 5430, World Bank,
Washington, DC
Kriz K (2010) Mapping different geographies. Springer, Heidelberg
Lee DR, Neves B (2009) Rural poverty and natural resources: improving access and sustainable
management. background paper for IFAD 2009 rural poverty report. http://www.ifad.org/rural/
rpr2008/chapter2/2.pdf. Accessed 12 Feb 2013
Lewit E, Kerrebrock N (1997) Population-based growth stunting. The future of children. Child
Poverty 7(2):149–156
Nelson A (2000) Accessibility, transport and travel time information. CIAT hillsides project report,
International Center for Tropical Agriculture (CIAT), Cali
Nelson A (2009) Travel time to major cities: a global map of accessibility. Joint Research Centre,
European Commission, Brussels. http://bioval.jrc.ec.europa.eu/products/gam/index.htm.
Accessed 12 Feb 2013
Ong´ayo AO (2008) Political instability in Africa. Where the problem lies and alternative perspectives.
African Diaspora Policy Centre, Amsterdam. http://www.diaspora-centre.org/DOCS/Political_
Instabil.pdf. Accessed 18 Jan 2012
Polachek SW, Sevastianova D (2010). Does conflict disrupt growth? Evidence of the relationship
between political instability and national economic performance. IZA discussion paper no
4762, Institute for the Study of Labor, Bonn
5 Mapping Marginality Hotspots 83

Sachs J (2005) The end of poverty: how we can make it happen in our lifetime. Penguin Press,
New York
Syrquin M (2011) GDP as a measure of economic welfare. ICER working paper no. 3/2011,
International Centre for Economic Research, Prague
The Economist (2011) The sun shines bright. The Economist, 3 December 2011, 401(8762):77–79.
http://www.economist.com/node/21541008. Accessed 12 Dec 2011
Thomas MA (2009) What do the worldwide governance indicators measure? Eur J Dev Res
22(1):31–54
Uchida H, Nelson A (2008) Agglomeration index: towards a new measure of urban concentration.
Background paper for the world development report 2009: reshaping economic geography,
World Bank, Washington, DC. http://siteresources.worldbank.org/INTWDR2009/Resources/
4231006-1204741572978/Hiro1.pdf. Accessed 12 Feb 2013
UNDP (2010) Human Development Report 2010: the real wealth of nations: pathways to human
development. United Nations Development Programme, New York
van Velthuizen H, Huddleston B, Fischer G, Salvatore M, Ataman E, Nachtergaele FO, Zanetti M,
Bloise M (2007) Mapping biophysical factors that influence agricultural production and rural
vulnerability. Food and Agriculture Organization of the United Nations, Rome
von Braun J, Hill RV, Pandya-Lorch R (2009) The poorest and hungry: a synthesis of analyses and
actions. In: von Braun J, Hill RV, Pandya-Lorch R (eds) The poorest and hungry: assessments,
analyses, and actions. International Food Policy Research Institute, Washington, DC, pp 1–61
World Bank (2009) Governance indicators. World Bank, Washington, DC. http://info.worldbank.
org/governance/wgi/index.asp. Accessed 12 Feb 2013
World Bank (2010) World development indicators. World Bank, Washington, DC. http://data.
worldbank.org/data-catalog/world-development-indicators. Accessed 12 Feb 2013
World Bank (2011) World Bank atlas method. World Bank, Washington, DC. http://data.worldbank.
org/about/country-classifications. Accessed 12 Feb 2013
Yohannes Y, Fritschel H, Olofinbiyi T, Nestorova B, Menon P, Ruel MT, von Grebmer K, von
Oppeln C, Towey O, Golden K (2010) 2010 Global Hunger Index: the challenge of hunger: focus
on the crisis of child undernutrition. International Food Policy Research Institute, Washington, DC
Chapter 6
The Poorest: Who and Where They Are?

Akhter U. Ahmed, Ruth Vargas Hill, and Farria Naeem

Abstract  This chapter provides a global quantitative perspective on where the world’s
poor and particularly the poorest live, and the extent of progress made in the reduction
of income poverty close to and far below the international poverty line (US$1.25/day
per capita) over the past two decades. The characteristics of the ultra-­poor are
identified based on a quantitative assessment across developing countries. Poverty
reduction from 1990 to 2008 was quite equal between those who are somewhat
below that poverty line and the ultra-poor—in fact it slightly favored the ultra-­poor.
This suggests that the theory of enduring poverty traps may not be holding true for
those in ultra-poverty in recent years, as this pattern is different compared to earlier
global findings. For interventions to reach the ultra-poor (i.e., those living on less
than US$0.63/day) effectively, geographically marginal households should be
targeted; and the low-levels of education, and in the case of Asia, landlessness should
be taken into account.

Keywords  Poverty • Ultra-poverty • Poverty traps • Targeting • Exclusion

6.1  Where Do the Poor Live?

Although poor people live in almost every society they are often concentrated
geographically in certain areas. The severity of poverty varies across countries
and the geographic patterns of concentration are not static. Moreover while the
poor share certain common characteristics, the profile of the poor often differs

A.U. Ahmed (*) • F. Naeem

International Food Policy Research Institute (IFPRI), Dhaka, Bangladesh
e-mail: [email protected]
R.V. Hill
International Food Policy Research Institute (IFPRI), Washington, DC, USA

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 85
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_6,
© The Author(s) 2014
86 A.U. Ahmed et al.

among societies. Knowledge of where the poor live, how severe their poverty is, and
other characteristics can be critical for designing effective policies to meet their
needs and improve their welfare.
In this chapter we have provided a general review of where the world’s poor and
particularly the poorest live, the extent of progress made in poverty reduction, and
the characteristics of the poor based on a quantitative assessment across selected
developing countries.1 Our research shows that despite much progress reducing
poverty worldwide, a substantial number of the world’s poorest people are being left
behind. Millions of people across the developing world are still solely preoccupied
with survival and hunger is a reality in their everyday lives. Today around 1.3 billion
people in the developing world subsist on less than US$1.25/day and 234 million
live on less than US$0.63/day. Our findings also indicate that compared to the poor
with incomes closer to the poverty line of US$1.25/day, the poorest typically belong
to socially excluded groups, lived in remote rural areas, and had less education,
fewer assets, and less access to markets.
In September 2000 the Millennium Declaration was adopted by 189 member
states of the United Nations. The heads of states confirmed their countries’ com-
mitments to achieving the eight Millennium Development Goals (MDGs) by 2015.
The first MDG is to reduce the proportion of people that lived in poverty and hunger
in 1990 by half by 2015. The MDG indicator of extreme poverty is living on less
than US$1.25/day,2 which we used in this chapter to identify the regions where the
world’s poor live and trends in poverty from 1990 (the base year for the MDGs)
until 2008. We disaggregated this measure of poverty to look at the locations and
changes in the welfare of those living below US$0.63/day and by doing this we
captured trends in the severity of extreme poverty.
In 1990 the developing world had a population of 4.43 billion, of which 1.91 billion
(43.1 %) lived on less than US$1.25/day. From 1990 to 2008 the number of people
in developing countries grew by 1.32 billion and the number of people living on
less than US$1.25/day declined. Of the developing world’s 5.75 billion people in
2008, 1.29 billion (22.4 %) lived on less than US$1.25/day (Fig. 6.1). The regional
composition of the developing world’s poor changed remarkably over the
18-year period (Fig. 6.1).3 East Asia and the Pacific’s share of the world’s poor
declined by more than half to only 22 % during that period. In contrast South Asia’s

1 
This chapter was largely drawn from Ahmed et al. (2007) along with analyses of an updated
dataset.
2 
The World Bank defines the extreme poor of the world as those who live on less than US$1.25/day,
measured at the 2005 purchasing power parity (PPP) exchange rates. The measures of poverty
used in this chapter come from PovcalNet (World Bank 2011). PovcalNet is an interactive
computational tool that has been developed by a team at the World Bank Research Group that
allows users to calculate poverty measures for different poverty lines and country groupings based
on household survey data.
The developing world was divided into six regions: East Asia and the Pacific, South Asia, Sub-Saharan
3 

Africa, Latin America and the Caribbean, Eastern Europe and Central Asia, and the Middle East
and North Africa (in decreasing order of total population). We have removed China from the
East Asia and Pacific region to observe China’s performance separately.
6  The Poorest: Who and Where They Are? 87

Fig. 6.1  Where the poor lived: 1990 and 2008 (Ahmed et al. 2007, updated 2012)

share of the extreme poor increased from 32 to 44 % and Sub-Saharan Africa’s
share doubled over the same 18-year period.
The difference in trends of the quantity of extreme poor living in East Asia and
the Pacific versus South Asia and Sub-Saharan Africa is remarkable. While East
Asia and the Pacific had a larger number of poor (926 million) than South Asia
(617 million) in 1990, East Asia and the Pacific had 287 million fewer people in
extreme poverty than South Asia in 2008. Indeed East Asia and the Pacific is the
only region that experienced a substantial decline in the numbers of those living on
less than US$1.25/day (from 926 million to 284 million) between 1990 and 2008.
South Asia saw a modest decline of 46 million and the number of extreme poor
actually increased in Sub-Saharan Africa by about 96 million (from 290 million in
1990 to 386 million in 2008). The total number of extreme poor in Sub-Saharan
Africa became larger than in East Asia and the Pacific during this period.
The proportion of the developing world’s population living on less than US$1.25/
day fell from 43.1 % in 1990, the base year for the MDGs, to 22.4 % in 2008, sug-
gesting that the poverty component of the first MDG has virtually been met at the
global-level 7 years before the target date of 2015. Regional progress, however, has
been uneven. The decline in the global proportion of people living in extreme poverty
has been largely driven by East Asia and the Pacific, and to some extent by South Asia.
Indeed East Asia and the Pacific overachieved the poverty MDG—extreme poverty
(below US$1.25/day) in the region fell by 42 % points from 56.2 % in 1990 to 14.3 %
in 2008. Extreme poverty also declined substantially in South Asia from 53.8 to
36.0 % over 1990–2008. In Latin America and the Caribbean the decline was from
12.2 to 6.5 %. Although Sub-Saharan Africa experienced some decline in extreme
poverty from 1990 to 2008, the decline was very limited (from 56.5 to 47.5 %).
88 A.U. Ahmed et al.

6.2  L
 ooking Beneath the US$1.25/day Line:
Subjacent and Ultra-poverty

While the MDGs classify extreme poverty as living on less than US$1.25/day, we
disaggregated those living on less than that amount into two groups. We chose
cut-­off points that split the distribution into two meaningfully sized groups while at
the same time using simple, equally spaced cut-off points. We designated the two
groups as those living in ‘subjacent’ and ‘ultra-poverty’ as follows4:
• Subjacent poverty were those living on US$0.63–1.25/day
• Ultra-poverty were those living on less than US$0.63/day
Out of 1,289 million poor living below US$1.25/day in 2008, 1,055 million
(82 %) were subjacent poor and 234 million (18 %) were ultra-poor. Figure 6.2
shows that most of the world’s ultra-poor lived in Sub-Saharan Africa. Most of
Asia’s extreme poor lived just below the US$1.25/day line—only a small minority
there was ultra-poor.
Figure 6.3 shows the trends in ultra-poverty proportions in three major regions
from 1990 to 2008. Again we see that the declines were much more rapid in East
Asia and the Pacific than elsewhere. The proportional declines in ultra-poverty from

2008 Ultra poor: 234 million

Europe and
Central Asia
East Asia and 0%
Pacific
Latin America
7%
and the
Caribbean
8%

Middle East and

North Africa
0%

South Asia
15%

Sub-Saharan
Africa
Fig. 6.2  Where those 70%
in subjacent and ultra-poverty
lived: 2008 (Ahmed et al.
2007, updated 2012)

Although the poverty gap ratio and the distribution-sensitive squared poverty gap ratio could be
4 

used to measure the depth and the severity of poverty respectively, we used the head-count measure
of poverty because its interpretation is straightforward.
6  The Poorest: Who and Where They Are? 89

Percentages of regional populations living on less than US$ 0.63/day

30
Sub-Saharan Africa
27

24

21

18
Percent

East Asia and

15
Pacific
12

6 South Asia

0
1990 1993 1996 1999 2002 2005 2008

Fig. 6.3  Trends in ultra-poverty proportions of regional populations, 1990–2008 (Ahmed et al.
2007, updated 2012)

1990 to 2008 by region were as follows: East Asia and the Pacific from 13.4 to
0.9 %; South Asia from 8.0 to 2.2 %; and Sub-Saharan Africa from 24.5 to 20.0 %.
The East Asia and the Pacific region experienced substantial economic growth
during this period and also had initial conditions such that the growth benefited
many people living in poverty (i.e., growth elasticity of poverty reduction was very
high). South Asia also achieved remarkable economic growth rates during the
period, but was less able to convert this growth to reductions in poverty. Sub-Saharan
Africa saw both limited economic growth and poverty reduction during this period.
Given population growth throughout this period, Sub-Saharan Africa has seen
increases in the number of people living in ultra-poverty even though the proportion
of the population living in ultra-poverty has declined. Sub-Saharan Africa’s high
proportion of ultra-poverty and limited growth and progress in reducing poverty
calls into question the idea that business-as-usual will create improvements in
well-­being in a timely manner for a large share of the world’s absolute poorest.
Indeed the continued prevalence and severity of poverty in Sub-Saharan Africa is
one of the major contemporary ethical challenges to the global society.5

5 
Poverty traps have been identified in Madagascar (Barrett et al 2006), Kenya (Barrett et al.
2006), South Africa (Adato et al. 2006), and Cote d’Ivoire (Barrett et al. 2001), but have not been
identified in Russia (Loshkin and Ravallion 2004), China (Jalan and Ravallion 2001), and Mexico
(Antman and McKenzie 2005), although the authors found considerable persistence of poverty
over time in Mexico.
90 A.U. Ahmed et al.

Reductions in poverty, 1990–2008

US$ 1.25/ Subjacent Ultra
day poverty poverty poverty
0%

-5%
-5%
Percentage point change

-6%

-10%

% point decline in
poverty:
-15% -14% equitable growth vs.
-16% actual

-20% Equitable growth

-21%
Actual decline
-25%

Fig. 6.4  Percentage point changes in poverty from changes in subjacent and ultra-poverty: 1990–
2008 (Ahmed et al. 2007, updated 2012)

Under theoretical standard growth models the convergence hypothesis implies

that gains should come quickest to those living in ultra-poverty. Theories of poverty
traps link severe poverty with slow improvements in welfare. The severity of poverty
and the limited progress in reducing it indicate that the poorest in Sub-Saharan
Africa may be trapped in poverty as some recent literature suggests (Azariadis and
Stachurski 2005; Sachs 2005; Collier 2007). Micro-level evidence of poverty traps
has been found for a number of countries in Sub-Saharan Africa, while little evidence
has been found for countries in other regions of the world where the severity of
poverty is lower (Russia, China, Mexico).
If poverty traps exist, those in ultra-poverty may be so poor that optimal behavioral
choices cause them to rise out of poverty much more slowly than those who are less
poor. If this is the case, then gains may come more quickly to those living just under
US$1.25/day in subjacent poverty and the people furthest from the extreme poverty
line—those in ultra-poverty—will experience less poverty reduction. How can we tell
whether those in ultra-poverty have fared better or worse than those closer to the line?
While panel data is needed to decisively answer this question, we calculated
the amount that subjacent and ultra-poverty would have been reduced (or increased in
some cases) if poverty reduction had come from across-the-board income growth by
the same amount to get an indication of this from national poverty data. We com-
pared this ‘equal growth scenario’ poverty reduction with the amount of poverty
reduction that actually took place. The ‘equal growth scenario’ poverty reduction is
shown as a white bar next to the actual change in poverty proportions in Fig. 6.4
6  The Poorest: Who and Where They Are? 91

(the appendix describes how this was calculated). For example, if the 21 percentage
point decrease in global poverty had come from universal income growing by the
same amount, there would have been a decline in subjacent poverty of 16 percentage
points and a fall in ultra-poverty of five percentage points.
We found that the incidence of poverty among those just below the US$1.25/
day poverty-line fell two percentage points less than it would have had all incomes
grown equally, whereas the incidence of ultra-poverty fell one percentage point
more than it would have had all incomes grown equally. This finding suggests that
poverty reduction was quite equal between the subjacent poor and the ultra-poor—in
fact it marginally favored the ultra-poor. It is encouraging to see that the theory
of poverty traps may not be holding true for those in ultra-poverty. This is in
contrast to the trend that was observed from 1990 to 2004, when ultra-poverty fell
much more slowly than subjacent poverty (Ahmed et al. 2007), which was more
consistent with theoretical expectations of the existence of poverty traps. Part of
the reason for this change is that we are using the 2005 purchasing power parity
(PPP) exchange rate rather than the 1993 PPP rate upon which the previous results
were based.

6.3  Ranking Countries by Concentration of Poor

The distribution of the people living below US$1.25/day in the developing world
shows that poverty concentration by country was quite similar in 1990 and 2008.
The global distribution of poor remained highly concentrated, with ten countries
accounting for 78 % in 2008 (Fig. 6.5) compared to 85 % in 1990 (Fig. 6.6).

40
% of all poor in the developing world

36

30

21
20
14
10 8
5 4 4 3 2 1 1
0

Fig. 6.5  Top ten countries in terms of the percentage of people living below US$1.25/day in 2008
(Ahmed et al. 2007, updated 2012)
 92 A.U. Ahmed et al.

40
% of all poor in the developing world
37

30
25

20
15

10
5
4 4 3 3 2 1 1
0

Fig. 6.6  Top ten countries in the number of people living below US$1.25/day in 1990 (Ahmed
et al. 2007, updated 2012)

However, a distinct pattern in country-wise poverty reduction emerged when we

looked into changes in the numbers of poor over time.
In 2008, 14 % of the global poor lived in China, compared to 37 % in 1990. This
high number of poor in China is due in part to sheer population size. China has had
remarkable success in extreme poverty reduction—from 680 million in 1990 to only
173 million in 2008.
A comparison of Figs. 6.5 and 6.6 indicates that Brazil, Ethiopia, and Vietnam
were among the ten countries with the highest proportions of poor in 1990; but
were replaced by Kenya, the Philippines, and Tanzania in 2008. Between 1990
and 2008 there was an upward movement in the distribution of global poor in
some Sub-­Saharan Africa countries such as Nigeria and the Democratic Republic
of Congo. The ranking of countries with the highest numbers of people living in
poverty also changed; in 1990 the top three countries were China, India, and
Indonesia, but in 2008 they were India, China, and Nigeria.
Further disaggregation of poverty data within countries showed that global
extreme poverty was highly concentrated in rural areas of India and China. In 2008
40 % of the total developing world’s poor lived in rural areas of India and China—
27 % in rural India and 13 % in rural China. A recent World Bank study (2009)
found a widening rural–urban poverty gap in China, particularly for the western
provinces where the incidence and severity of poverty is highest. Similarly the
divergence in rural–urban poverty rates in India has been an impediment to its overall
poverty reduction (Fig. 6.7).
Ultra-poverty was also highly concentrated in a few countries. Ten countries
accounted for 75 % of the world’s total ultra-poor in 2008. Nigeria alone accounted
for 23 % of the world’s ultra-poor. Nigeria’s ultra-poor population (50 million)
accounted for almost half of Nigeria’s total poor in 2008. Similarly for the
6  The Poorest: Who and Where They Are? 93

% of all poor in the

30 27

developing world
20
13
10

0
India-Rural China-Rural

Fig. 6.7  Percentage of the world’s extreme poor (below US$1.25/day) living in rural areas of
China and India in 2008 (Ahmed et al. 2007, updated 2012)

30
% of all ultra poor in the developing world

25
23
20
16

12
10
6 5
3 3 3 2 2

Fig. 6.8  The percentages of the global poor population represented by the top ten countries
in terms of the number of ultra-poor living below US$0.63-a-day in 2008 (Ahmed et al. 2007,
updated 2012)

Democratic Republic of Congo, its share of the global ultra-poor was 16 %, which
accounts for more than half of the country’s total poor.
On the other hand, although India accounted for 12 % of the world’s total
ultra-­poor, the absolute number of ultra-poor was around 28 million, which was
only 2.4 % of the total population of the country. Similarly for China, the share of
ultra-­poor was 6 % of global-ultra poor, but the absolute number of ultra-poor
(13 million) was only 1 % of China’s total population in 2008. The small fraction of
ultra-poor with respect to the total populations of China and India indicates that
despite the concerted poverty reduction efforts made by these nations the ultra-poor
have been hard to reach. A World Bank (2009) poverty assessment of China notes
that it has become harder to eliminate the remaining poverty in the country because
the extremely poor are highly dispersed. Addressing this problem would require
shifting from area-based poverty reduction approaches to more innovative and
streamlined household-based targeting approaches (Fig. 6.8).
94 A.U. Ahmed et al.

6.4  Who Are the Poor?

Understanding the characteristics of the world’s poorest is important for designing

effective policies to meet their needs and improve their welfare. In this section we
contribute to this understanding by analyzing household data and reviewing the
results of empirical research in 20 countries: Burundi, Ethiopia, Ghana, Kenya,
Malawi, Mozambique, Rwanda, Senegal, Zambia, Bangladesh, India, Pakistan,
Sri Lanka, Laos, East Timor, Vietnam, Tajikistan, Peru, Guatemala, and Nicaragua.
The characteristics considered here were limited to those that can be compared
across countries, at least to the extent possible.

6.4.1  Spending on Food, Fuel, Housing, and Health Care

Across income groups and regions, expenditures on food represented the highest
share of household budgets. In general poorer households and those in rural
areas spent a relatively higher proportion of the family budget on food than others,
but the differences were not large. Expenditures on fuel represented the second
highest expenditure share in Bangladesh, India, and Pakistan, while housing costs
represented the second highest share in all three sample countries in Latin America
and in Tajikistan. No clear pattern between health care expenditures and poverty
emerged across these countries, which is a worrisome finding because poverty
assessments for these countries have repeatedly found that ill health is associated
with poverty (Ahmed et al. 2009).

6.4.1.1  Remoteness

The poorest and most food-insecure households were located furthest from roads,
markets, schools, and health services. In addition to being an indicator of wealth, an
electricity connection also indicates, to a certain extent, the ‘connectedness’ of
households to roads, markets, and communications infrastructure, and the resulting
income-earning opportunities and public services. Consistently across countries,
extremely poor households had considerably lower access to electricity than those
living above the poverty line (Ahmed et al. 2009).

6.4.1.2  Education

Education has been shown to have significant positive impacts on agricultural

productivity, employment, access to credit, use of government services, personal and
children’s health, and education outcomes. In nearly all study countries the proportion
of uneducated adult males was almost double or more among the ultra-poor than for
people living above the poverty line. Quality primary education can provide children
6  The Poorest: Who and Where They Are? 95

from poor families with the tools to move out of poverty. In all study countries,
however, the evidence was the same: children from poorer families were less likely to
go to school. Without education the future of children living in extreme poverty will
likely be a distressing perpetuation of their current experience (Ahmed et al. 2009).

6.4.1.3  Landlessness in Rural Areas

The ownership or control of productive assets is an important indicator of livelihood

quality because such assets generate income. In all parts of Asia the landless were
highly represented among the poorest. For example, nearly 80 % of the ultra-poor in
rural Bangladesh did not own land. In Sub-Saharan Africa, however, little difference
was found between the incidence of landlessness among the poorer and less poor
households, and in some cases the reverse was true. This corresponds to findings of
other studies that in Sub-Saharan Africa the poorest often own some land (but too
little), but lack access to other key assets and markets (Ahmed et al. 2009).

6.4.1.4  Excluded Groups

In each of the 20 countries considered in this study some social groups—but not the
societal majorities—had consistently higher prevalence of poverty and hunger.
Individuals in groups excluded from regional progress against poverty remained
among the poorest in Asia. For example, in Laos the prevalence of poverty was
more than twice as high among the minority Mon-Khmer than the majority Lao,
and in Vietnam the incidence of extreme poverty was more than six times higher
among ethnic minorities that among the Kinh and Chinese ethnicities. In India
disadvantaged castes and tribes (referred to as Scheduled Castes and Scheduled
Tribes) were overrepresented among the ranks of the extremely poor, particularly
among those living in ultra-poverty. This was more commonly the case for Scheduled
Tribes than for Scheduled Castes (Ahmed et al. 2009).
In Sub-Saharan Africa access to land and other resources depends on member-
ship in groups of common descent, which results in outsiders having difficulty
accessing resources and securing stable livelihoods. This is true in Senegal
where refugees from Mauritania and displaced people from the Casamance were
more likely to remain in poverty than native Senegalese. The genocide in Rwanda
also displayed the importance of ethnicity in determining access to resources
(Ahmed et al. 2009).

6.4.1.5  Women

Some weak evidence supports the hypothesis that female-headed households are
overrepresented among the ultra-poor, but in general, significant differences were
not found. Examining only the differences between male-headed and female-headed
96 A.U. Ahmed et al.

households obscures the reality that, within households headed by men, the welfare
of women and girls may be lower than that of their male counterparts. Some
studies in South Asia have shown that within households, women consume sig-
nificantly less food and sometimes less high-quality food such as meat and eggs
(Ahmed et al. 2009).

6.5  Conclusion

We have sought to answer two crucial questions—‘who are the world’s poorest?’
and ‘where are they concentrated?’ We approached these questions by identifying
the regions of the world in which deprivation was most severe and noting where
progress has been achieved, and also by examining the characteristics of the
extremely poor. The overall goal of this research was to strengthen the empirical
basis upon which policymakers can make informed policy choices for reducing
extreme poverty. The proportion of the developing world’s population in extreme
poverty, defined here as those living on less than US$1.25/day, fell from 43.1 % in
1990 (the base year for the MDGs), to 22.4 % in 2008. This suggests that the poverty
component of the first MDG had virtually been met at the global-level 7 years
before the target date of 2015.
Regional progress, however, was uneven. The decline in global extreme
poverty has been largely driven by East Asia and the Pacific, aided by South Asia.
Indeed East Asia and the Pacific overachieved the poverty MDG, as extreme
poverty in the region fell by 42 percentage points from 56.2 % in 1990 to 14.3 %
in 2008. Extreme poverty also fell substantially in South Asia, from 53.8 to
36.0 % over 1990–2008. In Latin America and the Caribbean the decline
was from 12.2 to 6.5 %. Although Sub-Saharan Africa experienced some decline
in extreme poverty from 1990 to 2008, the decline has been very limited (from
56.5 to 47.5 %).
Disaggregating those living on less than US$1.25/day into two groups allowed
us to consider changes in the severity of extreme poverty: Those in subjacent
poverty (between US$0.63 and US$1.25/day) and those in ultra-poverty (less than
US$0.63/day). Using this disaggregation we have shown that 234 million of the
world’s 1.3 billion extremely poor people lived in ultra-poverty in 2008. Most of
the world’s ultra-poor lived in Sub-Saharan Africa. Most of Asia’s poor lived in the
subjacent category—only a small minority there was ultra-poor.
Our findings suggest that poverty reduction from 1990 to 2008 was quite equal
between the subjacent poor and the ultra-poor—in fact it marginally favored the
ultra-poor. It is encouraging to see that the theory of enduring poverty traps may
not be holding true for those in ultra-poverty. This is in contrast to the trend that
was observed from 1990 to 2004, when ultra-poverty fell much more slowly than
subjacent poverty, which was more consistent with theoretical expectations regarding
the existence of poverty traps.
6  The Poorest: Who and Where They Are? 97

The distribution of people living below US$1.25/day in the developing world

shows that extreme poverty was highly concentrated in a few countries—ten
countries accounted for 78 % of global poor in 2008. Ultra-poverty was also highly
concentrated—ten countries accounted for 75 % of the world’s total ultra-poor in
2008. In 2008 Nigeria accounted for 23 % of the world’s ultra-poor.
Understanding who the poorest are is crucial for the design of effective inter-
ventions to improve their welfare. Without context-specific and timely information,
it is difficult to design programs that fit their needs. It is thus important to broaden
the collection of and access to accurate data on the world’s poorest.
The evidence presented in this chapter suggests that for interventions to
reach those living on less than US$0.63/day effectively, remote households that
are traditionally excluded from resources and markets should be targeted, and
both the low-­levels of education and in the case of Asia, landlessness should be
taken into account. Our findings suggest that interventions to protect the poor
against health shocks, to address the exclusion of social groups, to prevent child
malnutrition, and to enable investments—particularly in education—are essential
for helping them rise out of poverty.

 ppendix: Subjacent and Ultra-poverty

A
Decomposition Analysis

To calculate the change in extreme poverty that would have resulted from equal
growth in all incomes it was assumed that incomes within countries and regions
were log-normal distributed (a common assumption in the inequality literature, see
Bourguignon 2003 and Klasen and Misselhorn 2006), such that the distribution of
income in 1990 was log-normal distributed with a mean, y and a standard devia-
tion s . Bourguignon (2003) shows that in this case the poverty rate Pt can be cal-
culated from only the mean and standard deviation of income by:

 log z − log y  s 
Pt ( z ) = Π  +  (6.1)
 s  2

Where z is the poverty line and Π is the cumulative normal distribution. When
incomes are log normally distributed the standard deviation can be calculated from
the Gini coefficient of income (G) by:

 G +1
s = 2 Π −1   (6.2)
 2 

Where Π -1 is the inverse of the cumulative normal distribution.

Using Eq. (6.1) it is possible to determine the growth in income commensurate
with the observed change in dollar/day headcount poverty between 1990 and 2004
98 A.U. Ahmed et al.

by calculating the estimated level of mean income y2004 that would give the observed
headcount poverty estimate of P2004 (1.08) if the standard deviation of the distribu-
tion ( s 1990 ) had stayed the same:

 s 
log y2004 = log (1.08) − Π −1 [ p2004 (1.08)] − 1990  s 1990
 2 

Using this estimate of y2004 , headcount poverty estimates P2004 (0.81) and
P2004 (0.54) can be determined and from this expected change in subjacent and
ultra-poverty can be generated.
Country Gini indices were taken from World Bank (2011). For regions the standard
deviation of the regional distribution of income was taken directly from Besley and
Burgess (2003). Milanovic’s (2002) estimates of the distribution of world income
from household survey data were used to estimate the Gini of the developing world.

Open Access  This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Adato M, Carter M, May J (2006) Exploring poverty traps and social exclusion in South Africa
using qualitative and quantitative data. J Dev Stud 42(2):226–247
Ahmed AU, Hill RV, Smith LC, Wiesmann DM, Frankenberger T (2007) The world’s most
deprived: characteristics and causes of extreme poverty and hunger. 2020 discussion paper 43,
International Food Policy Research Institute, Washington, DC
Ahmed AU, Hill RV, Smith L, Frankenberger T (2009) The poorest and hungry: characteristics and
causes. In: von Braun J, Hill RV, Pandya-Lorch R (eds) The poorest and hungry: assessments,
analyses, and actions. International Food Policy Research Institute, Washington, DC, pp 107–116
Antma, F, McKenzie D (2005) Poverty traps and nonlinear income dynamics with measurement
error and individual heterogeneity. World Bank Policy Research working paper 3764, World
Bank, Washington, DC
Azariadis C, Stachurski J (2005) Poverty traps. In: Aghion P, Durlauf S (eds) Handbook of
economic growth. North-Holland, Elsevier, Amsterdam/San Diego, pp 295–384
Barrett C, Bezuneh M, Aboud A (2001) Income diversification, poverty traps and policy shocks in
Cote d’Ivoire and Kenya. Food Policy 26(4):367–384
Barrett C, Marenya P, Mcpeak J, Minten B, Murithi F, Oluoch-Kosura W, Place F, Randrianarisoa
JC, Rasambainarivo J, Wangila J (2006) Welfare dynamics in rural Kenya and Madagascar.
J Dev Stud 42(2):248–277
Besley T, Burgess R (2003) Halving global poverty. J Econ Perspect 17(3):3–22
Bourguignon F (2003) The growth elasticity of poverty reduction: explaining heterogeneity across
countries and time periods. DELTA working paper 2002–03, Department and Laboratory of
Applied and Theoretical Economics, Paris
Collier P (2007) The bottom billion: why the poorest countries are failing and what can be done
about it. Oxford University Press, Oxford
Jalan J, Ravallion M (2001) Household income dynamics in rural China. Policy Research working
paper series no 2706, World Bank, Washington, DC
6  The Poorest: Who and Where They Are? 99

Klasen S, Misselhorn M (2006) Determinants of growth semi-elasticity of poverty reduction.

Mimeo, University of Göttingen, Göttingen
Lokshin M, Ravallion M (2004) Household income dynamics in two transition economies. Stud
Nonlinear Dyn Econometr 8(3):374–389
Milanovic B (2002) True world income distribution, 1988 and 1993: first calculations based on
household surveys alone. Econ J 122:51–92
Sachs J (2005) The end of poverty: economic possibilities for our time. Penguin Press, New York
World Bank (2011) PovcalNet: an online poverty assessment tool. World Bank, Washington, DC.
http://iresearch.worldbank.org/PovcalNet/index.htm. Accessed 6 Jun 2012
World Bank (2009) From poor areas to poor people: China’s evolving poverty reduction agenda.
An assessment of poverty and inequality in China. Poverty Reduction and Economic
Management Department/World Bank, Washington, DC
Chapter 7
Targeting the Poorest and Most Vulnerable:
Examples from Bangladesh

Nusha Yamina Choudhury and Christa Räder

Abstract Programs to alleviate poverty and hunger are most effective when they
reach the poorest and most vulnerable, however, identifying target groups is often
difficult. This chapter focuses on geographic and beneficiary targeting mecha-
nisms for safety-nets, along with the strengths and limitations of each approach.
Different modalities of beneficiary targeting of the poorest and most vulnerable
groups were examined with examples from the United Nations World Food
Programme safety-net initiatives in Bangladesh. The chapter highlights the impor-
tance of strengthening targeting mechanisms in government operated safety-net
programs and improving relevant governmental capacity.

Keywords Bangladesh • Social safety nets • Intervention targeting • GIS

7.1 Introduction: Why Targeting?

The identification of the poorest and most vulnerable is primarily linked to how we
define and characterize them. In Bangladesh most definitions of poverty incorporate
income, expenditures, and food consumption dimensions. The Household Income
and Expenditure Survey (HIES) conducted by the Bangladesh Bureau of Statistics
(BBS) defines poverty by the Cost of Basic Needs (CBN) and Direct Calorie Intake
(DCI) methods. The CBN defines poverty based on per capita expenditures on basic
food and nonfood items, and differentiates between the poor and extremely poor

N.Y. Choudhury (*) • C. Räder

World Food Programme, Dhaka, Bangladesh
e-mail: [email protected]; [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 101
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_7,
© The Author(s) 2014
102 N.Y. Choudhury and C. Räder

using upper and lower poverty lines.1 The DCI defines poverty and extreme poverty
based on daily per capita food consumption below 2,122 and 1,805 kcal respec-
tively. In the measurement of poverty, diet is an important element, hence both
methods have a certain degree of overlap.
Over the past decade Bangladesh has achieved significant reductions in poverty as
measured by the CBN and DCI methods. In this paper we focus on the CBN method
of poverty measurement because it is widely preferred over the DCI method (BBS
2003). Based on the CBN method, the prevalence of poverty was as high as 57 % at
the beginning of the 1990s, declined to 50 % in 2000 and to 40 % in 2005. Subsequent
estimates from BBS in 2010 revealed that poverty had declined even further to 32 %.
The prevalence of extreme poverty was also reduced from 34 % in 2000 to 25 % in 2005.
Despite reductions in the prevalence of poverty the absolute number of poor and
marginalized remains high due to population size. Out of a population of 128 million
in 2000, 64 million were poor and 44 million were extremely poor. By 2005 the
population of Bangladesh had grown to 140 million population, of which the poor
and extremely poor accounted for 56 and 35 million respectively.
In addition to the large number of poor, geographical variation in per capita poverty
measurement and the severity of poverty are also concerns. The eastern regions of
Bangladesh have increasingly benefited from integration with major urban centers,
namely Dhaka and Chittagong, in contrast to the more isolated west and southwest,
which are economically lagging and where poverty levels are higher. This disparity
raises questions about the values of a regional versus a nationwide focus of poverty
reduction initiatives.
In poverty reduction efforts the targeting choice between poor versus extremely poor
presents another challenge to the design of programs and their implementation strate-
gies. In the recent past most of Bangladesh’s national policies, as well as mainstream
development programs, addressed poverty, whereas extreme poverty received more
focus from international development partners. In Bangladesh extreme poverty and
marginality not only denote a lack of income, but are also strongly linked to increased
exposure to risk of natural and economic shocks, limited or no access to social and eco-
nomic entitlements, and exclusion from mainstream development activities.
While the national government of Bangladesh and its development partners have
prioritized poverty on their agenda, limited availability of resources to address the
40–60 million people living in poverty remains a daunting challenge. In poverty and
vulnerability reduction efforts the effective utilization of limited resources is a key
challenge for development practitioners, especially when choices have to be made and
criteria for prioritization need to be defined. This challenge leads to three major ques-
tions: Who are the most vulnerable? Why are they vulnerable? Where do they live?
It is here where targeting and the prioritization of areas and beneficiaries become
necessary for the implementation of effective and efficient safety net programs.

1
The upper poverty line corresponds to households in absolute poverty whose food consumption
expenditures are at or below the food poverty line (based on the cost of a basic diet that provides
2,122 kcal/person a day). The lower poverty line corresponds to households in extreme poverty
whose total expenditures are equal to the food poverty line.
7 Targeting the Poorest and Most Vulnerable: Examples from Bangladesh 103

7.2 Geographic Targeting

When resources are limited, or issues and problems have regional dimensions,
geographic targeting is the most efficient way to allocate resources. With geographic
targeting, eligibility for benefits is first and foremost determined by location. Poverty
maps can be used to focus programs in areas of the country or to allocate resources
by subnational jurisdictions.
In Bangladesh the United Nations World Food Programme (WFP) initiated
the spatial targeting of programs during the 1980s. In the beginning resource
allocation maps were produced for the Vulnerable Group Development (VGD)
program. These maps were partly based on perceptions of poverty and partly on
indicators associated with poverty. Later with the availability of poverty maps in
2004 the targeting of programs became more robust by incorporating data from
surveys and censuses.

7.3 Mapping of the Poorest and Most Vulnerable

Programs to alleviate poverty and hunger are most effective when they provide
assistance in areas where the poorest reside and where socio-economic development
lags behind. In the past the ability to identify lagging areas was limited. Geographic
targeting became a popular approach for resource allocation and decision making
with the development of appropriate analytical and statistical methods, and the
availability of information on poverty and hunger indicators.
Poverty mapping is an exercise to estimate poverty prevalence at a level where
a typical household income and expenditure survey cannot produce statistically
reliable poverty estimates due to high sampling errors. In Bangladesh official
poverty rates are not estimated below the division level.2 Various poverty mapping
methods were devised to disaggregate the higher administrative level poverty
data to a lower administrative or geographic level, and to overcome the increasing
imprecision of poverty estimates arising from the disaggregation. The Small Area
Estimation (SAE) poverty mapping method developed by Elbers et al. (2003),
has been widely tested and validated throughout the world as the most efficient
for mapping poverty. One advantage of this method is that in addition to estimating
poverty prevalence it also estimates the expected standard error as well.
With technical support from Massey University in New Zealand, the BBS and
the WFP produced absolute poverty and extreme poverty maps for 2001 at the
Upazila (sub-district or county) level based on the SAE method using a 5 % sample

2
Divisions are the largest geopolitical administrative unit in Bangladesh, followed in diminishing
scale by districts, sub-districts or Upazilas, and unions. There are seven divisions, 64 districts, over
500 Upazilas, and over 4,000 unions in Bangladesh.
104 N.Y. Choudhury and C. Räder

from the 2001 Population Census3 (BBS 2001) and the 2000 HIES (BBS 2003).4
The method takes advantage of the strengths of both sources. In the case of the HIES
data, its strength is that direct measures of poverty (i.e., income and expenditure
data) are available. In the case of the census data its strength is associated with its
scale; data was collected from every household in the country as opposed to being a
sub-sample of a larger population.
Both maps became outdated. With the publication of the 2005 HIES data (BBS
2007) there was growing demand for upgraded poverty maps. In response to this demand
the BBS, the World Bank, and the WFP produced Upazila-level poverty maps for 2005
using the 2001 census and the 2005 HIES datasets (BBS et al. 2009a, b, 2011). The
2004 Population Sample Census, which was representative at the district level, was also
used to validate the findings of the poverty-mapping exercise (BBS 2005; Fig. 7.1).
The poverty map and extreme poverty map are useful for different purposes. One
possible application of the extreme poverty map would be for the prioritization of
areas identified with a high prevalence of ‘extreme poverty’ in terms of resource
allocations. Areas with a higher prevalence of extreme poverty are also areas where
vulnerability, deprivation, and food insecurity are likely to be more severe, therefore
planning interventions to address the worst situations might benefit more from con-
sideration of the extreme poverty map as opposed to the poverty map (Fig. 7.2).

7.4 Utilization of Poverty Maps for Prioritizing

Areas for Social Safety Nets

The key focus of WFP’s safety net and development programs are the extremely
poor. The WFP in Bangladesh is a pioneer in the use of poverty mapping tools for
geographic targeting and resource allocation, both for its own efforts as well as for
governmental social safety net programs. The major preventative social safety net
efforts that the WFP operates in Bangladesh are ‘school feeding,’ ‘community nutri-
tion,’ VGD,5 and ‘disaster risk reduction’ programs. In addition to these regular
programs, the WFP also implements natural disaster relief and recovery programs.
Through its school feeding program the WFP provides micronutrient fortified
biscuits to primary school students. The program goals are to increase class atten-
dance and reduce micronutrient deficiencies among school children. The program is
operational in selected Upazilas that have high-levels of extreme poverty and poor
performance on education indicators.
The VGD program is a nationwide safety net that is now exclusively operated by the
Ministry of Women and Children’s Affairs, with technical and facilitation support from

3
During the poverty mapping exercise, the complete Population Census 2001 data (BBS 2001) was not
available therefore a 5 % sample from the census was used. Later after the complete census data became
available the 5 % sample data was verified and no significant variation was found between the two.
4
The International Rice Research Institute also produced a rural poverty map using the 2001
Population Census data and the 2000 HIES data.
5
The WFP operated the VGD program until the end of 2010 and in 2011 transferred program
operations entirely to the government.
7 Targeting the Poorest and Most Vulnerable: Examples from Bangladesh 105

Fig. 7.1 Bangladesh poverty map 2005 (BBS/WB/WFP 2009a)

the WFP. All VGD program participants are women from the poorest households. The
major entitlement of this program is food, which is complemented with training on
income generating activities and nutrition. Due to limited resources, a certain num-
ber of extremely poor households are selected from each Upazila as beneficiary
households and these numbers are determined using the extreme poverty map.
The community nutrition program provides a package of services to certain
groups that are vulnerable to nutrient deficiencies. In particular the program targets
moderately undernourished children aged 6–59 months and pregnant and lactating
women, from areas with a high prevalence of poverty. To enhance the effectiveness of
the community nutrition program it is integrated with other social safety net programs.
The WFP disaster risk reduction program enhances community resilience to differ-
ent types of natural disasters and the effects of climate change, while strengthening
106 N.Y. Choudhury and C. Räder

Fig. 7.2 Bangladesh extreme poverty map 2005 (BBS/WB/WFP 2009b)

household food security. To do this, community infrastructure such as roads, embankments,

and flood and cyclone shelters are built and maintained. These interventions are mostly
targeted in disaster-prone areas with a high prevalence of extreme poverty.

7.5 Geographic Targeting: Prioritizing Assistance Areas

for the School Feeding, Community Nutrition,
and Disaster Risk Reduction Programs

Bangladesh has been quite successful in increasing primary school enrollment,

however, achieving quality universal primary education by 2015 remains a challenge
due to irregular attendance, low primary school completion levels, and the poor
7 Targeting the Poorest and Most Vulnerable: Examples from Bangladesh 107

Extreme Primary education completion rates

poverty 1. Very low 2. Low 3. Moderate 4. Good 5. Better
rates <38% 39–49% 50–60% 61–71% >71%
1.Very high 1st Priority: 1st Priority: 2nd Priority: 3rd Priority: 3rd Priority:
>44% Very severe Severe Moderately Less Less Severe
severe Severe
2. High 1st Priority: 1st Priority: 2nd Priority: 3rd Priority: 3rd Priority:
43–33% Very severe Severe Moderately Less Less Severe
severe Severe
3. Moderate 2nd Priority: 2nd Priority: 2nd Priority: 4th Priority: 4th Priority:
32–23% Moderately Moderately Moderately Least Least Severe
severe severe severe Severe
4. Low 4th Priority: 4th Priority: 4th Priority: 4th Priority: 4th Priority:
22–11% Least Least Severe Least Severe Least Least Severe
Severe Severe
5. Very Low 4th Priority: 4th Priority: 4th Priority: 4th Priority: 4th Priority:
<10% Least Least Severe Least Severe Least Least Severe
Severe Severe

Fig. 7.3 Area prioritization matrix for the WFP school Feeding program (WFP 2011a)

quality of education. Both the government and NGOs, including the WFP, are
working to improve enrollment and attendance in primary education through
various interventions. In this context the WFP attempts to prioritize areas for the
school feeding program based on primary education outcomes and poverty
prevalence.
Areas were prioritized for the school feeding program based on two indicators:
the percentage of primary school students reaching grade five (DPE 2008) and the
prevalence of extreme poverty (BBS et al. 2009b). Both indicators were divided
into five categories with specific thresholds. The thresholds were determined by
applying a Geographic Information System (GIS) based data classification tech-
nique called ‘natural break.’ Upazilas with greater than 33 % poverty levels and
primary education completion of less than 49 % were identified as top priority
areas for the school feeding program. Upazilas with 32–23 % poverty levels and
50–60 % primary education completion were given second priority. In Fig. 7.3 is
a summary of the method of geographic prioritization for the school feeding
program, followed by an illustration (Fig. 7.4) of the prioritization process for the
school feeding program as guided by the matrix. For the sake of clarity only the
scenarios for the Rajshahi and Rangpur divisions in the northwest of Bangladesh
have been presented here.
Using similar methods, area prioritization was determined for two other WFP
safety net efforts, the ‘disaster risk reduction’ and ‘community nutrition’ programs.
The cut-off values that were used to classify poverty estimates and other indicators
were also determined using the GIS ‘natural break’ data classification system. The
thresholds may vary depending on the scenario (i.e., the highest and lowest values
in the database).
108 N.Y. Choudhury and C. Räder

Fig. 7.4 Area prioritization for the school feeding program in the divisions of Rajshahi and
Rangpur (WFP 2011b)

Geographic prioritization for the disaster risk reduction program was based on
two indicators, the prevalence of extreme poverty and vulnerability to flooding and
cyclones. This second indicator was derived from historical data. The final results
from the analysis showed that high-priority areas for disaster risk reduction activi-
ties were along the major rivers and coastal region. During the analysis it was
observed that areas prone to severe disasters also had a very high prevalence of
absolute and extreme poverty, indicating a strong association between natural
disasters and poverty (Figs. 7.5 and 7.6).
Due to the absence of Upazila-level nutrition or diet information the targeting
for the community nutrition program was based only on the extreme poverty map
due to the strong linkage between poverty and malnutrition. To more clearly
observe the impacts of the program it was decided not to implement the program
in areas where the government-operated National Nutrition Programme was
being implemented.
Because the WFP works with limited resources, most programs are implemented
in top priority areas only. Often even more selective prioritization becomes
essential due to funding shortfalls, however, the WFP advocates for the other
priority areas to donors and development partners. Recently the Ministry of Primary
and Mass Education utilized the WFP school feeding priority map in order to plan
the implementation of their own program, ‘School Feeding in Poverty-prone
Areas,’ in the remaining high-priority areas where the WFP programs were unable
to operate.
 7
Targeting the Poorest and Most Vulnerable: Examples from Bangladesh

Fig. 7.5 Prioritization of areas for the disaster risk reduction program (WFP 2011c)
109
110 N.Y. Choudhury and C. Räder

Fig. 7.6 Priority areas for the community nutrition program (WFP 2011d)

7.5.1 Resource Allocation for National Safety Net Programs

Guided by Poverty Estimations

The extreme poverty map has been used extensively by the WFP and the Government
of Bangladesh to plan resource allocation for nationwide safety-net efforts like the
VGD and ‘100-Days Employment Generation’ programs. In such programs area
prioritization was not an issue and all Upazilas were entitled to benefits, however,
the number of extreme poor selected from within each Upazila for program enti-
tlements was a major concern. The extreme poverty map provided a neutral basis for
determining the total beneficiary numbers, where the numbers were proportional to
7 Targeting the Poorest and Most Vulnerable: Examples from Bangladesh 111

the per capita rates of extreme poverty in each Upazila. For instance, in the VGD
program there were 750,000 households nationwide that were entitled to benefits.
Based on poverty estimates, Upazilas where extreme poverty levels were higher had
a larger number of beneficiary households selected.

7.5.2 Advantages and Limitations of Geographic Targeting

There are many advantages to geographic targeting: it is administratively simple, there

is no bias in the selection of areas, political compromises may be avoided, and the most
appropriate areas can be selected. One disadvantage of geographic targeting is that it
will perform poorly as a prioritization method when poverty or the desired target popu-
lation is not spatially concentrated. Performance also depends on the accuracy of the
map, a concern that is becoming less relevant over time as the small-area estimation
technique has advanced in its sophistication and application. The poverty and extreme
poverty maps only geographically target vulnerable areas, leaving marginal populations
living in less poverty-prone areas without social program entitlements.

7.6 Beneficiary Targeting: Identification of the Poorest

and Most Vulnerable in the WFP Safety Net Programs

Geographic targeting only identifies areas where assistance is needed. It provides little
insight into the characteristics of the beneficiaries for whom the social safety nets are
intended. Therefore geographic targeting is followed by beneficiary targeting.

7.6.1 Criteria Used to Target the Poorest and Most Vulnerable

The WFP safety-net programs target the poorest most food insecure households, as
well as nutritionally vulnerable social groups such as children under the age of 5,
and pregnant and lactating women. WFP usually provides food and cash assistance
to poor, food insecure households, while nutritionally vulnerable groups receive
nutritious or fortified food combined with nutrition training. In protective and
promotional safety-net programs, the food or cash entitlements are usually transferred
to one individual per household, but the entitlement benefits are shared among
all household members. Certain generic criteria are used by the WFP and other
stakeholders to identify the poorest and most food insecure households. The selection
criteria are usually applied in combination and are conditional, for example a household
should meet three out of five of the following criteria:
• Land ownership of less than 0.06 ha
• Female-headed households (widowed, separated, divorced, or with disabled
husband)
112 N.Y. Choudhury and C. Räder

Preparation of a preliminary Verification of the list: Finalization of the list:

list:
Around 25% of the With satisfactory verification,
A union-level VGD committee households enlisted in the the list is sent to the Upazila-
comprised of locally elected preliminary list are verified level VGD committee comprised
representatives (i.e., Union by a team comprised of the of elected representatives, NGO
Parishad chairperson and Upazila chairperson and representatives, and Upazila-
members) selects extremely poor board members, NGO level government officials.
households/individuals with the representatives, community Upazila VGD committees review
help of ward-level local leaders, and government the preliminary list and the
representatives. The selection is officials who visit the verification report prior to final
based on a set of criteria prepared selected households. approval.
and circulated by the WFP and
the relevant government ministry.

Fig. 7.7 Selection of the poorest and most vulnerable households by local government in the joint
WFP-government partnership program VGD (WFP 2009)

• Extremely-low and/or irregular household income from day or casual labor and
virtually no productive assets
• Household members affected by chronic food shortages (i.e., who often skip
meals due to insufficient food)
• Poor housing conditions (low quality construction materials, sanitation facilities,
and household goods).
In addition to these conditional criteria, some mandatory selection criteria are
also considered during the selection of beneficiaries. The most commonly used
mandatory criteria are:
• Household members do not benefit from other social safety-net programs
• The selected participant must be 18–49 years of age and physically able to
participate in program activities

7.6.2 Beneficiary Selection Process

The selection of the poorest and most vulnerable households and individuals based
on the criteria is either performed by local government, or by NGOs and the
communities. In both cases the WFP oversees the selection process. The two illus-
trations summarize the selection process for program beneficiaries in social safety
nets that address food security (Figs. 7.7 and 7.8).
The identification of nutritionally vulnerable individuals by the WFP community
nutrition program follows a different process. This program is entirely operated by
the WFP and NGOs without government involvement. The NGOs initially select
female volunteers called village nutrition promoters from communities and train
them to select beneficiaries. The village nutrition promoters then select undernourished
children under age 5 and pregnant and lactating women from the communities
through a screening process. The NGOs conduct sample verification of the initially
selected beneficiaries and finalize the list.
7 Targeting the Poorest and Most Vulnerable: Examples from Bangladesh 113

The NGO conducts “Participatory

Rapid Appraisals” in the community The NGO conducts a mini-survey to capture
using social-mapping and wealth- baseline information on the poorest listed
ranking techniques to identify participants, analyzes the survey data, and
individuals from the poorest households shortlists the most needy participants based
and prepares the preliminary list of on the selection criteria.
participants.

The list is reviewed and finalized

The final list is endorsed by the Upazila after sample verification of the
committee (comprised of local representatives) listed participants by the WFP
with NGO participation. and government counterparts.

Fig. 7.8 Selection of participants from the poorest households by NGO partners in a joint WFP-
government partnership Disaster Risk Reduction program (WFP 2007)

7.6.3 Limitations and Challenges in Targeting

the Poorest and Most Vulnerable

Properly targeting the poorest and neediest is an essential requirement for social
safety nets to function effectively. A World Bank (2006) study reported that 27 % of
the VGD beneficiaries and 47 % of the Primary Education Stipend6 Programme
beneficiaries in Bangladesh were not poor. An evaluation of the government oper-
ated Employment Generation Programme that ostensibly targeted the poor also
found that one-third of the program participants were from middle-income families
(MoFDM 2009).
The exclusion of individuals who are most in need of social safety nets due to
resource constraints is unavoidable. However, providing benefits to people who
do not belong to the intended target population (inclusion and exclusion errors) is a
key concern for all stakeholders. The major reasons for inclusion and exclusion
errors in beneficiary selection are as follows:
• Political influence, social and personal biases lead to the inclusion of unintended
beneficiaries.
• A general set of selection criteria applied throughout the country may leave out
area-specific characteristics of poverty.
• Some of the beneficiary selection criteria are debatable on the ground of exclusion.
For instance certain safety net programs that target reproductive or working age
women do not consider participants over 49 years of age. This could prevent
vulnerable women beyond this age who are mentally and physically fit from
participating in social safety nets. Alternatively physical fitness could be substituted
for age criteria.

6
The Primary Education Stipend Programme is a state-run social safety net for primary school
students from poor households who achieve satisfactory exam results and regularly attend school.
114 N.Y. Choudhury and C. Räder

The involvement of local governments can enhance government proprietorship

in the implementation of the program, though it also presents many risks: favorit-
ism, bias towards the better-off, and the denial of benefits to the neediest community
members. Such a system bears the risk of exacerbating and perpetuating existing
patterns of social exclusion. Monitoring of the selection by an unbiased third
party may reduce inclusion errors to some extent. Moreover sensitizing local
communities to the selection procedures may create greater local awareness of
selection anomalies.
Community-based targeting often uses a group of community members who are
not eligible for the transfer programs to decide who in the community should
benefit. The mere presence of influential members of the community during the
selection process can influence the choices. In general, community-based targeting
has more advantages than drawbacks. It relies on local information about individual
circumstances, which is generally more accurate than information available from
any other method. This approach also introduces local definitions of need and
welfare. Hence community-based targeting tends to be most appropriate where
local communities are well-established and cohesive.
The targeting mechanisms described here are applicable to, and have been
practiced in rural situations. In the design of interventions for urban poor living in
slums both geographic and beneficiary targeting are complex. Targeting urban
populations entails additional challenges such as the lack of information on slum
conditions and residents, the frequent shifting of slum households, and the diversified
livelihoods of the urban poor, etc. Currently development partners working for the
urban poor in Bangladesh have developed various targeting mechanisms that need
to be consolidated and strengthened.

7.7 Conclusion

The concepts of identifying the poorest areas and the poorest and most vulnerable
populations are gaining popularity among development practitioners, despite criti-
cisms of the spatial restriction of social safety net programs to certain areas. The
process of targeting the poorest areas and individuals still lacks rigor, especially in
government operated social safety nets, diminishing the efficiency and effectiveness
of programs. Technical expertise on the utilization of poverty maps in Asia at this
point is still largely confined to the WFP. Such capacities need to be transferred to
governmental and non-governmental counterparts in order to help improve target-
ing mechanisms on broader levels. Defining and articulating the targeting criteria
and process to the public provides transparency and accountability, which are essen-
tial for reducing the chances of misappropriation of social safety net benefits.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.
7 Targeting the Poorest and Most Vulnerable: Examples from Bangladesh 115

References

BBS (2001) Population census report 2001. Bangladesh Bureau of Statistics, Dhaka
BBS (2003) Report of the household income and expenditure survey 2000. Bangladesh Bureau of
Statistics, Dhaka
BBS (2005) Population sample census 2004. Bangladesh Bureau of Statistics, Dhaka
BBS (2007) Report of the household income and expenditure survey 2005. Bangladesh Bureau of
Statistics, Dhaka
BBS/WB/WFP (2009a) Bangladesh poverty map 2005. Bangladesh Bureau of Statistics/World
Bank/World Food Programme, Dhaka/Washington, DC
BBS/WB/WFP (2009b) Bangladesh extreme poverty map 2005. Bangladesh Bureau of Statistics/
World Bank/World Food Programme, Dhaka/Washington, DC
BBS/WB/WFP (2011) Updating poverty maps: Bangladesh poverty maps for 2005. Bangladesh
Bureau of Statistics/World Bank/World Food Programme, Dhaka/Washington, DC
DPE (2008) Primary education survey 2008. Bangladesh Ministry of Mass and Primary Education,
Directorate of Primary Education, Dhaka
Elbers C, Lanjouw JO, Lanjouw P (2003) Micro-level estimation of poverty and inequality.
Econometrica 71(1):355–364
MoFDM (2009) 100 Days employment generation program, implementation guideline 2009.
Bangladesh Ministry of Food and Disaster Management, Dhaka
WFP (2007) Operational guidelines for the planning and implementation of the World Food
Programme—assisted enhancing resilience component of the Country Programme for
Bangladesh 10410.0 (2007–2011). World Food Programme, Dhaka
WFP (2009) Vulnerable group development programme implementation guideline. World Food
Programme, Dhaka
WFP (2011a) Technical note on prioritizing upazilas for school feeding programme. World Food
Programme, Dhaka
WFP (2011b) Priority area map for school feeding programme. World Food Programme, Dhaka
WFP (2011c) Priority area map for enhancing resilience programme. World Food Programme, Dhaka
WFP (2011d) Priority area map for community nutrition programme. World Food Programme, Dhaka
World Bank (2006) Social safety nets in Bangladesh: an assessment. Bangladesh Development
series no 9. World Bank, Washington, DC
Chapter 8
Correlates of Extreme Poverty
in Rural Ethiopia

Degnet Abebaw and Assefa Admassie

Abstract This chapter examines the determinants of extreme poverty in rural Ethiopia
at the household level using indicators that reflect consumption expenditures, dietary
calorie intake, and household assets. The descriptive analyses results indicated that
ultra-poverty in the household consumption dimension was positively associated with
distance from educational and health facilities, roads, and other infrastructure. The
results of an econometric model showed that ultra-poverty was positively and signifi-
cantly associated with household size and the age of the household head, and inversely
associated with the ownership of farming assets and livestock.

Keywords Ethiopia • Ultra-poverty • Consumption expenditures • Nutrition • Asset

ownership

8.1 Introduction

In recent years Ethiopia has experienced rapid economic growth (MoFED 2011).
Given the country’s widespread poverty, high and sustained economic growth is
needed to improve the welfare of the population. According to official estimates
(MoFED 2011) each 1 % increase in national economic growth decreases the popu-
lation living in poverty by 1.7 %.
Since the early 1990s the Ethiopian government has undertaken various policy
reforms to reduce deep rooted poverty and food security problems. The current
government formulated and implemented the 1994 Agriculture-Led Industrialization
Strategy (ADLI), the 1995 Constitution (FDRE 1995), and other economy-wide as
well as sector, gender, and location oriented development strategies and programs.

D. Abebaw (*) • A. Admassie

Ethiopian Economic Policy Research Institute, Ethiopian Economics Association,
Addis Ababa, Ethiopia
e-mail: [email protected]; [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 117
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_8,
© The Author(s) 2014
118 D. Abebaw and A. Admassie

The government also began the transition from a planned to a free-market economy
immediately after taking power in 1991.
Many of the country’s poverty reduction efforts receive financial and technical
support from development partners, including bilateral and multilateral donors,
NGOs, civic societies, and the private sector. Despite the many efforts to reduce
poverty and food insecurity in Ethiopia, the country is still one of the poorest coun-
tries in the world, with low per capita income and public health service capacities,
and high illiteracy and malnutrition rates. According to the United Nations
Development Programme, Ethiopia’s human development index score ranked 157th
among 169 countries in 2010 (UNDP 2011). While economic growth is necessary
to reduce poverty and food insecurity, the extent to which the many forms of poverty
are reduced also depends on the equitable distribution of the returns of economic
growth. Poverty is still common, particularly in rural areas (MoFED 2008), and the
reduction of income discrepancy has been limited. If segments of the population are
left aside in the process of development because of political, cultural, or environ-
mental factors, then merely increasing the pace of economic growth is unlikely to
reduce poverty or food insecurity.
While the reduction of poverty is the indicated goal of various government poli-
cies and programs there is little empirical evidence of their impacts on the well-
being of the poorest, whose condition we refer to in this paper interchangeably as
either extreme poverty, ultra poverty, or marginality. The main purpose of this paper
was to examine the determinants of extreme poverty in rural Ethiopia at the house-
hold level using indicators that reflect consumption expenditures, dietary calorie
intake, and household asset ownership.

8.2 Marginality: Conceptual Issues

The definitions of marginality are diverse and many. Mehretu et al. (2000, 89)
defined marginality as “a complex condition of disadvantage that individuals or
communities may exercise because of vulnerabilities which may arise from
unequal or inequitable environmental, ethnic, cultural, social, political and eco-
nomic factors.” Gatzweiler et al. (2011, 3) defined marginality as “an involuntary
position and condition of an individual or group at the margins of societal, politi-
cal, economic, ecological and biophysical systems, preventing them from access
to resources, assets, services, restraining freedom of choice, preventing the devel-
opment of capabilities and eventually causing extreme poverty.” According to
Gatzweiler et al. (2011) marginalized social groups are products of multiple fail-
ures of markets, institutions, and policies.
Gurung and Kollmair (2005) described marginality through two major dimen-
sions, societal and spatial. The societal dimension includes demography, reli-
gion, culture, social structure, economics, and politics that limit access to
resources by individuals or groups. In this respect the focus of discussion is
often on the underlying causes of exclusion, inequality, and social injustice
8 Correlates of Extreme Poverty in Rural Ethiopia 119

(Gans 1993; Massey 1994; Sommers et al. 1999; Brodwin 2001; Leimgruber
2004; Gurung and Kollmair 2005). Marginalization may be discrimination or
suppression on the basis of the societal factors mentioned above in terms of access
to livelihood options, participation in social life, and/or public space (Brodwin
2001; Larsen 2002a, b; Gurung and Kollmair 2005). Societal marginality can
arise because of population dynamics, political instability, intensification of agri-
culture, degradation of land resources, poor access to technologies, and limited
industrial growth (Gurung and Kollmair 2005).
The spatial dimension of marginality typically results from limited access to
centers of development due to location and distance (Sommers et al. 1999;
Leimgruber 2004; Müller-Böker et al. 2004; Gurung and Kollmair 2005). In this
dimension of marginality the impacts of physical location on the well-being of indi-
viduals or groups and the geographical area itself are considered. There is also an in
situ level of spatial marginality or unequal development within relatively small geo-
graphical areas (Sommers et al. 1999).
Societal biases regarding gender, ethnicity, age, or social hierarchy are more
prevalent in locations that are remote and isolated from major economic and ser-
vice centers (Müller-Böker et al. 2004). The International Food Policy Research
Institute (IFPRI) characterized the ultra-poor as people living in remote rural
areas furthest from roads, markets, schools, and health services with few assets
and little education or access to credit, or those who are excluded because of their
ethnicity, gender, or disability (IFPRI 2007). The Moro, a remote West African
tribe whose members have little education, land, or other assets have been charac-
terized as ultra-poor (Ahmed et al. 2007; von Braun et al. 2009). The predominant
social groups among the ultra-poor are women, the disabled, the elderly, and
orphans (Ahmed et al. 2007).
Marginality may be dynamic and relative, emerging and evolving over time.
Kirkby (2000) asserted that marginality can reinforce itself, but may also be over-
come. The ultra-poor may remain poor because of complex factors and/or from
being trapped in situations that perpetuate poverty (IFPRI 2007). Marginality is a
relative term and may exist in any nation despite economic status; however, the
depth and nature of marginality are a function of a nation’s economic, political,
social, and technological development.
Mehretu et al. (2000) distinguished between contingent and systemic marginal-
ity. They describe contingent marginality as a condition of being adversely affected
by competitive inequality in free-market dynamics. This form of marginality affects
those that are least prepared to successfully negotiate in the marketplace due to liv-
ing in remote locations, inadequate labor skills, or a lack of useful information
about opportunities (Castells 1989). Systemic marginality is created by socially
constructed non-market biases often used by a dominant social class to maintain
political control, social exclusion and/or economic exploitation over subordinate
classes (Gans 1993; Mingione 1996). Tribal-based exclusionary marginalization in
Rwanda, Ethiopia, and Sudan are excellent examples of social constructions result-
ing in systemic marginality (Palmer 1977). Policies designed to redress contingent
marginality may be inherently inappropriate for situations of systemic marginality.
120 D. Abebaw and A. Admassie

8.3 Data and Empirical Approach

8.3.1 Data and Geographic Context

The data used for this paper were obtained from the Central Statistical Agency’s
(CSA) Welfare Monitoring Survey (WMS) conducted in 2004 (CSA 2007a) and
Household Income Consumption and Expenditure (HICE) survey conducted in
2004 and 2005 (CSA 2007b). The WMS provided data on various socioeconomic
indicators, including: access to health, education, and nutrition, the utilization of
basic facilities and services, and other non-income aspects of well-being. The HICE
survey provided data on household income, consumption, and expenditures. With
the exception of the regional state of Gambella and parts of Afar and Somali that
were all excluded from the WMS and HICE efforts, the datasets covered a nation-
wide sample of 36,352 (24,152 rural and 12,200 urban) and 21,600 (9,494 rural and
12,106 urban) households respectively. Our analyses focused on the rural household
data from both surveys.
We combined and analyzed the WMS and HICE data on household consumption
expenditures, dietary calorie intake, and asset ownership to examine the determi-
nants of extreme poverty in rural Ethiopia. The correlates used to characterize the
poverty outcomes were a host of household demographic and socioeconomic vari-
ables. Most of the data used in our empirical analyses were derived from 9,494
households that were common observation units in both the WMS and HICE efforts.

8.3.2 Empirical Approach

Both qualitative and quantitative methods can be used to analyze marginal poverty.1
Due to data availability we used quantitative methods, including both descriptive
analysis and econometric estimation, to characterize the extreme poverty of the
sample households. We used three household well-being indicators to analyze pov-
erty across the survey households: Adult-Equivalent Consumption Expenditure
(AECE), Adult-Equivalent Caloric Intake (AECI), and asset ownership.
Household criteria for extreme poverty within the value range of each of these
indicators were defined as follows. Based on AECE values, households were des-
ignated as extremely poor if consumption expenditures2 were in the bottom

1
Gurung and Kollmair (2005) identified various qualitative and quantitative indicators of the
degree of marginality. Qualitatively, marginality could be assessed in terms of violations of social
and legal norms such as child labor, gender inequality, and social and human rights exclusions.
Assessing the existence and magnitude of marginality through qualitative techniques is beyond the
scope of this paper because of the lack of appropriate data.
2
Consistent with other studies (MoFED 2008), we used a nutrition (calorie) based approach
(Dercon and Krishnan 1998).
8 Correlates of Extreme Poverty in Rural Ethiopia 121

quintile of the distribution. Regarding AECI, the cut-off point for extreme poverty
was designated at 1,600 kcal/day per adult. The asset ownership indicator was
based on range of important farming and household assets (Baker and Schuler
2004). We also attempted to accommodate several explanatory household variables
in the models based on marginality literature and the data available from the WMS
and HICE survey datasets. The household variables considered in the econometric
analysis were the following:
Education (Hheduc). The education level of the household head is often a key deter-
minant of household welfare and was measured in the number of years of school
completed.
Household size (Hhsize). Land is state property in Ethiopia and per capita land-
holding sizes are declining in rural areas, therefore we expected large households
(measured by the number of members) to face greater difficulty in meeting nutri-
tional needs.
Marital status of household head (Hhmar). Marriage is a form of social capital and
therefore we expected households with single heads (denoted using 0) to be at
higher risk of extreme poverty than married household heads (denoted using 1).
Gender (Hhgender). Female-headed households in Ethiopia face multiple disadvan-
tages and thus we expected these households (denoted using 0) to be more vulner-
able to extreme poverty than male-headed households (denoted using 1).
Age of household head (Hhage). The age of the household head was used as a proxy
for life stage and productivity because older household heads were expected to have
greater professional knowledge and experience, and a larger amount of family labor
available. We also included a quadratic-term (age square) to capture negative
impacts of age after a certain age threshold due to elderliness.
Livestock size (Livestock). Livestock are one of the few common assets and an
important source of capital in rural Ethiopia. The relative size of livestock was
used in two of the Probit regressions as an explanatory variable with the expecta-
tion that households with healthier livestock would be less associated with
ultra-poverty.
Farm assets (Farm_material). This index value was calculated through a factor
analysis as a continuous variable based on household ownership of agricultural
tools and equipment, which we expected to have an inverse relationship with
extreme poverty.
Radio ownership (Radio). One of the few means of getting current information rel-
evant to the productivity and well-being of rural households in rural Ethiopia is via
radio, therefore we used a dummy variable (1 = radio ownership, 0 = otherwise) to
observe if there was an association between radio ownership and extreme poverty.
Regional dummies. We also included regional dummy variables to observe whether
regional differences in geographic, historical, cultural, or socioeconomic factors
influenced marginality.
122 D. Abebaw and A. Admassie

Table 8.1 Regional Regional state or city Number of households Percentage

distribution of the rural
Tigray 851 9.0
household sample
Afar 419 4.4
Amhara 2,029 21.4
Oromiya 2,325 24.5
Somali 484 5.1
Benishangul-Gumuz 537 5.7
S.N.N.P. 2,000 21.1
Harari 288 3.0
Addis Ababa 276 2.9
Dire Dawa 285 3.0
Total 9,494 100.0
Based on WMS and HICE data from CSA (2007a, b)

8.4 Descriptive Assessment of Poverty Situation

Of the total rural household sample (9,494), 77 % (7,351) were male-headed and
23 % (2,143) were female-headed. The majority of households (88 %) were headed
by people aged 15–65 years and the remaining 12 % had older heads. Households
with married heads accounted for 78.4 % of the sample, 18.7 % were divorced,
separated, or widowed and the remaining 2.9 % had never married. Approximately
73 % of the household heads were illiterate, less than 20 % had completed primary
school, and only 8.4 % had attended secondary education or above.
The sample households were located in eight regional states and two city admin-
istrations of Ethiopia. The three most populous states, Oromiya, Amhra and SNNP,
accounted for two-thirds of the sample. Sample households from rural areas of the
two city administrations, Addis Ababa and Dire Dawa, constituted around 6 % of
the total sample. The remaining 28 % were drawn from the other regional states, the
least developed of which (Afar, Benishangul-Gumuz, and Somali), accounted for
around 15 % of the total (Table 8.1).

8.4.1 Consumption Expenditures and Dietary Calorie Intake

Consumption expenditures and caloric intake are two important indicators of house-
hold well-being. According to HICE survey data the mean household AECE in
2004–2005 was Birr 2,217.26 or US$256.33 (see Table 8.2). The distribution
revealed significant disparity, ranging from Birr 1,078.46 (US$124.68) in the lowest
quintile to Birr 4,018.07 (US$464.07) in the highest quintile, indicating a nearly
four-fold difference among rural households. Table 8.2 presents the AECI values
separated by AECE quintiles to examine the association between consumption
expenditures and caloric intake among the survey households. Households in the
8 Correlates of Extreme Poverty in Rural Ethiopia 123

Table 8.2 Consumption expenditures and calorie intake distributions by AECE quintile
Consumption expenditure Daily calorie intake
AECE quintile (average in Birr) (average in kcal)
1 1,078.46 1,825
2 1,549.97 2,485
3 1,981.34 3,012
4 2,458.64 3,332
5 4,018.07 3,899
Average 2,217.26 3,332
Based on HICE data from CSA (2007b)

Table 8.3 Regional distribution of daily household AECI values

Net daily kilocalories
Less than Between 1,600 More than Deviation from
Regional state or city 1,600 and 2,200 2,200 national mean
Tigray 10.9 24 65.1 −6.8
Afar 24.7 29.8 45.5 −7.5
Amhara 15.5 17.4 67.1 −9.8
Oromiya 9.1 19.5 71.4 +4.7
Somali 8.3 26.5 65.3 −3.4
Benishangul-Gumuz 18.8 19 62.2 −5.6
SNNPR 9.1 19.2 71.7 +5.7
Harari 4.7 15.7 79.7 +15.1
Addis Ababa 9.1 23.6 67.3 −1.7
Dire Dawa 7.2 20.4 72.4 +4.2
Nationwide 11.2 16.2 69.6 –
Based on WMS and HICE data from CSA (2007a, b)

lowest AECE quintile also had AECI values that were about half that of the highest
AECE quintile. The greater disparity of AECE values compared to AECI values
suggests that poorer households rely on cheaper and higher calorie dietary items
(Table 8.3).
Table 8.4 presents the regional distribution of expenditure-based and diet-based
poverty in Ethiopia. As the table shows, diet-based poverty was nearly as high as
expenditure-based poverty in all regions. Mean expenditure-based poverty was
39 % in 2004–2005. Mean diet-based poverty was 38 % during the same period.
As indicated in Table 8.4 the relative amounts of extreme poverty based on
AECE and AECI appear to be the highest in the regional states of Tigray,
Benishangul-Gumuz, Somali, Afar, and Amhara. With the exception of Afar these
states also had the greatest overall poverty gap and squared poverty gap index val-
ues in the country (MoFED 2008), indicating income shortfalls for meeting basic
nutritional need.
124 D. Abebaw and A. Admassie

Table 8.4 Regional poverty variation in Ethiopia

Expenditure-based Diet-based Overall Overall squared
Regional state or city poverty poverty poverty gap poverty gap
Tigray 0.51 0.48 0.104 0.032
Afar 0.429 0.436 0.078 0.021
Amhara 0.404 0.391 0.104 0.036
Oromiya 0.372 0.371 0.075 0.024
Somali 0.452 0.439 0.099 0.03
Benishangul-Gumuz 0.458 0.459 0.106 0.035
SNNP 0.382 0.369 0.071 0.022
Harari 0.206 0.184 0.033 0.007
Addis Ababa 0.299 0.316 0.052 0.012
Dire Dawa 0.398 0.384 0.063 0.015
Nationwide 0.393 0.385 0.085 0.027
Adapted from MoFED (2008)
Diet-based poverty includes households with daily dietary intake of 1,600 kcal or less per adult

Table 8.5 Major reasons for not seeking health care among rural households by AECE quintile
No need Financially
AECE quintile Gender to consult incapable Distance Other
1 Male 38.8 9.6 1.1 0.55
Female 37.4 10.5 1.2 0.69
Total 76.3 20.1 2.3 1.24
2 Male 40.5 8.1 1.1 0.81
Female 38.3 9.1 1.2 0.73
Total 78.8 17.2 2.4 1.54
3 Male 39.9 7.1 1.2 0.88
Female 40.2 8.5 1.1 0.96
Total 80 15.6 2.3 1.85
4 Male 41.8 6.2 0.9 0.67
Female 39.6 8.3 1.3 1
Total 81.4 14.4 2.2 1.69
5 Male 40.3 7.2 1 1.03
Female 39.1 9.1 0.8 1.34
Total 79.5 16.3 1.8 2.36
Based on WMS and HICE data from CSA (2007a, b)

8.4.2 Health and Medical Services

Household health conditions and the utilization of health care varied among AECE
quintiles. The incidence of illness was higher in the bottom quintile than the top
quintile; however, health care service utilization was greater for the top than for the
bottom quintile households. This suggests that access to health care is inversely
related to income in rural areas of Ethiopia. The main reasons cited for not seeking
health care despite justifiable need included: the lack of financial resources, dis-
tance, and lack of awareness of health care services (see Table 8.5). An inverse
8 Correlates of Extreme Poverty in Rural Ethiopia 125

Table 8.6 Percentage of childbirths by type of assistance and AECE quintile

AECE quintile
Support provider 1 2 3 4 5 Total
Health care professionals 0.7 1.3 1.9 4.5 26.6 5.7
Traditional attendant 36.8 27.5 26.8 25.4 21.9 28.1
Relative or others 55.2 64.9 66.7 65.2 46.6 60.5
None 6.9 6.1 4.4 4.7 4.7 5.4
Total (%) 100 100 100 100 100 100
Based on WMS and HICE data from CSA (2007a, b)

Table 8.7 Educational status of household heads by AECE quintile

Secondary
AECE quintile Illiterate Primary (1–6) Primary (7–8) and above Other Total
1 74.2 18.3 2.3 0.8 4.4 100.0
2 72.0 20.5 2.5 1.1 3.8 100.0
3 73.0 17.8 3.1 1.7 4.4 100.0
4 71.3 20.0 2.6 2.0 4.1 100.0
5 66.4 22.3 3.2 4.1 2.8 100.0
Based on CSA (2007a)

relationship between health care and household consumption expenditure was also
observed for childbirth services.
In general the percentage of births attended by health professionals in Ethiopia is
extremely low by global standards (World Bank 2005). As might be expected, the
vast majority (88.7 %) of women gave birth at home with the assistance of relatives
and traditional midwives rather than at medical facilities attended by health care
professionals (see Table 8.6). Births assisted by health care professionals tended to
increase with asset ownership. For instance the percentage of births assisted by
health professionals for women in the fifth quintile was over 25 percentage points
higher than for women in the lowest quintile. This suggests that apart from supply-
side policy interventions to expand health care facilities, demand-side policy inter-
ventions for increasing household assets may contribute to increased health care
utilization among pregnant women.

8.4.3 Educational Status

Generally speaking the education and literacy of household heads increased from
the lowest to the highest AECE quintile. The percentage of literate household heads
increased from around 26 % in the lowest quintile to around 34 % in the highest
quintile. The proportion of household heads with secondary schooling or above
increased by more than four-fold from the lowest to highest quintiles (see Table 8.7).
126 D. Abebaw and A. Admassie

Table 8.8 Probit estimations based on household consumption expenditures

Probit regression Marginal effect
Variables Coefficient SE Coefficient SE
Hhsize 0.2924 0.0099*** 0.0702 0.0024***
Hhgender −0.0524 0.0634 −0.0127 0.0157
Hhage 0.0152 0.0073* 0.0029 0.0017*
Hhagesquare −0.0001 0.00007 −0.00002 0.00002
Hheduc −0.1562 0.0321*** −0.0375 0.0077***
Hhmar −0.076 0.068 −0.0186 0.0171
Radio −0.3505 0.0476*** 0.0757 0.091***
Livestock −0.2355 0.0293*** −0.0565 0.00703***
Farm_material −0.0772 0.0274*** −0.0185 0.00661***
Regional dummy variables
Tigray 0.0042 0.113 0.0102 0.0288
Afar −0.0627 0.1273 0.0146 0.02681
Amhara 0.0935 0.1060 0.0023 0.02681
Oromiya −0.5643 0.1073*** −0.01164 0.0187***
Somali 0.0836 0.1211 0.03208 0.0313
BenishangulGumuz −0.4344 0.1256*** −0.0845 0.0190***
SNNPR −0.2136 0.1072** −0.0480 0.02256**
Harari −0.7145 0.1567*** 0.1175 0.0154
Dire Dawa −0.0557 0.1353 −0.013 0.01629***
Constant −2.2182 0.1912***
Number of observations 8,825
Log Pseudo likelihood −3,676.76
Wald chi2(10) 1,155.39
Pseudo R2 0.165
Percent of households predicted correctly 70.4
Based on WMS and HICE data from CSA (2007a, b)
***,**, * denote statistical significance at 1, 5, and 10 %, respectively, Addis Ababa is the reference
region

8.4.4 Probit Estimation Results and Discussion

We used three binary probit models to estimate the correlates of extreme poverty as
measured by the three household well-being indicators; consumption expenditure,
calorie intake and asset ownership. We defined the dependent variables of these
models as follows. For consumption expenditure the dependent variable was coded
one if the household was in the lowest AECE quintile and zero otherwise. For calo-
rie intake the dependent variable was coded one if the daily household AECI was
below 1,600 kcal and coded zero if otherwise. For asset ownership the dependent
variable was coded one for households in the lowest index quintile and coded zero
if otherwise. The estimation results are presented in Tables 8.8, 8.9 and 8.10. The
overall fit of the estimated models was good. The pseudo R2 values ranged 0.11–
0.20 and correct predictions varied 66–74 %.
8 Correlates of Extreme Poverty in Rural Ethiopia 127

Table 8.9 Probit estimations based on household dietary caloric intake

Probit regression Marginal effect
Variables Coefficient SE Coefficient SE
Hhsize 0.2152 0.0102*** 0.0354 0.0017***
Hhgender −0.0272 0.0709 −0.0045 0.0119
Hhage 0.0081 0.0081 0.0013 0.0013
Hhagesquare −0.00006 0.00008 −0.000010 0.00001
Hheduc −0.0093 0.03278 −0.00153 0.00539
Hhmar −0.2255 0.075*** −0.0406 0.0147***
Radio −0.1089 0.0500** 0.0172 0.00755**
Farm_material −0.3351 0.0311 −0.0055 0.0051
Livestock −0.1995 0.0330*** −0.0328 0.0054
Regional dummy variables
Tigray 0.2871 0.1425*** 0.0549 0.0311*
Afar 0.8303 0.1499*** 0.2095 0.0503***
Amhara 0.6416 0.1341*** 0.1329 0.0336***
Oromiya 0.1125 0.1347 0.0192 0.0239
Somali 0.1381 0.1577 0.0246 0.0303
Benishangul-Gumuz 0.5629 0.1468*** 0.1253 0.0413***
SNNPR 0.1788 0.1361 0.03145 0.0256
Harari −0.1842 0.1950 −0.0269 0.0250
Dire Dawa −0.07121 0.1761 −0.0122 0.03155
Constant −2.594 0.2226***
Number of observations 8,824
Log Pseudo likelihood −2,812.26
Wald chi2(17) 648.48
Pseudo R2 0.1133
Percent of households predicted correctly 66.8
Based on WMS and HICE data from CSA (2007a, b)
***, **,* denote statistical significance at 1, 5 and 10 %, respectively, Addis Ababa is the reference
region

Household expenditure-based poverty. Table 8.8 presents the results of the probit
analysis of the correlates of household expenditure-based poverty. Larger house-
holds were more likely to be ultra-poor than smaller households ceteris paribus. The
effect of household head gender was not significant. The marital status of the house-
hold head had a negative association with extreme poverty but was not significant. As
expected, education of the household head had a negative but not statistically signifi-
cant association with household ultra-poverty. The ownership of a radio had a nega-
tive and significant relationship with ultra-poverty. Household asset ownership was
negatively associated with ultra-poverty. The estimation results also revealed strong
associations between location and ultra-poverty: most of the regional dummy vari-
ables had negative and significant associations with the dependent variable.
Household diet-based poverty. Table 8.9 shows the relationships between
household variables and diet-based extreme poverty. About 12 % of the sample
households qualified as marginalized based on caloric intake. In contrast to the
128 D. Abebaw and A. Admassie

Table 8.10 Probit estimation based on household assets

Probit regression Marginal effect
Variables Coefficient SE Coefficient SE
Hhsize −0.1892 0.0111*** −0.0440 0.0024***
Hhgender −0.5974 0.0512*** −0.1621 0.0157***
Hhage −0.0335 0.0061*** −0.0078 0.0015***
Hhagesquare 0.00030 0.00006*** 0.00007 0.00001***
Hheduc 0.2386 0.0246*** 0.0555 0.0057
Hhmar −0.3884 0.0571*** −0.1017 0.0166***
Regional dummy variables
Tigray −0.1553 0.1121 −0.0696 0.0195***
Afar 0.4188 0.1206*** 0.1164 0.0405***
Amhara −0.4059 0.1076** −0.0834 −0.0194***
Oromiya 0.1418 0.1069 0.0318 0.0231
Somali 0.2312 0.1257* 0.0596 0.0375*
Benishangul-Gumuz 0.2034 0.1222* 0.0518 0.0337
SNNPR 0.02162 0.107 0.0051 0.0253
Harari 0.1297 0.1404 0.0283 0.02861
Dire Dawa 0.1949 0.1392 0.041 0.0264
Constant 1.509 0.1717***
Number of Observations 8,860
Log Pseudo likelihood −3,549.22
Wald chi2(15) 1,336.06
Pseudo R2 0.1995
Percent of households predicted correctly 72.02
Based on WMS and HICE data from CSA (2007a, b)
***, **, * denote statistical significance at 1, 5, and 10 %, respectively, Addis Ababa is the refer-
ence region

consumption expenditure model, most of the regional dummy variables had

positive and significant relationships, indicating greater likelihood of diet-based
extreme poverty among households in the regional states of Tigray, Afar,
Amhara, and Benishangul-Gumuz compared to rural households located near
the capital. There was a negative and significant relationship between married
household heads and diet-based extreme poverty.
Household asset-based poverty. Unlike the expenditure and caloric intake based
model results, the age, educational status, and gender of the household head all had
unexpected association signs with extreme poverty with respect to household assets.
The probability of asset-based extreme poverty was less likely among female-
headed households compared to male-headed households. The incidence of asset-
based extreme poverty was also more prevalent for households with literate heads.
Compared to the capital city, households in Afar, Somali, and Benishangul-Gumuz
were more likely to be ultra-poor based on asset ownership, whereas those in
Amhara were less likely. The variability of the relationships of the regional dummy
variables with asset-based extreme poverty may be due to geographic variability in
the functional value of farming assets. In most parts of Afar, Somali, and some parts
of Benishangul-Gumuz farming equipment is relatively unimportant because most
8 Correlates of Extreme Poverty in Rural Ethiopia 129

of the people in these regions are pastoralists who rely little on farm tools and equip-
ment that are vital for sedentary crop-based agriculture.

8.5 Conclusion

The descriptive results indicated that ultra-poverty based on the household expenditure
dimension was positively associated with distance from educational and health facili-
ties, roads, and other infrastructure. The descriptive results also revealed considerable
overlap of the different indicators of ultra poverty or marginality among the sample
households. In particular, ultra-poor households in terms of expenditures also tended to
be ultra-poor in terms of dietary caloric intake and asset ownership. The incidence of
ultra poverty exhibited significant variation across the different regions in the country.
The results of the probit models indicated several interesting results. Ultra pov-
erty based on the household expenditure dimension was positively and signifi-
cantly associated with household size and the age of the household head, and
inversely with the ownership of farming assets and livestock. Ownership of a
radio and the education level of household heads were both inversely associated
with expenditure-based extreme poverty. Many of the key factors that were asso-
ciated with expenditure-based extreme poverty were similarly associated with
diet-based extreme poverty; however, the relationships of education of the house-
hold head and farming assets did not have statistically significant relationships
with diet-based extreme poverty. There was a negative and significant relationship
between married household heads and diet-based extreme poverty. Unlike the
results for expenditure-based and diet-based extreme poverty, household size had
a negative and statistically significant association with asset-based ultra-poverty.
There was also a negative and statistically significant association between female
household heads and asset-based poverty. Also contrary our expectation, the inci-
dence of asset-based extreme poverty was positively associated with the educa-
tion level of the household head. In all three poverty dimensions most of the
regional dummy variables were strongly associated with extreme poverty, sug-
gesting considerable marginality in rural areas throughout the country.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Ahmed AU, Hill RV, Smith LC, Wiesmann DM, Frankenberger T (2007) The world’s most
deprived: characteristics and causes of extreme poverty and hunger. 2020 discussion paper 43,
International Food Policy Research Institute, Washington, DC
Baker J, Schuler N (2004) Analyzing urban poverty: a summary of methods and approaches. World
Bank Policy Research working paper no 3399, World Bank, Washington, DC
130 D. Abebaw and A. Admassie

Brodwin P (2001) Marginality and cultural intimacy in a trans-national Haitian Community.

Occasional paper no. 90, Department of Anthropology, University of Wisconsin, Milwaukee
Castells M (1989) The information city. Blackwell, Oxford
CSA (2007a) Welfare Monitoring Survey 2004. Statistical report on basic population characteris-
tics, education, health, nutritional status and child care. Statistical bulletin, Central Statistical
Authority, Addis Ababa
CSA (2007b) 2004–05 Household income, consumption and expenditure survey. Statistical
bulletin 394, Central Statistical Authority, Addis Ababa
Dercon S, Krishnan P (1998) Changes in poverty in rural Ethiopia 1989–1995: measurement,
robustness tests and decomposition. Center for the Study of African Economies WPS/98–7,
Centre for the Study of African Economies, University of Oxford, Oxford
FDRE (1995) Constitution. Federal Democratic Republic of Ethiopia, Addis Ababa
Gans HJ (1993) From underclass to undercaste—some observations about the future of the postin-
dustrial economy and its major victims. Int J Urban Reg Res 17:327–335
Gatzweiler F, Baumüller H, von Braun J, Ladenburger C (2011) Marginality: addressing the root
causes of extreme poverty. ZEF working paper no 77, Center for Development Research,
University of Bonn, Bonn
Gurung GS, Kollmair M (2005) Marginality: concepts and their limitations. IP6 working paper,
Swiss National Centre of Competence in Research North South, Bern
IFPRI (2007) Taking action for the World’s Poor and Hungry People: the 2020 conference at a
Glance, Beijing, 17–19 October 2007. International Food Policy Research Institute,
Washington, DC
Kirkby J (2000) Land degradation and development, optimism or pessimism? SAPES Trust,
Harare
Larsen JE (2002a) Who cares about and for marginal people? Danish Social Science Research
Council, Copenhagen. http://www2.ihis.aau.dk/gep/publicationer/nr3.pdf. Accessed 12 Apr
2011
Larsen JE (2002b) Spatialization and culturalization of social policy: conducting marginal people
in local communities. Paper presented at the area-based initiatives in contemporary urban pol-
icy conference, Copenhagen, 17–19 May 2001. Danish Building and Urban Research
Association, Copenhagen
Leimgruber W (2004) Between global and local: marginality and marginal regions in the context
of globalization and deregulation. Ashgate Publishing, Farnham
Massey D (1994) Space, place and gender. University of Minnesota Press, Minneapolis
Mehretu A, Pigozzi BW, Sommers LM (2000) Concepts in social and spatial marginality. Geogr
Ann 82(2):89–101
Mingione E (1996) Urban poverty in advanced industrial world: concepts, analysis and debates. In:
Mingione E (ed) Urban poverty and the underclass. Blackwell, Oxford
MoFED (2008) Dynamics of growth and poverty in Ethiopia (1995/96–2004/05). Ministry of
Finance and Economic Development, Addis Ababa
MoFED (2011) Ethiopia five years (2010/11–2015/16): growth and transformation plan. Ministry
of Finance and Economic Development, Addis Ababa
Muller-Böker U, Geiger D, Kansakar VBS, Kollmair M, Molesworth K, Suleri A (2004) Sustainable
development in marginal regions of South Asia. In: Hurni H, Wiesmann U, Schertenleib S (eds)
Research for mitigating syndromes of global change: a trans-disciplinary appraisal of selected
regions of the world to prepare development—oriented research partnership. Schlaefli and
Maurer Publishing, Bern
Palmer RH (1977) Land racial discrimination in Rhodesia. University of California Press, Berkeley
Sommers LM, Mehretu A, Pigozzi BW (1999) Towards typologies of socio-economic marginality:
north/south comparisons. In: Jussila H, Majoral R, Mutambirwa C (eds) Marginality in space—
past, present and future: theoretical and methodological aspects of cultural, social and eco-
nomic parameters of marginal and critical region. Ashgate Publishing, Farnham
UNDP (2011) Human development report 2011. United Nations Development Programme, New
York
8 Correlates of Extreme Poverty in Rural Ethiopia 131

von Braun J, Hill RV, Pandya-Lorch R (2009) The poorest and the hungry: a synthesis of analyses
and actions. In: von Braun J, Hill VR, Pandya-Lorch R (eds) The poorest and hungry: assess-
ments, analyses, and actions. International Food Policy Research Institute, Washington, DC,
pp 1–61
World Bank (2005) Ethiopia: a country status report on the health and poverty. Volume II, report
no. 28963-ET, World Bank, Washington, DC
Chapter 9
Examining the Circle of Attachment Trauma,
Shame, and Marginalization: The Unheard
Voices of Young Kutchi Girls

Manasi Kumar

Abstract This chapter offers a psychological understanding of the experience of

social marginality as viewed from the perspective of young girls from the Indian
province of Gujarat. Secure attachments are one of the primary ‘capabilities’ that
have direct bearing on an individual’s sense of identity and freedom. Insecure
attachments, particularly dismissing kinds, lead to inhibitions in personality devel-
opment and build up layers of shame and self-doubt. The author examines how the
psyche is tormented by repeated experiences of social marginalization in the form
of dismissal at the hands of family, and how shame becomes an abiding emotion—
creating further doubts, disenfranchisement, and alienation.

Keywords Psychological attachment • Emotions • Identity • Freedom

9.1 Introduction

Let me start by explaining how I arrived at this undertaking. I am neither a development

practitioner nor an economist; rather I am a psychologist interested in social issues
such as poverty, gender inequity, environmental degradation, violence, etc. My
interest in real-life events and their ramifications entail a bifocal examination: one
aspect pertaining to how the outside world changes and impinges on the individual
and another on how individual mental frames and ways of thinking, emoting, and
feeling change behaviors and perceptions towards the world. My doctoral work at
University College in London from 2005 to 2009 examined the differential impacts
of natural disasters and social violence on the mental health of children, particularly

M. Kumar (*)
Department of Psychiatry, College of Health Sciences,
University of Nairobi, Nairobi, Kenya
e-mail: [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 133
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_9,
© The Author(s) 2014
134 M. Kumar

on the attachment representations of children. I was particularly interested in

understanding how massive trauma events trigger further trauma and psychopathol-
ogy in children, and whether social trauma has distinctive clinical signatures in that
change among communities, families, and individuals from both the inside and the
outside. The findings of that work are not the focus here, but rather I present a subset
of findings about post-disaster adversity and the social marginalization of young
girls who I interviewed in the Kutch district of the western Indian province of
Gujarat. I worked there with children who had survived the debilitating 2001 Gujarat
earthquake. In order to better understand childhood in the rural Gujarati/Indian con-
text, I interviewed children from both working-class and middle-class families,
which also offered a control sample that I needed for my disaster trauma research.
Visiting schools in the villages of Lodai and Khengarpur, the city of Bhuj, and in
Adriyana in the Surendranagar district was an eye-opening experience for me.
While the young boys played with my laptop and camera, and surrounded me asking
questions about where I was from, what I would do in my research, and generally
sharing with me about their lives in the schools and villages; the girls seemed
disinterested, withdrawn, and rarely came forward to greet me. In my ignorance
I overlooked this for a while and thought that perhaps the boys were simply less
inhibited and that perhaps the girls were not as well-educated or accustomed to
outsiders. My ‘attachment interviews’ were video recorded and those involving girls
were uneventful, as they spoke very little there were often long pauses, and a pro-
found silence would accompany anything they described. They did not make eye
contact, there was no liveliness in their tone or discussion content, and I was always
struck by something heavy building-up inside myself when I listened to these
children. Besides practical details such as the video camera losing power, the inter-
views were getting accomplished without much content and I was getting behind
schedule because single interviews lasted more than an hour. One interview after
another felt similar. After nearly 30 interviews with the girls and three to five group
exercises with the children my experiences confirmed to me that something was
indeed problematic with the girls here. My first reaction was to get irritated by their
reluctance, thinking of them as somehow less intelligent or engaging than the boys,
who seemed more free spirited and childlike. In psychology as in other disciplines
such as anthropology, a researcher’s own feelings and intuition are powerful tools
for understanding individuals and social interactions.
There were other difficulties that contributed to this situation, such as the impos-
sibility of hiring a female translator for the work with the young girls. No female
translator could be contracted because women in this region do not normally travel
or work outside their homes or the confines of their own village or town. There was
a local saying, “no women stepping outside the darbargarh” (an ancient fort inside
the nearby city of Bhuj), that also seemed to be an obvious boundary for women,
symbolizing their confinement inside the old city. Several girl participants were hesi-
tant to be interviewed or tested by a male translator, and the translator, as an adult
male, brought a certain power dynamic and unique cultural angularities into the test-
ing situation that appeared to me as if they could potentially interfere with deriving
the information needed from the participants. On several occasions I realized that a
9 Examining the Circle of Attachment Trauma, Shame, and Marginalization… 135

benign question such as inquiring whether the child remembered when the quake had
occurred was instead put to the child as a general knowledge question, requiring the
child to reproduce the exact time, date, and epicenter of the quake. This approach
made the children more anxious, making the assessment more difficult; however,
becoming aware of such dynamics motivated me to intervene myself. I decided after
three or four of such interviews that I would do the remaining testing myself despite
the language barriers and this approach was successful in the long run.
Even though no gender differences were found in the psycho-social adjustment
of the children as a result of the actual testing, the young girls appeared much more
fragile and needed more time to think and provide answers or form opinions than
the boys. At times it appeared that they had never been asked to share their personal
experiences or day-to-day problems, and that sudden testing on psychosocial adjust-
ment seemed to puzzle them. Some girls were so nervous that their hands would
shake and I was aghast to see them covering their faces before me, making me feel
as if I too became perceived as a ‘domineering male presence’ before them. For
these reasons I pulled out the interviews with the Kutchi girls and studied them
carefully after I returned to my office in London.

9.2 Attachment: The Concept and Its Social Application

Attachment in this research context is understood as a concept that permeates

culture and is embedded within a dialectical intrapsychic/intersubjective matrix
(Green 2000). Attachment is part of a sense of self that is developed through the
influence of the other. It is not only the need for the other, but also about what kind
of presence the other has had or made in a person’s life. As children mature, attach-
ment security becomes an attribute of the person rather than of relationships
(Fonagy 2001, 2005; Thompson and Raikes 2003). Attachment is an innate need
and also one of the important capabilities that sustain human relationships. At the
same time it is also embedded within an interpersonal equation, since the presence
of a vital other is needed to animate and activate relational ties. In their work on
moral philosophy Nussbaum (2004) and Lear (2007) have discussed how social
values and organization can be understood by examining the emotional capabilities
of individuals.
Attachment research over the last 30–40 years has shown that a caregiver’s
presence, contact, and comfort efficiently appease a child’s emotional state
(Harlow 1959; Bowlby 1969; Sroufe and Waters 1976; Ainsworth et al. 1978). In this
process the adults around children endow them with cultural information and
meanings that not only pass on language, social skills, and traditions etc., but that
also provide a safety net to survive in a complex world. Children listen to and are
greatly influenced by the stories narrated by their significant others and the adults
around them (Bruner 1990), and soon adopt this experienced quality of storytell-
ing. This capacity to share information and narrate, exchange, and negotiate ideas
and experiences forms the basis of later socio-emotional development.
136 M. Kumar

Research on emotions has shown that negative emotions stimulate social

comparison, narration, and conversation to fuel cognitive work and to allow a
metabolizing process. The social-representation theory of Moscovici (1984) and the
social-constructionist approaches of Bruner (1990) are examples of how distressed
individuals seek social contact and how conversations provide a medium to digest
unfamiliar or threatening situations or objects. In distressing situations the attach-
ment mechanisms get activated (Bowlby 1969) and serve two purposes: direct anxi-
ety reduction and increased cognitive clarity. It is thought that the proximity and
care of caregivers can restore these functions when such are required. An emotional
experience, particularly a negative one, stimulates social sharing of the experience
in both interpersonal as well as collective contexts. Repetitive communication
allows an emotional experience to get registered and lends itself to the evolution of
its mental representation. It is hypothesized that the quality and amount of sharing
of one’s experiences, especially negative experiences, reveal how children have
dealt with these issues in their interpersonal and social environments. Researchers
working on emotions of shame and guilt have also found that unshared emotional
episodes elicit more intense feelings of shame and guilt than shared ones (Finkenauer
and Rimé 1998).

9.3 Insights on Attachment Trauma, Shame, and Poverty:

A Discourse Analysis of Interviews with Young Girls
in Kutch

9.3.1 Methods

I quantitatively analyzed Child Attachment Interview (CAI) narratives of 12

selected girl participants from the earthquake group. I selected the participants
on the basis that their narratives provided a broad sweep of themes that repre-
sented the entire CAI sample. For this chapter a simple treatment of narratives
using thematic analysis (using actual sentences as themes), in keeping with prin-
ciples of grounded theory (Glaser and Strauss 1967) and a phenomenological
approach extended by psychological anthropology (Rosaldo 1984; Shweder and
LeVine 1984; Shweder 1991; Berry et al. 1997) were favored over other qualita-
tive techniques. Phenomenology seeks the psychological meanings that consti-
tute a phenomenon through investigating and analyzing past examples of the
phenomenon within the context of the participant’s lives (assuming that the
capacity to live through events or respond to different situations greatly exceeds
the capacity to know exactly “what we do” or “why we do what we do”).
Narratives are also seen as the performance of the self or as a story of identity
(Parker 2004). The analysis of the CAI transcripts helped unfold processes where
identity and self were threatened and/or a certain kind of identification was either
denied or affirmed.
9 Examining the Circle of Attachment Trauma, Shame, and Marginalization… 137

9.3.2 The Child Attachment Interview: The Tool

The CAI explores the way young children (8–14 years of age) think about
themselves and how they represent relationships with their parents. It includes
about 19 semi-structured questions to be pursued with the child during the course of
the interview. Shmueli-Goetz et al. (2008, 4) presented the CAI as an instrument
where “a better compromise between indirect assessment of representations and
simply using an adult interview and coding system” has been reached. Additionally,
it is designed to overcome the ‘measurement roadblock’ in attachment (i.e., the lack
of validated assessment instruments and clinical interventions with pre-pubescent
children). The interview helps assess children’s internal models of attachment
relationships by directly questioning them about their experiences with, and percep-
tions of, their primary caregivers.
The majority of the CAI scales are intended to assess the child’s overall current
state of mind with respect to attachment, a state of mind that is assumed to be
reflected in the narrative as a whole (Target et al. 2003). These scales are: Emotional
Openness, Preoccupied Anger, Idealization, Dismissal, Self Organization/Disorgan-
ization, Balance of Positive and Negative References of Attachment Figures, Use of
Examples, Resolution of Conflicts, and Overall Coherence. Three of these scales
(Preoccupied Anger, Idealization, and Dismissal) are rated separately for mothers
and fathers. All scales have value ranges from 1 to 9 denoting the lowest to highest
score respectively. All scales were coded from 1 to 9 with anchor points at odd
values illustrated with examples of episodes characteristic of that level, as well as a
description of the ways in which that aspect of the child’s representations was
manifested across an entire interview. In a nutshell, the CAIs help illuminate both
the contents of the mind and its conversation with the other.

9.3.3 Sample Characteristics

Approximately 12 interviews were conducted with girls from rural Gujarat. The
mean age of the interviewees was 11.7 years old (range = 10–14). Interviews were
typically conducted in schools at four locations across the district of Kutch: Bhuj,
Lodai, Khengarpur, and Khavda. On some occasions (such as during school holi-
days) interviews were conducted in other places such as local community halls. The
first few interviews were done with the help of a translator and subsequently my
limited but increasing familiarity with Kutchi and Gujarati enabled me to conduct
them myself. Although all 12 girls were selected from the earthquake affected areas
of Kutch, none of them were severely affected mentally or physically by the quake.
The children lived in kutcha (makeshift construction) houses and worked at home
doing household chores, looking after younger siblings and embroidering cloth at
home to help make ends meet. The average monthly income of the girl’s parents was
around 3,000 rupees (US$800) for households with an average of five to six mem-
bers. Approximately 500–700 (US$133–187) additional rupees were contributed by
138 M. Kumar

children who are old enough to embroider or engage in other paid work such as
making wristbands, brooms, local cigarettes, etc. Physically the girls appeared very
weak and severely anemic, and they often suffered from conditions such as chest
congestion, cough, ‘chikungunya’ (Alphavirus) infections, or were recovering from
waterborne parasite infections.

9.4 Analysis of the Attachment Interview Material

The section below highlights key themes that came up in the 12 interviews with the
Kutchi girls. In comparison to those conducted with young boys, the girls’ inter-
views were more cryptic, hesitant, and shorter. The nonverbal gestures and facial
expressions of these girls by contrast conveyed a lot more than their speech did.
What were revealed in the interviews were multiple narratives strewn with shame;
uncertainty about one’s identity, fate, or future; and the inhibited speech conveyed
their status as disenfranchised units of the family and community in general. Names
of the participants have been changed below and abbreviated, and the numbers indi-
cate the ages of the children interviewed.
Characteristic questions and responses on CAIs: shame, inhibition, and
uncertainties.
• How would you describe yourself? Who are you as a person?
– “I am expected to get along well with my family and be kind to them.”
(FB, 12)
– “I don’t like to fight with others.” (FB, 10)
– “I am a quiet girl.” (FB, 10)
– “I am a good girl.” (FK, 12)
– “Sometimes I think and sometimes I don’t think.” (FB, 11)
– “I would like to study so that I never feel that I did not realize my potential
like even if I get married no one is able to boss or maltreat me.” (FB, 13)
– “I do not like to be fashionable.” (FB, 13)
• How would you describe your relationship with your mother?
– “Doesn’t give me any money (but she buys me things).” (FB, 14)
– “I never go out without her.” (FB, 14)
– “I do whatever she tells me to do.” (FB, 10)
– “She is happy when I behave well.” (FB, 11)
– “Like if mummy tells me to collect all the trash, I do that. If she tells me to
wash the utensils I do that too, and if she goes out and she gets late and tells
me to do the entire house work, I do that as well.” (FB, 9)
– “When I don’t play with my little brother, then she scolds me.” (FB, 9)
– “It’s good. I sweep and do all the work.” (FK, 10)
– “I am my mother’s daughter.” (FB, 11)
– “I like working with her in the kitchen.” (FB, 11)
9 Examining the Circle of Attachment Trauma, Shame, and Marginalization… 139

– “Mum likes it when I work in the house.” (FB, 11)

– “I like sitting and talking with mother (about) household gossip.” (FL, 13)
– “When papa tells me not to come to school and mummy sends me—I like
that.” (FB, 12)
• How would you describe your relationship with your father?
– “It’s good, I don’t know more.” (FB, 11)
– “Papa keeps on saying that girls can’t do anything … he doesn’t understand.
Papa doesn’t believe that a girl can do something. I want him to realize that
we can.” (FB, 13)
– “Whatever I say, my father gets it for me.” (FK, 12)
• In what ways would you like to be like your mother?
– “My mother never wishes ill for others, listens to others, keeps me with her
and keeps her pain to herself, doesn’t tell anyone that I have this problem.”
(FB, 10)
– “Yes, but I would not want to remain a housewife like her.” (FB, 12)
– “Yes, I don’t know.” (FB, 10)
• Do you feel loved by your parents?
– “I have never felt this with papa, but with my mother yes,…whenever she
talks so—I feel she wants to push me out of the house…that she doesn’t want
me there.” (FB, 15)
• Do your parents fight? How does that make you feel?
– “Never.” (FK, 11)
– “I don’t know about that.” (FB, 10)
– “When father shouts at mother, I feel very scared.” (FB, 11)
– “Yes my father doesn’t listen to my mother. He drinks and there are many
other things that he doesn’t listen to what mummy says to him—so they fight.”
(FB, 15)
• Do you have any wishes?
– “No.” (FL, 10)
– “I don’t know.” (FL, 12)
– “I want to be a madam (memsahib, which literally means an English dame).”
(FB, 12)
– “I want to be a teacher, pilot, and scientist.” (FB, 10)
– “I want to be an inspector. The rest is up to God. I want to be someone, let’s
see where my destiny takes me. Because my parents want me (to become)
married. I want to do any work—though I want to be an inspector—but actu-
ally I will take up any work as long as there is some work available. I want to
be someone.” (FB, 15)
– “I want to study till 10th grade.” (FK, 12)
– “If my father lets me study, then I would like to do a job.” (FL, 12)
140 M. Kumar

Common patterns and trends that emerge from the case vignettes described
above are:
1. Paucity of material/discourse—reflecting a reality of life?—or an issue of the
ease and agency in qualitative research?
2. Absent gazes, prolonged silences, smiles, and monosyllables
3. Difficulty in putting very basic everyday activities or problems into words or a
comprehensible narrative (poor literacy/educational background, verbal facility)
4. Lack of imagination—inability to garner enough material to express needs,
thoughts, desires, or wishes
5. Idealized responses—Ferenczi’s (1988) ‘wise baby’—traumatized child—par-
ticularly on questions about parents
6. Psychic equivalence (Fonagy et al. 2002)—“symbolic equation” (Segal 1988)

9.5 Discussion: Explorations into the Aporias and Absences,

Re-examining Subjectivities

The patterns of response here suggest some sort of deficit in thinking and emoting.
Both of these capacities are compromised in situations of high stress or amidst
continual negligence and dismissal of personal needs.

9.5.1 Particular Quality to the Narratives of Absence

If the interviews are seen as ‘forms of self-telling,’ then at least two formations of
self can be discerned from the material above. Freud’s psychic reality was not so
much about how life was, but rather how it was recounted and interpreted to be.
So the stories here also provided fodder to look beneath the mask and see the layered
nature of these young girls’ psychic reality. Two response patterns could be
discerned through an attempt to differentiate narrative modes of thought based on
narrative discourse.

9.5.1.1 Embalming of Absence and Invisibility

This narrative response was mentally and bodily connected to shame—concreteness

and invisibility—compelling mental numbness. Shame, like any self-conscious
emotion, requires an organism’s own sensitivity, is related to external appearance,
and is directed externally. Shame becomes a heightened consciousness of self, an
unusual and distinct form of self-perception. The self is seen as small, helpless,
frozen, and emotionally hurt (Lewis 2003). Such a self-perception or internal attri-
bution often emanates from harsh socialization or when there exists a high degree of
punishment for failure (or reward for success). The experience of shame makes the
9 Examining the Circle of Attachment Trauma, Shame, and Marginalization… 141

self both the subject and the object: the self is embroiled in a battle with itself only.
Research also shows that there tends to be less sharing when people feel shame
acutely (Shweder 2003; Rimé 2009).
Another quality of their narratives was the concreteness of metaphors (Grubrich-
Simitis 1984), in that an open-ended quality of fantasy life or imagination was
absent. Fixed gazes and expressions carry an object-like, unalterable quality.
“Metaphorization” and “mentalization” are possessions of the ego, and when these
are impaired it speaks about the functioning of the ego. Their absence then is also
colored with ‘real’ mental capability deprivation.

9.5.1.2 Daunting Presence of Dismissive Defensive Strategy

It is important to differentiate here that the latter is different from dismissive

attachment. It is not that the children were dismissive of the relationships/parents, but
that they were dismissive of their own neediness, desires, and the disturbing things
that were happening around them. The style of functioning could be dismissive—not
in terms of undermining the relationships, but in dismissing the need to be cared for
and the feelings of loss associated with it. This is what the sample showed. There
seems to be a precocious ‘wise-baby’ who deflects the emotional life to a place where
it cannot be felt (Winnicott 1970; Ferenczi 1988) and takes on a maternal function in
an inverted way. The loss of mental states (Lear 2007), the lack of openness, and a
certain solipsism are also about problems in this relational matrix, where parenting
and socialization experiences have been hurtful, insensitive, frightening, etc.

9.5.2 Is Inhibition a Symptom or Different from a Symptom?

The responses above also allude to restrictiveness, constraint in the ability to express
and share stories of their lives. This may be due to either a paucity of vital experi-
ence or a marring of the ability to make sense and engage with one’s experiences.
It may also be due to both of these factors.
One text that engages with this restricted ability is Freud’s (1926) “Inhibitions,
symptoms, and anxiety,” where he began by discussing how inhibition is different
from a symptom (or symptom formation). According to Freud, inhibition has a spe-
cial relation to function, where it is a normal restriction of a function and at times
may be closer to being a symptom and carry a pathological dimension. Inhibitions
akin to symptoms are products of disturbed functioning of the ego. Apart from
disturbances in the ego, functioning due to sexual or nutritional needs, ego strength
is known to get disrupted when in conflict with the demands of either the super-ego
or the id. Freud (1926) traced the generalized inhibitions of the ego to impositions
on the ego as a measure of precaution, or as brought about as a result of depletion in
energy. He was emphatic to point out that, unlike a symptom, inhibitions take place
and act upon the ego itself. This might also mean that inhibitions are closer to the
142 M. Kumar

conscious (or belong to the domain of the pre-conscious) than symptoms per se.
In this sense one’s socialization and familial surroundings might play a more active
role in causing or exacerbating such restrictions in the sense of self. It seemed that
these young girls were fighting their own selves and inner persecutory parents/
family in this process.
Freud (1926) conceptualized shame as an inhibiting emotion that led to repres-
sion, therefore giving it a central importance in the development and maintenance
of psychopathology. In a bid to solidify his drive model (eros and thanatos as “life”
and “death” drives), he underplayed the role that shame/inhibition plays in repression
(Scheff 2000). As his model became more and more intrinsically driven, shame
ceased to occupy a place in the literature.

9.5.3 Domesticity and Docility—(En)gendered Identifications

It is not in doubt that the norms and practices that affect adults also affect children;
cultural practices inhibiting women’s movement and access to resources are
examples. The responses given by these girls depicted the burgeoning unpaid work
burden and in many ways their narratives are also ways in which culture and its
practices transmit themselves. What might vary is the cultural and linguistic per-
spective or narrative form in which it is formulated and expressed (Bruner 2004).
Whether it was paid or unpaid domestic work or work outside the house it was clear
that work was a critical factor in the lives of these children. The largest inequality
took place in household work. The interviews made the enormous burden of work
on these girls quite explicit, apart from highlighting the rewards/punishments that
are associated with it. Love, care, and acceptance were bartered for work. The
excessive exposure to household work in many ways obstructs other potential func-
tions, such as knowledge seeking, education, hygiene, health, leisure activities, etc.
It was quite explicit from the children’s narratives that the gendered social norms
and traditions gave preference to boys and prioritized education for boys over girls.
It leads to the question of whether being a girl was something shameful or at least
culturally devalued. Being respected and treated with dignity also belong to the
domain of well-being, and prolonged discrimination weakens ego strength, trust,
and a sense of security that is crucial for secure attachments.

9.6 Intergenerational Import and Transmission

of Attachment, Trauma and Poverty

The Kutchi girls that I interviewed seemed to face a continual disenfranchisement

of their voices, needs, and desires. Attachment trauma in the young girls of rural
Gujarat was precisely this inability and failure of their families to adequately nur-
ture psychological and social capabilities in the girl children, and this trauma was
9 Examining the Circle of Attachment Trauma, Shame, and Marginalization… 143

transmitted intergenerationally. A few implications of the points described above

need to be made clear:
• The literature on both poverty and trauma points out that these two themes are
multidimensional and intergenerational in their import. Psychologists studying
massive trauma events such as the Nazi atrocities against the European Jews during
World War II, describe how the third and fourth generations after the original
survivors continue to suffer from pain, guilt, and loss. Similarly development
economists have shown how poverty is a trap in which people live in poverty, die
in poverty, and pass it on to subsequent generations. Attachment also is intergen-
erational in that, if not dealt with, insecurities keep repeating and the same rela-
tional patterns pass them on from one generation to the next. This intergenerational
cycle of the transfer of poverty/trauma/attachment needs to be studied altogether
and in-depth.
• Considering that poverty is understood today as capability deprivation, attach-
ment at least as described in this paper is offered as a concept that denotes basic
human well-being or a need, and one that is akin to a capability that family/
caretaker/society nurture in an individual. As a capability it endows an individual
with a sense of security, safety, and relational contentment that are essential for a
healthy sense of self, identity, and purpose in life. Attachment trauma of the kinds
that these Kutchi girls exhibited suggests a life where shame and self-doubt were
prominent emotions. So acute was the feeling of shame that it annihilated the self
into a complete dismissal of personal needs—right from the need to be loved, to
be cared for, to the need for autonomy and identity. Development research could
potentially enrich the discourse on poverty by looking at attachment trauma as a
capability deprivation akin to how poverty and social marginalization are under-
stood today. A young child’s mind and its capacities are marred in an environment
that generates constant conflict between needs and discrimination, and along with
the lack of love, care, or warmth, takes away any possibility of viewing oneself as
a capable human being.
• Shame has not been examined adequately in either psychological or other social-
science literature. Amongst the few contributions made to our understanding of
this emotion is the work of Cooley (1992), who talked of shame as a social emo-
tion. In his concept of the “looking-glass self” Cooley talked about the process of
self-monitoring, where the processes involved are: the imagination of our appear-
ance to others, the imagination of their judgment of that appearance, and some
sort of self feeling—like pride or mortification. In the young girls, a deep sense of
inferiority developed due to continual discrimination and disengagement that is
also partly mirrored by their mothers (or other female figures) who likewise are
not necessarily able to reflect a sense of identity and independence. The “looking-
glass self” then only senses lack, inadequacy, and absence in one’s self, and that
is what these interviews captured. Something that these interviews conveyed was
that the change had already happened—the self was absent and silence echoed
around these young girls of Kutch. Shame is isolating, it is a disconnection from
social bonds because it is also an evaluative emotion unlike guilt, which is an
144 M. Kumar

individualistic emotion (Chodorow 1996). Lewis (1971) further explains shame

as a social emotion—where a social bond is threatened, calling it a “bodily and/or
mental response to the threat of disconnection from the other.” In this sense the
girls in my study were in such strong identification with the authority that mini-
mized their freedom and existence that any thinking or reasoning that differed
caused them great shame or discomfort.
Nurturing attachment also implies nurturing or developing a certain ‘intelligibility,’
which is a combination of sensitivities and capabilities in the parents’ mind about the
larger concerns for their child. Research evidence (Harper et al. 2003; Robeyns 2003;
Baneerjee and Duflo 2012) shows that sensitization and empowerment of women
enables them to act on their general preferences in making decisions that are in the
best interests of their children’s well-being. The transmission can be positive (such as
resilience). What is important is how and whether the real and perceived negative
effects can be overcome over the course of a lifetime and/or between generations,
and if not, what is it that prevents these outcomes.
The capabilities approach developed by Sen (1982) insists that the focus should
be on the real freedoms that people have for leading a fulfilling and meaningful life.
This approach refuses to make normative judgments based exclusively on income,
commodities, or material resources. Capabilities are considered people’s potential
function or their ability to exist and act. Thus as a means of enhancing the well-
being of its people, a society also needs to focus on the freedom of individuals to
achieve their intrinsic capabilities. Attachment can be seen as a function of a per-
son’s capabilities that are intergenerationally nurtured and passed on. Psychologists
think that children are endowed with the need to seek secure attachments, though at
the end of the day it is also the family and social milieu that instill and reinforce
attachment security. So there exists a dual process of initiation of the capability for
secure attachment—one which a human being is programmed to strive towards (like
love, trust, empathy, etc.), and on the other hand it is also to be harnessed and devel-
oped over the course of one’s lifetime through socialization. It is in this domain that
these young girls were contesting their struggle for survival, identity, and freedom
to achieve what they desire.

9.7 Post Disaster Adversities: A Coda

One measure of the long-term damage post disasters is to examine personal and
social adversities in the lives of survivors. The severity of post-event adversities
then is one index of ongoing and often compounded traumatization. Medical
anthropologists Kleinman (1995) and Young (1995) have argued that the actual
appreciation of the influence of social conditions has been compromised due to the
extreme popularity of post-traumatic stress disorder (PTSD) as a psychiatric/medical
diagnostic category and the consequent “pathologization” of the person as a whole
(see Caruth (1996) review of PTSD for a critical commentary on this theme).
9 Examining the Circle of Attachment Trauma, Shame, and Marginalization… 145

This tendency towards “psychologization” undermines the role of structural forces,

and how institutions and social systems control and exert tremendous influence on
individual lives (Scheper-Hughes 1992; Leys 2000; Parker 2007).
That life after trauma is more painful and difficult to lead than the traumatic
event itself is well known to psychologists. But in instances where individual or
communal trauma is made more potent by a breakdown of socio-economic or politi-
cal security, or by a lack of access to basic entitlements, rights, and privileges, the
psychological damage is manifold (Nieuwenhuys 1990; Scheper-Hughes 1992;
Erikson 1995; Young 1995; Kaplan 2005). Essential elements for recovery from
such experiences include:
• a healthy family environment where one is able to communicate their needs
(since every other need becomes subservient to the struggle for survival)
• the ability to empathize with the pain and suffering of others (without meaning
to underrate the quality of care that poor or materially less able parents may
provide to their children or the daily survival takes a toll on time, energy, and
resources that can be channeled towards a child’s well-being)
• the ability to think about moral and ethical choices associated with decision making
on family issues, such as forcing a child to drop out of school in order to work.
When the secure socio-cultural field around which the family is supposed to
function collapses, the adults and children in the family may experience an
enormous sense of loss and disorientation. Therefore along with the morality
that emanates from internal family structure, a social structure is needed to
enforce ethical and child-rights oriented decision making on behalf of parents
and caretaking adults.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Ainsworth MDS, Blehar MC, Waters E, Wall S (1978) Patterns of attachment: a psychological
study of the strange situation. Erlbaum, Hillsdale
Baneerjee AV, Duflo E (2012) Poor economics: a radical rethinking of the way to fight global pov-
erty. Random House, New Delhi
Berry JW, Poortinga YH, Pandey J (eds) (1997) Handbook of cultural psychology. Allyn and
Bacon, Boston
Bowlby J (1969) Attachment and loss: vol 1. Attachment. Basic Books, New York
Bruner J (1990) The narrative construction of reality. Crit Inq 18:1–21
Bruner J (2004) Life as narrative. Soc Res 71(3):691–710
Caruth C (1996) Unclaimed experience: trauma and narrative. John Hopkins University Press,
Baltimore
Chodorow NJ (1996) Theoretical gender and clinical gender: epistemological reflections on the
psychology of women. J Am Psychoanal Assoc 44S:215–238
Cooley CH (1992) Human nature and the social order. Transaction Publishers, Piscataway Township
146 M. Kumar

Erikson K (1995) Notes on trauma and community. In: Caruth C (ed) Trauma: explorations in
memory. John Hopkins Press, Baltimore, pp 183–199
Ferenczi S (1988) Confusions of tongue between adults and the child: the language of tenderness
and of passion. Contemp Psychoanal 21:196–216
Finkenauer C, Rimé B (1998) Socially shared emotional experiences vs. emotional experiences
kept secret: differential characteristics and consequences. J Soc Clin Psychol 17:295–318
Fonagy P (2001) Attachment theory and psychoanalysis. Other Press, New York
Fonagy P (2005) Attachment, trauma, and psychoanalysis: where psychoanalysis meets neuroscience.
Keynote lecture at the IPA 44th Congress on Trauma: New Developments in Psychoanalysis,
International Psychoanalytic Association, Rio De Janeiro, 28–31 July 2005
Fonagy P, György G, Jurist EL, Target M (2002) Affect regulation, mentalization and the development
of the self. Other Press, New York
Freud S (1926) Inhibitions, symptoms and anxiety. Standard edition of the collected works of
Sigmund Freud, vol 20. Hogarth Press, London
Glaser BG, Strauss A (1967) Discovery of grounded theory: strategies for qualitative research.
Transaction Publishers, Piscataway Township
Green A (2000) The intrapsychic and intersubjective in psychoanalysis. Psychoanal Q 69:1–39
Grubrich-Simitis I (1984) From concretism to metaphor—thoughts on some theoretical and
technical aspects of the psychoanalytic work with children of Holocaust survivors. Psychoanal
Study Child 39:301–319
Harlow H (1959) The nature of love. Am Psychol 13:573–685
Harper C, Marcus R, Moore K (2003) Enduring poverty and the conditions of childhood: life
course and intergenerational poverty transmissions. World Dev 31(3):535–554
Kaplan S (2005) Children in Africa with experiences of massive trauma: a research review. SIDA mono-
graph, Swedish International Development Cooperation Agency, Stockholm. http://www.multietn.
uu.se/hgs/staff_hgs/pers _homepages_hgs/kaplan_SIDA4629en_ Children_ in_Africa_w.pdf.
Accessed 20 Oct 2012
Kleinman A (1995) Writing at the margin: discourse between anthropology and medicine.
University of California Press, London/Berkeley/Los Angeles
Lear J (2007) Working through the end of civilization. Int J Psychoanal 88:291–308
Lewis HB (1971) Shame and guilt in neurosis. International University Press, New York
Lewis M (2003) The role of self in shame. Soc Res 70(4):1181–1204
Leys R (2000) Trauma: a geneology. University of Chicago Press, Chicago/London
Moscovici S (1984) The phenomenon of social representations. In: Farr M, Moscovici S (eds)
Social representations. Cambridge University Press, Cambridge
Nieuwenhuys O (1990) Children’s lifeworlds: gender, welfare and labour in the developing world.
Routledge, London
Nussbaum M (2004) Body of the nation. Boston review of literature. http://bostonreview.net/
BR29.3/nussbaum.html. Accessed 25 Oct 2012
Parker I (2004) Qualitative psychology: introducing radical research. Open University Press,
London
Parker I (2007) Revolution in psychology: alienation to emancipation. Pluto Press, London
Rimé B (2009) Emotion elicits the social sharing of emotions: theory and empirical review.
Emot Rev 1:60–85
Robeyns I (2003) Sen’s capability approach and gender inequality. Fem Econ 9(2–3):61–92
Rosaldo MZ (1984) Toward an anthropology of self and feeling. In: Shweder RA, LeVine RA (eds)
Culture theory: essays on mind, self, and emotion. Cambridge University Press, Cambridge
Scheff TA (2000) Shame and the social bond: a sociological theory. Sociol Theory 18(1):84–99
Scheper-Hughes N (1992) Death without weeping: the violence of everyday life in Brazil.
University of California Press, Berkeley
Segal H (1988) Introduction to the works of Melanie Klein. Karnac Books, London
Sen A (1982) Choice, welfare and measurement. Blackwell, Oxford
Shmueli-Goetz Y, Target M, Fonagy P, Datta A (2008) The child attachment interview: a psycho-
metric study of reliability and discriminant validity. Dev Psychol 44(4):939–956
9 Examining the Circle of Attachment Trauma, Shame, and Marginalization… 147

Shweder RA (1991) Thinking through cultures. Harvard University Press, Boston

Shweder RA (2003) Towards a deep psychology of shame. Soc Res 70(4):1109–1129
Shweder RA, Levine RA (eds) (1984) Culture theory: essays on mind, self, and emotion. Cambridge
University Press, New York
Sroufe LA, Waters E (1976) The ontogenesis of smiling and laughter: a perspective on the organi-
zation of development in infancy. Psychol Rev 83:173–189
Target M, Fonagy P, Shmueli-Goetz Y (2003) Attachment representations in school-age children:
the development of the child attachment interview (CAI). J Child Psychother 29:171–186
Thompson RA, Raikes HA (2003) Toward the next quarter-century: conceptual and methodological
challenges for attachment theory. Dev Psychopathol 15:691–718
Winnicott DW (1970) The maturational processes and the facilitating environment. Int Psychoanal
Press 64:1–276
Young A (1995) The harmony of illusions: inventing post-traumatic stress disorder. Princeton
University Press, Princeton
 Part III
Environmental Drivers of Marginality
Chapter 10
Poverty, Agriculture and the Environment:
The Case of Sub-Saharan Africa

Prabhu Pingali, Kate Schneider, and Monika Zurek

Abstract Marginal areas of sub-Saharan Africa (SSA) have historically offered low
productivity potential and low returns on investments in agricultural productivity
growth. Population and agricultural market dynamics in Africa are improving the
prospects for productivity-enhancing investments in this environment. In this chapter
the authors introduce an opportunity cost framework to demonstrate where agricultural
development is now an opportune strategy to reduce marginality in SSA and to guide
strategic priority setting for public investment for the sustainable improvement of
agricultural productivity. It then lays out policy and technology priorities for sustainable
development of marginal production environments.

Keywords Opportunity costs • Priority setting • Investments • Yield gaps • Agricultural

productivity growth

10.1 Introduction

Poverty in Sub-Saharan Africa (SSA) remains predominantly a rural phenomenon

and the vast majority of the rural poor are smallholder farmers (World Bank 2007).
Many of those farmers are concentrated in areas that have remained unattractive for
agricultural development and as a result, demonstrate extremely low agricultural
productivity. Such marginal areas have historically offered little return on investment
due to low population density or poor access to markets and thus remain marginal
precisely because investments to alter their condition have been unfavorable to the
public and private sectors (including the farmers themselves), and have not been
made. However, as population density increases and market dynamics open, new
opportunities and the potential for attractive returns on investment in agricultural

P. Pingali (*) • K. Schneider • M. Zurek

Bill and Melinda Gates Foundation, Seattle, WA, USA
e-mail: [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 151
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_10,
© The Author(s) 2014
152 P. Pingali et al.

development are becoming increasingly promising. In this chapter we introduce an

opportunity cost framework to demonstrate where agricultural development is now
an opportune strategy to reduce marginality in SSA and to guide strategic priority
setting for public investment to sustainably improve agricultural productivity and
thereby reduce marginality.
Marginal environments are areas that have been relatively unfavorable for agricul-
tural production due to one or more socioeconomic, technological, or biophysical
constraints. Farmers may face constraining factors in the agro-ecological environment
(agro-ecological zone or farming system), environmental resource availability,
remoteness (distance to markets and services, transportation or communication
infrastructure), market access, socioeconomic resources, or an unfriendly policy
environment. Some marginal environments are characterized by unfavorable conditions
in several of these domains, while others may be considered marginal due to more
limited sets of factors (Lipper et al. 2007). Low population density and poor market
access limit the attractiveness of investments to enhance agricultural productivity.
Today, as population density is increasing and market demand and infrastructure are
improving, investments in intensification are becoming increasingly attractive in
areas previously considered to be unfavorable.
Many of SSA’s marginal areas, while becoming increasingly favorable for investments
to enhance productivity, also pose unique challenges, with important technology
and policy implications for ensuring that productivity growth is both pro-poor and
sustainable. Poor farmers depend in large part on ‘orphan crops’ such as sorghum and
cassava, and face constraints due to poor natural resource endowments and an already
degraded resource base. Therefore, effectively reducing marginality through agricul-
tural productivity growth in SSA requires a technology focus on the crops and traits
that are important to the poor. Similarly, increasing population density and urbanization
pose both an opportunity for agricultural development as well as a constraint, as they
also increase pressure on agriculture and the natural resource base. These pressures
are further exacerbated by existing environmental degradation and the impacts of
climate change. Therefore agricultural development strategies must focus on sustain-
able intensification. Policy can help to foster the sustainability of productivity growth
by addressing the environmental constraints particular to marginal areas and by
providing the proper incentives for natural resource use that help reduce or reverse
degradation. At the same time, some areas will remain unfavorable for agriculture,
such as extremely arid, steep sloped or rocky areas. Policy intervention is also necessary
to help keep such areas out of agricultural production and find alternative uses, such
as providing ecosystem services.

10.2 Poverty and Marginal Environments

in Sub-Saharan Africa

Marginality with respect to agriculture has been conceptualized as areas that are less
favorable for agriculture due to both biophysical and socioeconomic factors. These
areas are regions that demonstrate low potential for agricultural production due to
10 Poverty, Agriculture and the Environment: The Case of Sub-Saharan Africa 153

fragile agricultural resource bases and/or limited access. Most less-favorable areas
are uplands, highlands, arid or semi-arid zones where low agricultural productivity
results from a variety of constraining environmental factors including limited soil
fertility, prohibitive slopes, or unfavorable temperature or hydrological conditions
with economic factors such as poor infrastructure or market access. Appropriate
production strategies in these areas include perennial and tree crops, rotating culti-
vation systems, and mixed cropping. The constraints of limited natural endowments
are aggravated to begin with by increasing degradation trends, such as soil erosion
and compaction, water stress, over-grazing, deforestation, and desertification in the
drylands (Ruben et al. 2007).
Quantifying the extent of land area and the size of populations found in mar-
ginal areas is challenging, as existing information on these subjects and on local
limiting factors are not mutually exclusive. However, looking at the above charac-
teristics in turn—resource endowment and degradation—provides some basic
context on marginal lands.
Resource Endowment: Nearly half (43 %) of the African continent is characterized
as dryland and therefore environmentally constrained in agro-ecological terms.
Drylands include arid, semi-arid, and dry sub-humid agro-ecological zones, as
defined by their average growing season length (UNDP 2009). The vast majority of
drylands—about 72 %—are in developing countries, where they are home to 13 %
of the developing world’s population (Millennium Ecosystem Assessment 2005).
In SSA the predominant farming systems in the drylands include migratory live-
stock herding, agro-pastoral systems, and rainfed mixed cropping (Dixon et al.
2001; Ruben et al. 2007; UNDP 2009). Drought and water stress are the predomi-
nant environmental constraints, exacerbated in many areas by land degradation
(Adhikari 2011).
Land Degradation: Degraded lands may be defined as areas demonstrating long-
term decline in ecosystem function, one aspect of which can measured in terms of
declining net primary productivity1 over time (Bai et al. 2008). As of 2003 more
than 188 million people in SSA depended directly on degraded lands for their liveli-
hoods and worldwide agricultural lands are substantially over-represented in their
share of degraded land area (Bai et al. 2008 and authors’ calculations). Specifically
among the drylands, degradation associated with agriculture is slightly more prevalent
in the dry sub-humid and semi-arid areas than in the arid regions (Adhikari 2011).
Estimates of agricultural productivity loss due to land degradation vary greatly and
most are based on expert opinion or assumptions, with few rigorous experimental
efforts (Ruben et al. 2007). While numerous studies have found high levels of soil
erosion and soil nutrient depletion in several African countries and particularly in
the drylands (see Ruben et al. 2007), many have criticized the methods and accuracy
of these efforts, leaving little agreement over the extent of the productivity gap
resulting from land degradation (Ruben et al. 2007).

1
The net rate at which vegetation fixes carbon dioxide from the environment, estimated by remotely
sensed proxy indicators.
154 P. Pingali et al.

10.2.1 Poverty and Marginality

Poverty coincides with marginal environments across SSA. Evidence across nations
demonstrates that not only are the poor mainly found in rural areas where agriculture
is the main source of income, but also that poverty rates are higher in areas that
can be classified as marginal for agricultural production (authors’ calculations;
HarvestChoice 2010; FAO 2013). In terms of absolute numbers, more people live
in agriculturally favorable environments, so the share of the rural poor living in
marginal environments is relatively smaller. However, there is debate regarding the
relative incidence and severity of poverty in different types of rural environments
and data on the geographic distribution of poverty is extremely limited (Ruben et al.
2007). This debate stems in part from different conceptualizations of marginality as
discussed above, though studies in Africa have consistently found that poverty prev-
alence and severity are often greater in areas less-favored for both environmental
and economic reasons across the continent (Ruben et al. 2007).
The relationship between poverty and marginal agricultural environments is even
more pronounced for the ultra-poor—those living on less than US$0.75/day—who
are often found in semi-arid and arid environments. The ultra-poor are a significant
share of the total poor in SSA, approximately 136 million people nearly all of whom
(93 %) are farmers (HarvestChoice 2010). Globally, the ultra-poor are becoming
ever more concentrated in SSA and South Asia, and are overrepresented in marginal
areas (Gatzweiler et al. 2011). Figure 10.1 presents the example of Ghana, where
the prevalence of poverty and ultra-poverty in rural areas is highest in the sub-humid
and driest regions of the country. Similar analyses have found this trend to be consis-
tent across many countries (HarvestChoice 2010).

10.2.2 Agriculture Is an Engine of Growth and Poverty

Reduction

Agriculture continues to be an important source of income for rural households across

all income levels. In fact a substantial body of evidence supports the relationship
between agricultural productivity growth and poverty reduction, demonstrating gen-
erally high poverty reduction elasticity for agricultural productivity growth (Hazell
2010; Pingali 2010). Worldwide agricultural growth has been consistently shown to
be more effective in reducing poverty than comparable growth in other economic
sectors; for instance, on average a 1 % increase in the agricultural growth rate has
been estimated to reduce poverty by 1.6 % more than equal growth in industrial sectors
and by three times more than in service sectors (Christiaensen and Demery 2007).
In SSA specifically, investment in agriculture contributes 4.25 times more
towards reducing poverty than comparable investments in any other sector (Pingali
2010). Furthermore, agriculturally driven growth generates a larger welfare effect
than non-agriculturally driven growth, especially for the poorest 20 % of the popu-
lation (World Bank 2007). Agricultural productivity growth is thus an important
 10

Fig. 10.1 Poverty prevalence by agro-ecological zone in Ghana (HarvestChoice 2010)

Poverty, Agriculture and the Environment: The Case of Sub-Saharan Africa
155
156 P. Pingali et al.

lever for poverty and hunger reduction, however, the policy environment must be
favorable for intensification in order to incentivize farmers to adopt technologies
and practices that increase their productivity (Johnston and Mellor 1961; Lipton
2005; World Bank 2007).

10.3 Agricultural Development Strategies

in Marginal Areas

Technological advancements and agricultural development interventions can reduce

marginality by addressing the contributing environmental and economic factors
where the potential for returns on investments in productivity growth is favorable.
Market access and population density largely determine where productivity enhancing
technologies and investments are attractive to both farmers and external investors
(public and private).
Population density largely determines the relative costs of land and labor, which
provides a framework for understanding where intensification is favorable to farmers
and what strategies—labor- or land-saving—are likely to be pursued (Pingali 2001).
Additional strategies, including both technological and policy interventions, are
necessary to improve the sustainability of approaches to productivity growth (Lipper
et al. 2007). In Fig. 10.2 we present the opportunity costs of land and labor within
a framework for understanding the evolution of production and intensification
strategies. We further employed this framework to articulate the appropriate strate-
gies for productivity growth and sustainability to reduce marginality and improve
agricultural performance in SSA.
As population densities rise and the opportunity cost of land increases, farmers
have more incentive to increase productivity through intensification strategies that
increase yield, such as using more or improved inputs and/or increasing the intensity
of cultivation practices. Multiple examples throughout history demonstrate that
population pressure has driven shifts from traditional rotational fallow systems to
annual and multi-crop cultivation systems (Boserup 1965). Similarly, as the opportunity
cost of labor increases, farmers have a growing incentive to employ labor-saving
technologies and practices. Market opportunities are important factors in determining
the relative scarcity of labor, as employment alternatives increase wage rates and
decrease the supply of labor. Factors such as trade integration (as it impacts prices
and access to markets), availability of off-farm employment, and overall economic
growth also influence the opportunity cost of agricultural labor, the incentive to
intensify, and the ability to do so via labor-related savings. These aforementioned
labor market dynamics often accompany rising population density, though not
always. Intensification may also be found even where population density remains
low, if market access is favorable or unique market opportunities (such as niche or
quality markets) exist (Pingali 2001).
10 Poverty, Agriculture and the Environment: The Case of Sub-Saharan Africa 157

Farming System: Mechanized large- Farming System: High value commercial

scale cereal production systems (e.g., agriculture (e.g., Israel)
Brazil)
Productivity Strategy: Reducing unit Productivity Strategy: Food safety and
cost of production quality, value added premium (e.g., labels)
High
Environment Strategy: High potential Environment Strategy: High competition
for payment based ecosystem between food production and other
services schemes (e.g., carbon ecosystem services, need to find win-win
Opportunity cost of LABOR

market mechanisms, payments for solutions (e.g., organic, high-value

and Economic Growth
Trade Integration,

ecosystem services) products)

Farming System: Traditional low Farming System: Intensive cereal crop

productivity subsistence systems systems (e.g., Punjab, India; potentially
(e.g., marginal environments in SSA) higher quality areas of SSA)
Productivity Strategy: Increasing and Productivity Strategy: Input efficiency and
stabilizing local food supplies diversification
Low

Environment Strategy: win-win Environment Strategy: Productivity

technologies crucial (e.g., integrated tradeoffs for environmental conservation
soil fertility management, and provision of ecosystem services (e.g.,
intercropping) long fallow periods, organic, conservation
agriculture)

Low High
Opportunity cost of LAND
Marginal Favorable

Fig. 10.2 Agricultural development strategies by opportunity cost of land and labor

10.3.1 Agriculture-Environment Interactions

in Marginal Areas

Tradeoffs between agriculture and natural resource needs in the environment can
be severe with increasing pressure on land, potentially allowing short-term
resource availability or productivity gains while undermining the long-term security
of ecosystem function (Hazell and Wood 2008; Barbier 2010). Evidence has shown
that unsustainable land use and degradation are more likely where population
growth occurs rapidly without sufficient simultaneous agricultural intensification
relative to that growth (Pingali 2001; Ruben and Kuyvenhoven 2003; Hazell
and Wood 2008). While much population growth is taking place in areas that are
favorable for agriculture, many marginal environments are witnessing similar
trends. For instance, in dryland areas human population growth increased 18.5 %
globally between 1990 and 2000, and experts have noted that degradation demonstrates
one of the greatest threats to both agricultural productivity and poverty reduction
across dryland areas worldwide (Reardon et al. 2002; Millennium Ecosystem
Assessment 2005; Adhikari 2011).
Degradation is likely to be most severe where the returns on investments in land
improvement are lowest and in regions with a relatively large proportion of marginal
lands (Pingali 2001). Intensive cultivation without the addition of capital and inputs
to conserve and replace natural resources and nutrients degrades the natural resource
158 P. Pingali et al.

base. As resource degradation worsens pressure on rural households increases, and

this may deepen poverty and thereby undermine the poverty reduction potential of
agriculturally driven growth strategies (Hazell and Wood 2008).
Policy plays a critical role in ensuring that an enabling environment for inten-
sification helps incentivize more sustainable approaches. Effective property rights
are a necessary foundation. Where farmers rights to their land are insecure they are
likely to opt for practices that yield the greatest benefits in the short-term, leading
to underinvestment in land improvements that preserve the natural resource base for
long-term productivity (Pingali 2001).

10.4 Agricultural Productivity in Sub-Saharan Africa:

Towards a ‘Green Revolution 2.0’

While the second half of the twentieth century was characterized by the Green
Revolution (GR) that generated widespread productivity growth and affiliated
poverty reduction that lifted millions out of poverty in Asia and Latin America,
Africa was largely left out (Pingali 2012). Crop yield growth over the period from
1961 to 2010 was slower in SSA than the global average for nearly all crops, with
exceptions only for a few crops (wheat and soybeans) of which Africa’s share
of global production is negligible (Pardey et al. 2012). Many areas of SSA have
abundant agricultural potential in terms of environmental resource availability,
labor, and knowledge, yet productivity has remained extremely low.
The GR package of innovations was most pertinent to areas of high land scarcity
that were amenable to rapid intensification. These were densely populated areas
with high rainfall or irrigated areas suitable for the major cereal crops: wheat, rice,
and maize. That strategy was not appropriate for Africa where population densities
were low, market infrastructure was weak, and the poor depended largely on ‘orphan’
crops, with little history of crop improvement research (Binswanger and Pingali
1988; Evenson and Gollin 2003; Webb 2009; Pingali 2012). However, the situation
is quite different today, with rising population densities and increasing demand as
emphasized already, some areas in Africa have land/labor ratios similar to those in
Asia at the start of the GR. This suggests that the incentive to intensify land use is
rising and it is already becoming evident that Africa is finally beginning to experi-
ence a GR “2.0” (Otsuka and Kijima 2010; Pingali 2012).
For SSA, intensification is thus a promising strategy to reduce marginality where
rising population densities and improving market access and opportunities offer
substantial returns. In these areas agricultural development must prioritize removing
the key constraints that contribute to their environmental and economic marginality,
including basic crop improvement for the ‘orphan crops’ that have little history of
agricultural research and development, management strategies appropriate to resource-
limited environments, and stress tolerance. Addressing agricultural-environmental
tradeoffs must focus on leveraging win-win opportunities that do not compromise
productivity in the short- or long-term.
10 Poverty, Agriculture and the Environment: The Case of Sub-Saharan Africa 159

Top Crop

Millet
14

Sorghum
2nd Crop

Other Pulses
12
3rd Crop

10

Maize
Millions of Hectares

Cassava
8

Sorghum

Bananas/Plantains
Maize
Bananas/Plantains
6
Sorghum

Other Oilseeds

Cassava

Sorghum
4

Maize
Millet

0
Arid Semi-Arid Sub-Humid Humid Highlands

Fig. 10.3 Top ranked tropical crops in Sub-Saharan Africa by major agro-ecological zone
(HarvestChoice 2010)

10.4.1 Substantial Yield Gaps Remain

Despite potential and budding GR 2.0 momentum, substantial yield gaps remain in
SSA. Poor productivity is a multi-faceted problem resulting in large part from low
investment in research, low input usage, limited market access, poor policy and
regulatory environments, and environmental constraints. Large public investments
in agricultural research and development (R & D) were responsible for the GR’s
success in increasing productivity and drastically reducing poverty and hunger
across much of Asia (Hazell 2010; Herdt 2010). However, such investments have
dropped off dramatically since those made during the GR period (1966–1985).
What crops are important to the poor? Across all agro-ecological zones of SSA the
most important crops in terms of area harvested include: sorghum, millet, cassava,
maize, oilseeds, pulse beans, and bananas/plantains. Of this list only maize has a
history of substantial R & D investment in crop improvement, the others typically
fall into the category of ‘orphan crops’ that have historically received little R & D
attention. Figure 10.3 illustrates the most important crops for the five major
agro-ecological zones of SSA according to production area (HarvestChoice 2010).
Sorghum, millet, and cassava are prominent across zones, but more so in the marginal
areas, particularly dryland zones.
Quantifying yield gaps. Yield gaps represent the most economically feasible poten-
tial for improving yields based on currently available technologies, though new
technologies may make closing this yield gap easier or more feasible with fewer
inputs. The percentage of that yield gap that can be ‘closed’ with existing technologies
 160 P. Pingali et al.

6000
Baseline
Potential
Gap
5000
4000
yield (kg/ha)
3000
2000
1000
0

Arid Semi-Arid Sub-Humid Humid Tropical Highlands

Fig. 10.4 Yield gaps for sorghum in Sub-Saharan Africa (HarvestChoice 2010)

and under a certain set of conditions provides an estimate for the overall potential
for yield gains. Across SSA there is a substantial estimated yield gap for all of
the three top ranked crops in every major agro-ecological zone. For example, as
Figs. 10.4 and 10.5 illustrate, estimated yield gaps in sorghum and cassava are
512–73 % respectively (HarvestChoice 2010).
Yield gaps for the other top ranked crops in SSA are equally high. For millet the
estimated yield gap is 404 %, for bananas and plantains it is estimated to be 81 %.
“Other oilseeds” and “other pulses” presented in Fig. 10.4 represent aggregate
categories of minor crops that are estimated to have an average yield gap close to
300 %. Even for maize, the only crop ranked among the most important in SSA
that has benefitted from substantial investment in crop improvement, a significant
average yield gap of 287 % exists as illustrated in Fig. 10.6.

10.4.2 Technology Priorities

Productivity constraints and the resulting yield gaps in SSA demonstrate substantial
market failures in the provision of R & D and infrastructure investments, where neither
private firms nor national governments have sufficient incentive to invest. The case
for international public sector investment in basic science and agricultural R & D to
serve the poor is clear: neither the private sector nor individual country governments
have sufficient incentives to invest in the necessary upstream research to develop
10 Poverty, Agriculture and the Environment: The Case of Sub-Saharan Africa 161

30000
Baseline
Potential
Gap
25000
20000
yield (kg/ha)
15000
10000
5000
0

Arid Semi-Arid Sub-Humid Humid Tropical Highlands

Fig. 10.5 Yield gaps for cassava in Sub-Saharan Africa (HarvestChoice 2010)
6000

Baseline
Potential
Gap
5000
4000
yield (kg/ha)
3000
2000
1000
0

Arid Semi-Arid Sub-Humid Humid Tropical Highlands

Fig. 10.6 Yield gaps for maize in Sub-Saharan Africa (HarvestChoice 2010)
162 P. Pingali et al.

appropriate technologies for the crops and traits that are important to the poor. The poor
do not offer a value proposition to the private sector and many of the benefits will be
accrued by others outside of a particular country’s borders. Therefore an investment
gap exists for such international public goods. Just as it was necessary to spur the GR,
the international public sector, such as the CGIAR, is needed to close the yield gaps
in SSA (CGIAR 2000). Reducing marginality through agricultural development is
possible with public investment focused on basic R & D and improved delivery of
goods and services to farmers, and to better understand and respond to their needs.
Upstream basic science and crop improvement. First, upstream R & D is needed to fill
the gap in basic crop improvement for the ‘orphan crops.’ These are the most impor-
tant crops to the poor in marginal areas of SSA and basic crop improvement can have
substantial impacts on improving the favorability of drylands for agricultural produc-
tion. Targeting R & D to appropriate agro-ecological conditions will maximize the
spillover benefits to other areas and increase resilience to climatic shocks.
In addition to shifting the yield frontier, public R & D efforts ought to focus on
developing stress-tolerant crop varieties with resistance to both biotic (e.g., pests,
disease) and abiotic (e.g., drought, flood) stresses. Such stresses characterize many
of SSA’s marginal environments and improving farmers’ resilience to such risks will
reduce their marginality and expand the areas that are favorable for sustainable
agricultural intensification. For example, submergence-tolerant rice and drought-
tolerant maize varieties provide options that reduce farmers’ risks and improve
incentives to invest in additional productivity enhancing technologies. Furthermore,
environmental risks and stresses are now becoming increasingly frequent, less
predictable, and shifting geographically as the climate changes.
Downstream innovation in delivery. Investment and innovation are also needed to
improve the science of delivering information and technology options to farmers, as
well as for gathering information from farmers to ensure that public sector invest-
ments are responsive to their needs. Adapting existing technologies for local
agro-ecological and environmental conditions the poor face offers particularly low-
hanging fruit for productivity gains in SSA. Increasing feed demand and emerging
biofuel markets also create new opportunities for farmers in more marginal areas to
produce crops for non-food markets (Binswanger-Mkhize and McCalla 2010).
However, increasing access to technologies and markets must be accompanied by
concurrent improvements in market strength and function. Ensuring that farmers
have access to the knowledge and products that can increase their productivity is
insufficient to achieve widespread productivity gains and increase favorability for
agricultural intensification if markets are unable to reliably provide the necessary
inputs, absorb increased outputs, and improve price stability.

10.4.3 Policy and Institutional Priorities

Public policy plays an important role at the national level in ensuring that productivity-
enhancing technologies and innovations both reach and benefit smallholder farmers.
10 Poverty, Agriculture and the Environment: The Case of Sub-Saharan Africa 163

Focusing on agriculture as an engine of economic growth and poverty reduction

requires policy interventions and investments at both the national and international
levels to ensure an enabling environment that is favorable for smallholders increasing
their productivity sustainably.
Focus on agriculture as an engine of economic growth and poverty reduction. In
recent years SSA country governments and international donors have demonstrated
renewed interest in agriculture as an engine of economic growth and poverty reduction
as evidenced by regional increases in donor aid and government commitments
for agricultural development (Pingali 2012). The trend is positive, but continued
attention is necessary to keep agriculture on the regional policy agenda. Sustainably
increasing agricultural productivity and reducing marginality and poverty in
SSA will require policy interventions to: (1) sustain investment in agricultural R &
D, (2) improve regulatory and trade policy, and (3) improve infrastructure.
Although public investment in agriculture has increased since 2006, and particularly
after the food price crisis of 2008, public sector R & D investment in SSA still remains
low despite the fact that countries there continue to rely on agricultural productivity
to drive hunger and poverty reduction (Lipton 2005; World Bank 2007). Agricultural
R & D in a large number of developing countries exhibited a negative growth rate
in recent decades. On average, public spending in SSA increased by only 0.6 % per
year from 1981 to 2000, however, half of the countries actually spent less in 2000
on agricultural R & D than a decade prior (Beintema and Elliott 2009).
In addition to limited public sector investment, poor regulatory environments
limit the incentives for private sector engagement along supply chains. A favorable
enabling environment for innovation requires policies and regulations that ensure
intellectual property rights that guarantee that private benefits can be captured as
returns on investments. Functional and transparent biosafety regulations are also
important for reducing risk and uncertainty that otherwise jeopardize potential
returns and discourage investment (Pingali 2012). Weak regulation and high transaction
costs for private sector development more broadly, such as a lack of licensing, price
limits, or existing cartels, limit market development from R & D to the distributions
of inputs and agricultural products that would be necessary to achieve broad
productivity, food security, and poverty reduction goals on the continent. Effective
regulation and transparency require effective institutions to encourage and govern,
for instance, the distribution of agricultural inputs and outputs. Such enabling factors
are necessary preconditions to grow and sustain agricultural markets, yet remain
weak or nonexistent in many countries (World Bank 2012). Lastly, existing policies
governing agricultural R & D and other technologies often inhibit accessing and
employing modern tools such as biotechnology to innovate on behalf of poor farmers,
to improve food security, and to protect environmental interests.
Some policies can actually counteract the impacts of official development
assistance, such as in the cases of trade and investment. African farmers actually
face the highest trade barriers in the world with respect to accessing the inputs that
they need and in getting their food to markets and consumers (World Bank 2012).
The lack of transparency around regulations over food safety and quality standards,
in addition to protectionist policies, impede the cross-border trade of inputs and
164 P. Pingali et al.

agricultural products within the region that could benefit producers as well as
strengthen markets, lower transaction costs, and lower food prices for consumers.
For example, tariffs and quotas vary and are often poorly communicated, which
creates market uncertainty that further limits cross-border trade and contributes to
food price volatility (World Bank 2012).
Rural roads and transportation infrastructure are limited in many countries of
SSA, and distribution networks for inputs and agricultural products are poor.
Transportation costs in Africa remain high, though not necessarily due primarily to
poor roads as is often presumed. A recent study from the World Bank found that
the lack of investment and competition in transport services is the primary factor
contributing to high transportation costs in SSA: cartels are still common in the
transportation sector and roadblocks continue to undermine efficiency, weakening
incentives to invest in modern logistics networks. The World Bank estimates that
reducing transportation costs by 50 % would lead to increases in agricultural GDP
of 7.0 % in Mozambique and 3.0 % in Malawi. Limited cross-border trade, due
mainly to the existing policy and regulatory environments discussed above, further
compounds the negative impacts of weak transportation networks and impedes the
ability of farmers to access markets and move goods from areas of surplus to centers
of demand (World Bank 2012).
Build local capacity for sustainable productivity growth. An enabling policy envi-
ronment for sustainable smallholder productivity growth must involve and be
responsive to the needs of farmers and local communities. At the local level a
community-driven development approach offers a model for capacity building and
local engagement that focuses on empowering local communities to lead their own
development processes. These incorporate five central components, including:
empowering communities and local governments, creating joint responsibility
and control between local and central authorities, improving accountability, and
developing capacity (Binswanger and Nguyen 2004). At the national level, invest-
ments in systems—both institutional reform and infrastructure investment—are
needed. For example, this includes ensuring that national scientific facilities have
access to modern equipment, investing in training and development of local scien-
tists and policymakers, and ensuring that the right information and data are avail-
able to policymakers and decision makers throughout supply chains in order to
foster informed decisions regarding agricultural development strategies.

10.4.4 Technology and Policy Priorities for Sustainable

Intensification

Promoting sustainable intensification strategies can reduce the threat of environmental

degradation and foster long-term productivity. While poverty and environmental
degradation often coexist, the impact of one on the other and the direction of that
causality are not yet well understood. While many argue that poverty is a driver of
environmental degradation, the evidence is not conclusive. Recent gains in poverty
10 Poverty, Agriculture and the Environment: The Case of Sub-Saharan Africa 165

reduction have not demonstrated parallel gains in sustainable resource management

(Hazell and Wood 2008). This underscores the critical importance for sustainable
intensification strategies that both preserve and enhance the natural resource base
in areas where agricultural development offers a promising strategy to reduce mar-
ginality and poverty. Sustainable intensification requires policies that provide incen-
tives to farmers for preserving and improving the natural resource base and that will
ensure the capacity for both short- and long-term productivity growth.
Create incentives for the sustainable use of natural resources. Farmers need the
proper incentives to use resources judiciously and efficiently. Low output prices due
to a lack of market access or market distortions undervalue natural resource and
other inputs, reduce farmers’ profitability, and decrease incentives for investment in
more sustainable farming practices or intensification (Ruben and Kuyvenhoven
2003). Across SSA smallholder access to and ability to invest in improved inputs is
severely limited. Poor market access and high costs lead to low investment in land
improvement. For example, SSA has seen extremely limited use (or the complete
lack) of fertilizers, resulting in declining soil fertility (Ruben et al. 2007). Average
fertilizer applications per hectare in SSA peaked in the 1980s at 10 kg/ha, returning
since to the mean 1970s level of 7.0 kg/ha during the years 2001–2007, which is the
period for which the most recent data are currently available (Binswanger-Mkhize
2012). However, as intensification rises, so do returns on soil fertility improvements
and other productivity investments such as small-scale irrigation. Additionally,
regulations and market mechanisms (i.e., payments for ecosystem services, carbon
finance) both offer potential avenues to change incentives for natural resource use,
however, building the capacity and informing the design of appropriate institutions
to implement and regulate these policy interventions in SSA is still needed.
Crop management practices can also be employed to strengthen the natural
resource base for more sustainable productivity growth. For instance, nutrient manage-
ment and plant protection can increase productivity while reducing environmental
degradation and potentially improving natural resource bases. For example, a variety
of agronomic, biological, and mechanical techniques exist to enhance soil fertility,
structure, and availability/uptake of nutrients. Win-win strategies in this regard
include integrated soil fertility management and intercropping. Crop rotation and
intercropping practices also offer protection against pre-harvest losses. Combining
biological, cultural, genetic, and chemical techniques through integrated pest
management can improve plant protection relative to inorganic pesticides, which
often lose their effectiveness as the species they target adapt. Additionally, water
management techniques such as household- and watershed-level planning to decrease
the variability of water availability are also productivity-enhancing measures, includ-
ing: water-harvesting techniques, surface water diversion, and irrigation (Ruben
et al. 2007). Many of these strategies are knowledge-intensive and must be compli-
mented with the aforementioned investment and innovation in downstream delivery
of information to farmers.
Intensify to release marginal lands from agriculture. Promoting intensification on
lands that have high agricultural potential (i.e., where returns on agricultural
166 P. Pingali et al.

development are favorable) will allow for the release of agriculturally marginal and
ecologically fragile lands from agricultural production, allowing them to provide
other ecosystem services such as increased carbon sequestration through carbon
stock regeneration. Where the opportunity costs of land and labor both remain
low—such as very arid, steep sloped, or rocky areas—returns on investment in
sustainable intensification will continue to be low as well. By moving these
marginal areas out of agricultural production, they will be available to provide other
ecosystem services.
Valuing Ecosystem Services. In the long-term our ability to value and pay for ecosystem
services could be a critical means of addressing agricultural-natural ecosystem
tradeoffs. In the short-term establishing the analytical groundwork and incorporating
the value of natural resources into policy decision making are important steps, however,
neither the metrics nor the institutional frameworks are sufficiently established to
directly pay or penalize smallholder farmers at the local level. Policy priorities in the
short-term should include macro-level interventions to establish an enabling environ-
ment, such as the removal of perverse subsidies to agricultural, fisheries, and energy
sectors that cause harm to people and the environment (Pingali 2012). These can be
complimented in the long term by appropriate payment mechanisms for ecosystem
services that will improve incentives for farmers to pursue the dual objectives of
productivity and maintaining natural resource bases at the production level. Investment
must also focus on the establishment of metrics that are meaningful scientifically
and the institutional infrastructure to implement successful payment for ecosystems
services schemes that could reach and benefit smallholder farmers.

10.5 Conclusion

The marginal areas of SSA have historically been unfavorable for agricultural produc-
tivity improvements due to environmental and/or economic constraints. Up to now
they have largely remained marginal because circumstances have not made investment
in unlocking the constraints underlying their marginal status attractive. Low returns
on potential investments for agricultural development in SSA have been driven
in large part by low population densities and poor market access. However,
demographic shifts in SSA are making productivity enhancing investments more
attractive in many areas of SSA, where demand is rising and the opportunity costs
of land and labor are increasing. As such, agricultural development now offers new
opportunities to reduce marginality and poverty, and to increase the sustainability of
agricultural production strategies. Capitalizing upon these opportunities requires
careful attention to the opportunity costs of land and labor, pursuing intensification
where it makes sense, and releasing marginal lands from agricultural pressures
where returns on productivity growth investments will continue to remain low in
order to allow these areas to provide natural ecosystem services.
While SSA offers new opportunities for decreasing marginality through agricultural
intensification, poor environmental endowments, widely degraded lands, and the
10 Poverty, Agriculture and the Environment: The Case of Sub-Saharan Africa 167

threat of climate change require that productivity growth be achieved as sustainably

as possible. Investing in targeted R & D both upstream and downstream to address
SSA’s particular constraints—especially focusing on the crops and traits that are
important to the poor and the particular environmental conditions they face—can
dramatically increase the continent’s agricultural productivity, lessen marginality,
and contribute to widespread reductions in poverty. Concurrent policy interventions
are necessary to create an enabling environment for productivity, including reducing
distortions in trade policies, improving infrastructure, and improving the regulatory
environment for scientific innovation. Population and market dynamics open new
opportunities for productivity growth in the near-term. Careful strategies to mitigate
agricultural-environmental tradeoffs will ensure that productivity potential is better
preserved in the long term.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Adhikari B (2011) Poverty reduction through promoting alternative livelihoods: implications for
marginal drylands. J Int Dev. doi:10.1002/jid.1820
Bai ZG, Dent DL, Olsson L, Schaepman ME (2008) Global assessment of land degradation and
improvement. Identification by remote sensing. Report 2008/0, International Soil Reference
and Information Center, University of Wageningen, Wageningen
Barbier EB (2010) Poverty, development, and environment. Environ Dev Econ 15(06):635–660
Beintema N, Elliott H (2009) Setting meaningful investment targets in agricultural research and
development: challenges, opportunities and fiscal realities. In: Proceedings of the expert meeting
on how to feed the world in 2050, Rome. Food and Agriculture Organization of the United
Nations, Rome, 24–26 June 2009
Binswanger-Mkhize HP, McCalla A (2010) The changing context and prospects for agricultural
and rural development in Africa. In: Pingali P, Evenson R (eds) Handbook of agricultural
economics. Elsevier BV, Amsterdam
Binswanger-Mkhize HP (2012) The role of agriculture in the structural transformation. In: Timmer
P, McMillan M, Badiane O, Rodrik D, Binswanger-Mkhize HP, Wouterse F (eds) Patterns of
growth and structural transformation in Africa: trends and lessons for future development
strategies. International Food Policy Research Institute, Washington, DC, pp 10–11
Binswanger HP, Nguyen T (2004) Scaling up community-driven development for dummies. World
Bank, Washington, DC
Binswanger HP, Pingali P (1988) Technological priorities for farming in Sub-Saharan Africa.
World Bank Res Obs 3:81–98
Boserup E (1965) The conditions of agricultural growth. Allen and Unwin, London
CGIAR (2000) CGIAR research priorities for marginal lands. Consultative Group on International
Agricultural Research Technical Advisory Committee, Food and Agriculture Organization of
the United Nations, Rome
Christiaensen LJ, Demery L (2007) Down to earth: agriculture and poverty in Africa. Working
paper no 2010/35, World Bank, Washington, DC
Dixon J, Gulliver A, Gibbon D (2001) Farming systems and poverty: improving farmers’ livelihoods
in a changing world. Food and Agriculture Organization of the United Nations and World
Bank, Rome/Washington, DC
168 P. Pingali et al.

Evenson RE, Gollin D (2003) Assessing the impact of the green revolution, 1960 to 2000. Science
300:758–762
FAO (2013) Rural Income Generating Activities (RIGA). Food and Agriculture Organization of
the United Nations, Rome. Available at http://www.fao.org/economic/riga/riga-database/ en/
Gatzweiler F, Baumüller H, von Braun J, Ladenburger C (2011) Marginality: addressing the
root causes of extreme poverty. ZEF working paper no 77, Center for Development Research,
University of Bonn, Bonn
HarvestChoice (2010) Commissioned research. International Food Policy Research Institute,
Washington, DC
Hazell P (2010) An assessment of the impact of agricultural research in South Asia since the green
revolution. In: Pingali P, Evenson RE (eds) Handbook of agricultural economics, vol 4. Elsevier
BV, Amsterdam, pp 3469–3530
Hazell P, Wood S (2008) Drivers of change in global agriculture. Philos Trans R Soc Lond B Biol
Sci 363(1491):495–515
Herdt R (2010) Development aid and agriculture. In: Pingali P, Evenson RE (eds) Handbook of
agricultural economics, vol 4. Elsevier BV, Amsterdam, pp 3253–3304
Johnston B, Mellor J (1961) The role of agriculture in economic development. Am Econ Rev
51:566–593
Lipper L, Pingali P, Zurek M (2007) Less-favoured areas: looking beyond agriculture towards
ecosystem services. In: Ruben R, Pender J, Kuyvenhoven A (eds) Sustainable poverty reduction
in less-favoured areas. CABI Publishing, Wallingford, pp 442–460
Lipton M (2005) The family farm in a globalizing world: the role of crop science in alleviating
poverty. International Food Policy Research Institute, Washington, DC
Millennium Ecosystem Assessment (2005) Ecosystems and human well-being: current state and
trends, vol 1. Island Press, Washington, DC
Otsuka K, Kijima Y (2010) Technology policies for a green revolution and agricultural transformation
in Africa. J Afr Econ 19:ii60–ii76
Pardey PG, Alston JM, Kang CC (2012) Agricultural production, productivity and R & D over the past
half century: an emerging new world order. Plenary paper presented at the International Association
of Agricultural Economists (IAAE) Triennial Conference, Foz do Iguacu, 18–24 August 2012
Pingali P (2001) Population and technological change in agriculture. In: Smelser NJ, Baltes PB
(eds) International encyclopedia of the social and behavioral sciences. Elsevier Science Ltd,
Oxford, pp 11742–11744
Pingali P (2010) Agriculture renaissance: making “agriculture for development” work in the 21st
century. In: Pingali P, Evenson RE (eds) Handbook of agricultural economics, vol 4. Elsevier
BV, Amsterdam, pp 3867–3894
Pingali P (2012) Green revolution: impacts, limits, and the path ahead. Proc Natl Acad Sci U S A
109:12302–12308
Reardon T, Barrett C, Kelly V, Savadogo K (2002) Policy reforms and sustainable agricultural
intensification in Africa. Dev Policy Rev 17:375–395
Ruben R, Kuyvenhoven A (2003) Investing in poor people in less-favoured areas: institutions,
technologies and policies for poverty alleviation and sustainable resource use. Paper presented
at the international conference on staying poor: chronic poverty and development policy,
University of Manchester, Manchester, 7–9 March 2003
Ruben R, Pender J, Kuyvenhoven A (2007) Sustainable poverty reduction in less-favoured areas:
problems, options and strategies. In: Ruben R, Pender J, Kuyvenhoven A (eds) Sustainable
poverty reduction in less-favoured areas. CABI Publishing, Wallingford, pp 1–62
UNDP (2009) Climate change in the African drylands: options and opportunities for adaptation
and mitigation. United Nations Development Programme, New York
Webb P (2009) More food, but not yet enough: 20th century successes in agriculture growth and
21st century challenges. In: Eiselen H (ed) Fiat panis: for a world without hunger. Hampp
Media/Balance Publications, Stuttgart
World Bank (2007) World development report 2008: agriculture for development. World Bank,
Washington, DC
World Bank (2012) Africa can help feed Africa: removing barriers to regional trade in food staples.
World Bank, Washington, DC
Chapter 11
The Marginal Poor and Their Dependence
on Ecosystem Services: Evidence from South
Asia and Sub-Saharan Africa

Pushpam Kumar and Makiko Yashiro

Abstract In this chapter the authors employ a meta-study to explore why it is

critical to address the degradation of ecosystems for poverty alleviation, especially
in South Asia and Sub-Saharan Africa. The authors also investigate the linkages
between ecosystem services and aspects of extreme poverty. Their findings suggest
that the poor are often more vulnerable to the loss of ecosystem function that
restricts the supply of natural goods and services. The poor depend upon ecosystem
services, but the nature of this dependence is necessarily not uniform throughout the
year. The poor also tend to benefit less from environmental conservation efforts than
those who are not poor. The dynamic patterns of dependence on ecosystem services
of the poor and their coping strategies require regionally specific and in-depth
evaluation.

Keywords Ecosystem services • Poverty alleviation • Vulnerability • Linkages •

Agriculture

11.1 Introduction

Poverty may be defined as pronounced deprivation in well-being, which not only

signifies material deprivation as measured by an appropriate concept of income or
consumption, but may also include low achievement in education and/or health,
high vulnerability to and/or exposure to risk, and a lack of socio-political repre-
sentation or powerlessness (World Bank 2000). Multiple dimensions of well-being
were also highlighted in the conceptual framework used for the Millennium

P. Kumar (*) • M. Yashiro

Ecosystem Services Economics Unit, Division of Environmental Policy Implementation,
United Nations Environment Programme, P.O. Box 30552, Nairobi 00100, Kenya
e-mail: [email protected]; [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 169
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_11,
© The Author(s) 2014
170 P. Kumar and M. Yashiro

Ecosystem Assessment (MA 2005a) including: security, basic materials for living,
health, social relations, and freedom of choice and action. The conflicting interests
among various stakeholders who depend on common ecosystem services and
products highlight the political economy in regions with high levels of poverty
like South Asia and Sub-Saharan Africa, emphasizing the fact that powerlessness
is an important dimension of poverty because it is the powerless who suffer most
when such conflicts arise. The fact that the poor are often more critically affected
by ecosystem degradation due to reduced access to resources and a lack of alter-
natives is usually evident (ESPASA 2008). Migration and finding alternative
livelihoods (under duress rather than choice) are the most prevalent coping strate-
gies for the natural resource dependent poor who lose these conflicts, but these
strategies often result in homelessness, lowered self-esteem, and feelings of alien-
ation and/or detachment.
In this paper we explored why, in this context, it is critical to address the degrada-
tion of ecosystems for poverty alleviation, especially in South Asia and Sub-Saharan
Africa. We also investigated the linkages between ecosystem services and aspects of
extreme poverty. Subsequently we synthesized literature evidence of the dynamic
relationships between the incidence of poverty and the state of ecosystems from the
regions of South Asia and Sub-Saharan Africa.

11.2 Why Is It Important to Address the Marginal

Poor to Achieve Poverty Alleviation?

There is great variation among countries in the achievement of poverty alleviation.

For example, among the five countries of South Asia (Bangladesh, Bhutan, India,
Nepal, and Pakistan), all except Pakistan have experienced a declining trend in the
incidence of poverty—in terms of national poverty lines—since the mid-1990s.
According to the 1997 Human Development Report issued by the United Nations
Development Programme, poverty is usually worse in drier zones relative to more
humid zones (UNDP 1997). The MA (2005b) also reported that in arid and semi-arid
lands, people’s dependence on ecosystem services is often high because of the
limited availability of alternative livelihood options in fragile environments. In India
arid and semi-arid regions include 125 districts in over 12 states that are officially
identified as drought prone areas under the Drought Prone Area Programme, and 32
of these districts have either a high or very high incidence of poverty. The non-
income dimensions of poverty are very much evident in drought prone regions:
livelihood security is low on account of the high instability of crop production and
there are significant social costs on account of large-scale inter-state population
movements that result from resulting food shortages (Mehta and Shah 2006).
In South Asia poverty is primarily a rural phenomenon, with the majority of the
population living in rural areas and primarily dependent on agriculture for income
and employment (World Bank 2008). The ability of the rural poor to sustain their
livelihoods is generally constrained due to adverse environmental conditions: high
11 The Marginal Poor and Their Dependence on Ecosystem Services… 171

ecological vulnerability, low productivity of natural resources, and limited access to

land and other resources (World Bank 2002). Poverty in forested regions of India,
however, is also linked to widespread entitlement failure (Mehta and Shah 2006).
The rural poor are also hampered by their lack of access to markets, reliance on rain-
dependent agriculture, and the prevalence of threats to food security. For the rural
poor access to a variety of natural resources is often critical for sustaining livelihoods
because they provide diversification options when environmental conditions change
(Koziell and Saunders 2001). Chronic rural poverty in semi-arid regions of India has
also been attributed to the negligible or inferior natural resource endowments that
restrict the ability to augment income (Singh and Binswanger 1993).

11.2.1 Vulnerability and Insecurity

A key dimension of poverty is vulnerability, which reflects a household’s resilience

in the face of shocks and the likelihood that a shock will lead to a decline in well-
being (World Bank 2008). Poor households are vulnerable to sudden and pronounced
fluctuations in income that may arise due to poor health, market fluctuations, and
natural calamities. Since most rural poor are dependent on agricultural livelihoods
they are automatically the most vulnerable to climate-change induced risks of crop
failure and livestock losses.
For example, vulnerability to natural disasters is fairly high in Bangladesh. During
seasonal floods women and children become particularly vulnerable to related health
impacts and wage laborers suffer from shortages of employment opportunities.
For many rural households in Bangladesh riverbank erosion is a constant threat to
their well-being. The impacts are severest among the landless and marginalized
farmers. Hutton and Haque (2004) suggested that the displaced, particularly women,
suffer mental stress because of social fragmentation and the difficulties of adjusting
to the urban areas where they often immigrate.
Unanticipated environmental consequences of development projects have also
often been a great source of misery for local communities. The construction of
upstream projects may create downstream environmental hazards that are detrimen-
tal to livelihoods. For instance, the construction of embankments in the Ganga-
Brahmaputra river basin to moderate flood impacts has caused large areas in the
basin to remain in a semi-permanent waterlogged state, seriously affecting human
health and agriculture (Bandyopadhya 2002).
The lack of explicit government policy on ecosystems services, the absence of a
suitable accounting system, the general lack of awareness, and the lack of research
evidence to convince policy-making authorities to recognize and institutionalize the
management of ecosystem services have jointly played a role in the degradation of
ecosystems in South Asia and Sub-Saharan Africa with negative impacts on the
poor. Aberrations in ecosystem function such as unexpected flooding, long droughts,
desiccation of springs, increased spread of invasive species on productive lands, and
reduced productivity of natural resources like non-timber forest products or fish
172 P. Kumar and M. Yashiro

usually have the greatest impact on the poorest people who directly depend on these
products for subsistence purposes.
Distortions take place in policy cases, like offering subsidies for fertilizers and
pesticides to farmers to enhance crop productivity (for poverty alleviation), that are
oblivious of the resultant negative impacts on soil quality. While the provision of
natural ecosystems services is diminishing, demand is constantly growing.
Furthermore, ecosystems are not treated in policy frameworks as ‘natural capital,’
with tremendous potential to generate employment and income.

11.3 Links Between Ecosystem Services and Poverty

The links between the status of ecosystems and human well-being, including the
importance of various ecosystem services and products to human well-being, have
been increasingly recognized in recent years (MA 2005a, b; ESPASSA 2008;
Shackleton et al. 2008; Tallis et al. 2011). Furthermore there is a general consensus
that poverty is a major contributor to environmental degradation (WCED 1987).
That report (also known as the “Brundtland Report”), as well as the definition of
sustainable development developed by the World Commission on Environment and
Development (WCED), were landmarks in the sense that they identified direct links
between environmental degradation and poverty in the context of economic devel-
opment in developing countries, and that they called for collective action to address
environmental threats and economic development goals as interrelated phenomena
(Speth and Haas 2006). The World Bank joined the consensus with the 1992 World
Development Report, which highlighted the dependence of poor families on natural
capital for meeting their short-term livelihood needs, often leading to environmental
degradation.
There is shared recognition that most natural ecosystems are in a state of decline,
with negative impacts on human well-being, especially for those living in extreme
poverty. The MA (2005a) findings highlighted significant negative trends in the
delivery of many ecosystem services and concluded that these declines were barri-
ers to achieving the Millennium Development Goals (MDGs). In fact many of the
regions facing severe problems related to the deterioration of ecosystem services
overlap with those facing significant challenges to achieving the MDGs.
Many of the world’s poorest people live in rural areas and are highly dependent
on ecosystem services and products, such as those that contribute to food produc-
tion via agriculture, fishing, and hunting. As previously mentioned, people living
in extreme poverty are highly vulnerable to ecosystem changes such as the avail-
ability or quality of water, or the loss of wetland and coral reef ecosystems that
might increase the likelihood of coastal floods or storm damage. In addition to the
goal of eradicating extreme poverty, the ability of ecosystems to supply services
also has strong relevance in addressing other aspects of the MDGs, such as the
goal of eradicating hunger, the reduction of child mortality, combating disease,
and ensuring environmental sustainability (MA 2005a). At the same time,
11 The Marginal Poor and Their Dependence on Ecosystem Services… 173

achieving many of the aspects of well-being that are linked with ecosystem
services will directly or indirectly contribute to the attainment of a number of the
MDGs (UNEP/IISD 2004).

11.4 Global Evidence

The relationships between poverty, the environment, and ecosystems have

commanded a significant amount of attention by academics, non-governmental
organizations (NGOs), development practitioners, and civil society. A number of
influential studies (Dasgupta 1997; Pearce 2005; Barbier 2010) have dispelled the
usual beliefs about the relationships between poverty and the environment rather
than establish any profound new insight into these relationships. Conventionally
it is assumed that the poor discount the future and are unable to act with regard to
long-term interests, and hence are responsible for mining the soil, depleting the
groundwater, and causing deforestation more than their wealthier counterparts
(Dasgupta 1995). Many others believe that affluence increases the use of forest
products, pollution generating products, and leads to a lifestyle that is more inimi-
cal to the existence of nature. Both perspectives are correct as well as wrong, as
they happen to be purely contextually and culturally specific, and are dependent
on: property rights regimes, the prevailing political climate, historical trends of
resource use, relevant social and religious practices, and the aspirations of con-
sumption fueled by the global market and investment conditions. While these
drivers of change remain relevant to the relationships between poverty and the
environment, the degree of association and dimensions of causality can be per-
ceived in various ways, and the particular environmental resources or attributes
linked to poverty also need to be considered.
It is important to mention that while natural resources remain self-explanatory,
poverty here refers to absolute and chronic poverty, where the material conditions
of people’s lives are well below designated cut-offs, which indicates the limited
ability to buy enough food for minimum caloric intake. Evidently this is a very
conservative method of defining poverty and many aspects of well-being fall out-
side uni-dimensional definitions. Therefore poverty in such cases is a measure of
the distance of people’s actual condition from a state considered ‘well,’ but it does
not necessarily reflect their well-being.
The facts remain that the unsustainable use of natural resources contributes to
poverty, and on the other hand that poverty can also contribute to environmental
degradation (Duraiappah 1998). To solve these problems both poverty alleviation
and environmental policies need to be addressed. Many developing countries view
the need for improved management of natural resources as part of the attack on the
underlying causes of degradation and depletion, specifically excessive population
growth and poverty, which are the main drivers of immigration into the remaining
areas of natural habitat. Strong programs to reduce population growth rates and cre-
ate jobs for unemployed and underemployed rural residents will be crucial to the
174 P. Kumar and M. Yashiro

long-term prospects of reducing environmental degradation, especially to reduce

deforestation (Pfaff et al. 2000).
Many of the world’s poor occupy the least resilient, most threatened environmental
areas (Pearce and Warford 1993). There is also some evidence that the poor are
more dependent upon common property natural resources than the rich (Jodha 1986).
The assertion is often made that poor people have a greater tendency to overexploit
natural resources like land, forests, and water, and thereby degrade them. This is
based on the belief that the poor deplete natural resources at a greater rate than their
more affluent counterparts because they have direct access to them and limited pros-
pects of gaining access to other types of resources and economic opportunities.
Additionally, because the poor struggle for their subsistence they are preoccupied
with day-to-day survival, and are assumed to have limited incentives and capability
for long-term planning or investing in management efforts that improve the sustain-
ability of natural resource use over the long-term (i.e., soil conservation efforts).
Thus the poor have little choice but to employ the natural resource use practices that
fulfill their immediate subsistence needs, even if these practices are detrimental to
those resources over the long-term.
Growing poor populations in precarious environments (i.e., ecologically fragile
areas) are a major cause of the severe environmental destruction that has been docu-
mented in developing countries by numerous recent studies (Comim et al. 2009).
The problems of poverty and environmental degradation are complicated and
resolving them is made vastly more urgent by the relentless increase in the number
of people living in developing countries. In many cases poor people must degrade
their environment just to make ends meet, but in doing so they take not only from
nature’s bounty, but also from the well-being of future generations.
In addition to being agents of environmental degradation, the poor are also often
the victims of environmental damage due to the fact that they often depend heavily
on natural resources for subsistence, and therefore are generally more affected by
environmental degradation than those who are not poor (Tallis et al. 2011). Due to
this strong dependence of the poor on natural resources the degradation of ecosys-
tems may affect their survival and ‘trap’ them in poverty (Comim et al. 2009). Natural
resources often serve as de facto safety nets for the poor, especially at times of stress
such as poor performance of subsistence agriculture (Pattanayak and Sills 2001).
A greater number of poor people suffer from extreme natural disasters due to the
facts that they predominate in the areas and under conditions that are more suscep-
tible to damage from these events, they lack the resources to cope with the impacts
of these events, and they have less access to social safety nets that alleviate the
impacts of these events (UNEP/IISD 2004).
Thus it is generally the poorest who suffer the most from the consequences of
pollution and environmental degradation, and degraded environments can acceler-
ate the process of impoverishment because the natural assets and common resources
that the poor depend directly become less available. They often derive a large part
of their livelihoods from nonmarket environmental resources, common grazing
lands, or forests where food, fuel, and building materials are gathered. The dimin-
ishment or loss of such resources therefore may particularly hurt the poorest and
11 The Marginal Poor and Their Dependence on Ecosystem Services… 175

undermine their future productivity. Moreover shortages of drinking water or fuel

wood tend to affect the poor more than the people with greater means. Environmental
degradation depresses people’s income by causing them to invest more effort into
routine household tasks such as fuel wood and water collection by decreasing the
productivity or availability of these natural resources from which they wrest a living.
The synergetic interactions between poverty and environmental destruction can lead
to a downward spiral of ecological deterioration that threatens the well-being of
many of the world’s poorest people (Dasgupta 2010).

11.4.1 Evidence from South Asia and Sub-Saharan Africa

In response to the growing recognition of the linkages between declining ecosystem

services and poverty, conservation organizations and development agencies have
been designing and implementing initiatives that focus on maintaining ecosystem
services. Although the concept of ecosystem services provides a new platform for
the challenge of aligning conservation with development objectives through
common links to human well-being, the success of on the ground efforts has been
limited. This is due partly to the perception among practitioners that it is difficult
to integrate these objectives consistently, as well as some skepticism within the
environmental community of the application of an ecosystem services-based
approach to conservation (Tallis et al. 2009). Furthermore the availability of robust
empirical studies on the nexus between poverty and the environment is limited,
particularly due to difficulties in obtaining relevant data from developing countries
(Dasgupta et al. 2005).
In the following sections we have highlighted reported examples of the depen-
dence on ecosystem services by people living under extreme poverty from countries
in South Asia and Sub-Saharan Africa in order to contribute to awareness of ecosys-
tem management efforts and poverty alleviation. The information on the links
between ecosystem services and poverty alleviation summarized in the following
sections is drawn from studies undertaken by the Consortium on Ecosystems and
Poverty Alleviation (CEPSA) in South Asia (ESPASSA 2008) and Sub-Saharan
Africa (Shackleton et al. 2008). The South Asia effort focused on India and the Hindu
Kush Himalayan Region including: Bangladesh, Bhutan, Nepal, and Pakistan. South
Asia is home to almost one-quarter of the world’s population, and has experienced
intensive development processes that in many cases have led to severe environmental
degradation. The Sub-Saharan Africa component of the study includes eight case
studies with in-depth analyses and local information from arid and semi-arid areas
of: Botswana, Mozambique, Namibia, South Africa, Swaziland, and Zimbabwe
(Shackleton et al. 2008). In arid and semi-arid lands people’s dependence on ecosys-
tem services is often high due to limited alternative livelihood options, the fragile
nature of these ecosystems, and relatively high risk involved in achieving livelihood
objectives (MA 2005b). A brief summary of the links between different types of
ecosystem services and poverty based on these studies is provided below.
176 P. Kumar and M. Yashiro

11.4.2 Evidence from South Asia

In India and the Hindu Kush Himalayan Region rural communities typically depend
on ‘provisioning services,’ or products provided by ecosystems. Forest ecosystems
provide resources used for energy, food, animal fodder, medicine, farm implements,
and various household assets, and grassland ecosystems provide grazing areas for
domestic as well as wild herbivores, although most of the original grasslands of
India’s Gangetic plains have disappeared due to population pressure.
The livelihoods of a large population of pastoral and agro-pastoral communities
in the arid and semi-arid regions of Bhutan, India, Nepal, and Pakistan depend on
common pasturelands and free-grazing livestock. The ‘provisioning services’
provided by freshwater wetlands, such as fish and other aquatic resources were also
important to the livelihoods of rural communities. Bangladesh, India, and Pakistan
have extensive coastlines that include wetlands and these areas often host dense
populations with a high incidence of poverty (ESPASSA 2008).
The wetlands of South Asia are threatened by encroachment, unsustainable
harvesting of aquatic resources, industrial pollution, agricultural runoff, and the
spread of invasive exotic species. While agro-ecosystems play an important role in
the provision of food and fiber for the rural poor, help preserve scenic rural land-
scapes, and ensure groundwater recharge, they also have negative impacts on
ecosystems such as nitrate run-off from cropland to downstream catchments and
soil erosion from overgrazed hillsides. Grasslands also provide such ecosystem
services as carbon sequestration, methane absorption, and the reduction of nitrogen
dioxide emissions. In addition to the provision of wood for fuel and construction, a
wide range of ecosystem services provided by mangroves were highlighted, including:
shoreline stabilization, storm protection, water quality control, micro-climate stabi-
lization, groundwater recharge, flood control, sediment and nutrient retention, and
providing essential habitat for important bird and marine species.
As is the case in Sub-Saharan Africa, the cultural services provided by ecosys-
tems are important in South Asian countries, for example forest and mangrove
ecosystems have recreational and tourism value, generating important income for
the poor. One example is the Sundarban Tiger Reserve, where part of the extensive
mangrove forest spanning the India-Bangladesh border provides livelihood oppor-
tunities based on ecotourism (Chopra et al. 2009; Kumar 2012). One of the key
findings of the consortium study (EPASSA 2008) in South Asia, however, was the
asymmetrical distribution of economic benefits favoring those who can invest and
operate in ecotourism over the poor.

11.4.3 Evidence from Sub-Saharan Africa

The poor in Sub-Saharan Africa depend heavily on various natural resources for
energy needs, food, medicines, construction, crafts, tools, and for ritual and cul-
tural purposes. According to the valuation studies conducted in arid and semi-arid
areas of the region reviewed by CEPSA (Shackleton et al. 2008), wild resources
11 The Marginal Poor and Their Dependence on Ecosystem Services… 177

such as fuel wood, wild foods, and construction materials have significant economic
value for rural households with the equivalent income share from these products
representing as much as one-third of total household incomes.
Many wild resources such as raw materials, medicines, and botanical oil extracts
are traded in local, national, and increasingly, international markets to generate
income. The development of natural products for specialized cosmetic, pharmaceuti-
cal, food, and health markets is increasingly seen as an opportunity to alleviate
poverty while maintaining ecosystem health. A significant proportion of the har-
vested natural products are for social, spiritual, and cultural purposes, exemplified by
the fact that in eastern parts of the cape of South Africa the amount of plant material
harvested for cultural uses exceeded that for utilitarian uses (Shackleton et al. 2008).
Natural products are often gathered, used and/or sold in times of crisis to hedge
income gaps and to meet specific needs such as school fees or celebration costs. This
kind of natural safety net helps the poor reduce household risk and vulnerability, and
their ability to cope with difficult events such as drought, disease, the escalation of
commodity prices, and conflict. Wild foods and medicinal plants were two of the most
common and important uses of natural products relevant to human well-being in the
region. The role of wild foods in ensuring household food security and of wild plants
used as traditional medicines are important to the poor, though the local extinction of
certain species due to commercial overharvesting was also reported in the region
(Shackleton et al. 2008). A meta-study undertaken by Vira and Kontoleon (2010)
highlighted evidence of the dependence of the poor on biodiversity for their income.
Fuel wood also plays an important role in support of the poor who are reliant on
it for household energy and therefore changes in the availability and access to fuel
wood can significantly affect their livelihoods and well-being. Fresh water provided
by rivers and wetlands is also essential for supporting the poor, for both domestic
purposes and for supporting agricultural-based livelihoods. Ensuring sufficient
quantities of fresh water is a challenge due to the lack of necessary infrastructure.
In relation to regulating services, soil fertility is a key factor that affects the crop
production capacity of the poor. Declines in soil fertility potentially deepen poverty
levels through various mechanisms, such as reducing crop yields that in turn
decreases sales and income, contributing to food insecurity, and diverting scarce
cash resources from other needs to purchase food and/or fertilizers. The flood con-
trol services provided by natural wetlands in arid and semi-arid areas of southern
Africa are important regulating services for the poor. The spiritual, aesthetic, and
recreational services provided by ecosystems were also highlighted by the study as
important for the poor (Shackleton et al. 2008).
The studies conducted by the Consortium on Ecosystems and Poverty in South
Asia and Sub-Saharan Africa (ESPASSA 2008; Shackleton et al. 2008), reported
knowledge gaps in the identification of the whole range of ecosystem services
and their values. While it is widely recognized that the well-being of the poor is
affected by changes in ecosystem services both directly and indirectly, very few
existing studies address these mechanisms. Furthermore the study in Sub-Saharan
Africa (Shackleton et al. 2008) highlighted the fact that most of existing work to
date has focused on natural resources, while other types of ecosystem services have
not been well captured. Due to the fact that the poor often live in marginal areas that
178 P. Kumar and M. Yashiro

are most susceptible to the negative impacts of declining ecosystem services such as
flooding, drought, poor air quality, and soil degradation, further work is required to
better understand these aspects. It was also noted in that report that ecosystem man-
agement projects implemented by various agencies did not adequately consider
poverty alleviation issues, highlighting the need for further research and communi-
cation of the importance and value of ecosystem services to convince policy-makers
and practitioners to integrate conservation and development efforts.

11.5 Conclusions and Lessons Learned

• The poor are often more vulnerable to the loss of ecosystem services that result
in diminished supplies of natural goods and services. There is often a socioeco-
nomic asymmetry in the benefits from conservation. Those who are not poor
often benefit more economically than the poor from environmental conservation
efforts, while the poor generally suffer more from environmental degradation.
A robust theoretical and implementable framework is required to conceptualize
the linkages between poverty and ecosystem services.
• The poor often depend upon ecosystem services, but the nature of this dependence
is usually not uniform throughout the year. In most areas there are cyclical annual
weather patterns that determine the timing of agricultural activities and natural
phenomena like flood and drought. The seasonal patterns of dependence of the poor
and their coping strategies require regionally specific and in-depth evaluation.
• In all of the reviewed cases of poverty-ecosystem linkages indicators of the
intensity and directionality of linkages were either diffuse or incoherent. Future
research should seek indicators that enable more coherent understanding of
directionality and appropriate policy approaches.
• We have a limited understanding of the links between poverty alleviation and the
range of ecosystem services. Information is particularly limited on how regula-
tory ecosystem services, many of which are critical for supporting the lives of the
poor, affect their well-being. Increased understanding of these linkages is critical
for understanding the relationships between ecosystem services and poverty.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Bandyopadhya J (2002) A critical look at the report of the World Commission on Dams in the context
of the debate on large dams on the Himalayan rivers. Int J Water Resour Dev 18(1):127–145
Barbier EB (2010) Poverty, development, and environment. Environ Dev Econ 15(6):635–660
Chopra K, Kapuria P, Kumar P (2009) Biodiversity, land-use change and human well being.
Oxford University Press, New Delhi
11 The Marginal Poor and Their Dependence on Ecosystem Services… 179

Comim F, Kumar P, Sirven N (2009) Poverty and environment links: an illustration from Africa.
J Int Dev 22:447–469
Dasgupta P (1995) An inquiry into well-being and destitution. Oxford University Press, New York
Dasgupta P (1997) Environmental and resource economics in the world of the poor. Resources for
the Future, Washington, DC
Dasgupta P (2010) Nature’s role in sustaining economic development. Philos Trans R Soc B
365:5–11
Dasgupta S, Deichmann U, Meisner C, Wheeler D (2005) Where is the poverty-environment
nexus? Evidence from Cambodia, Lao PDR, and Vietnam. World Dev 33(4):617–638
Duraiappah A (1998) Poverty and environmental degradation: a review and analysis of the nexus.
World Dev 26(12):2169–2179
ESPASSA Ecosystem Services and Poverty Alleviation Study in South Asia (2008) A situation
analysis for India and the Hindu Kush Himalayan region. The Energy and Resource Institute,
New Delhi
Hutton D, Haque EE (2004) Human vulnerability, dislocation and resettlement: adaptation
processes of river-bank erosion-induced displaces in Bangladesh. Disasters 28(1):41–62
Jodha N (1986) Common property resources and rural poor in dry regions of India. Econ Pol Wkly
21(27):1169–1181
Koziell I, Saunders J (2001) Living off biodiversity: exploring livelihoods and biodiversity, issues
in natural resources management. International Institute for Environment and Development,
London
Kumar P (2012) Impact of economic drivers on mangroves of Indian Sundarbans: an exploration
of missing links. Environ Dev Sustain 14(6):939–953
MA (2005a) Ecosystems and human well-being: general synthesis. Millennium ecosystem assessment.
Island Press, Washington, DC
MA (2005b) Ecosystems and human well-being: desertification synthesis. Millennium ecosystem
assessment. Island Press, Washington, DC
Mehta AK, Shah A (2006) Chronic poverty in India: overview study, CPRC working paper 7.
Chronic Poverty Research Centre, University of Manchester, Manchester
Pattanayak SK, Sills EO (2001) Do tropical forests provide natural insurance? The microeconomics
of non-timber forest product collection in the Brazilian Amazon. Land Econ 77(4):595–612
Pearce DW (2005) Investing in environmental wealth for poverty reduction, report prepared on behalf
of the poverty environment partnership. United Nations Development Programme, New York
Pearce DW, Warford J (1993) World without end: economics, environment, and sustainable
development. Oxford University Press, New York
Pfaff ASP, Kerr S, Hughes RF, Liu S, Sanchez A, Schimel D, Tosi J, Watson V (2000) The Kyoto
protocol & payments for tropical forest: an interdisciplinary method for estimating carbon-
offset supply and increasing the feasibility of a carbon market under the CDM. Ecol Econ
35(2):203–221
Shackleton C, Shackleton S, Gambiza J, Nel E, Rowntree K, Urquhart P (2008) Links between
ecosystem services and poverty alleviation: situation analysis for arid and semi-arid lands in
southern Africa. Consortium on Ecosystems and Poverty in Sub-Saharan Africa, Cape Town.
http://www.unpei.org/PDF/preliminaryassessments/Links-Ecosystem-Services-and-Poverty-
Alleviation.pdf. Accessed 15 Feb 2013
Singh RP, Binswanger H (1993) Income growth in poor dryland areas of India’s semiarid tropics.
Indian J Agric Econ 48(1):52–64
Speth JG, Haas PM (2006) Global environmental governance. Island Press, Washington, DC
Tallis H, Goldman R, Uhl M, Brosi B (2009) Integrating conservation and development in the
field: implementing ecosystem service projects. Front Ecol Environ 7(1):12–20
Tallis H, Pagiola S, Zhang W, Shaikh S, Nelson E, Stanton C, Shyamsundar P (2011) Poverty and
the distribution of ecosystem services. In: Kareiva P, Tallis H, Ricketts TH, Daily GC, Polasky
S (eds) Natural capital: theory and practice of mapping ecosystem services. Oxford University
Press, Oxford, pp 278–293
180 P. Kumar and M. Yashiro

UNDP (1997) Human development report 1997. United Nations Development Programme. Oxford
University Press, New York
UNEP/IISD (2004) Exploring the links: human well-being, poverty and ecosystem services.
United Nations Environment Programme/International Institute for Sustainable Development,
Nairobi/Winnipeg
Vira B, Kontoleon A (2010) Dependence of the poor on biodiversity: which poor, what biodiversity?
Paper presented at the CPRC international conference 2010, Chronic Poverty Research Centre,
University of Manchester, Manchester, 8–10 September 2010
WCED (1987) Our common future. World Commission on Environment and Development, United
Nations, New York
World Bank (2000) World development report 2000/2001: attacking poverty. World Bank,
Washington, DC
World Bank (2002) Linking poverty reduction and environmental management: policy challenges
and opportunities. World Bank, Washington, DC
World Bank (2008) Social dimensions of climate change: workshop report 2008. World Bank,
Washington, DC
Chapter 12
Land Degradation, Poverty and Marginality

Nicolas Gerber, Ephraim Nkonya, and Joachim von Braun

Abstract This chapter emphasizes the complexity and plurality of the types and
magnitudes of causal relationships between poverty and environmental degradation,
based on a review of the literature. The authors use case studies focused on the issue
of land degradation to illustrate these relationships. Land degradation (LD) is
influenced by natural and anthropogenic factors, including socioeconomic condi-
tions. LD is of importance to people because it decreases the provision of terrestrial
ecosystem services and the benefits they provide for human well-being. A key ques-
tion is whether lower levels of well-being lead to more or less destructive resource
use and management strategies. The authors call for a systematic and science-based
assessment of LD worldwide as a necessary first step toward the inclusion of LD in
global measures of well-being.

Keywords Ecosystem services • Global indicators • Land management • Agricultural

productivity • Environmental Kuznets curve • Land degradation

12.1 Introduction

Land degradation (LD) has affected many parts of the world for centuries and is
reported to be increasing in extent and severity. LD has negative consequences on
the productivity of land and the ability to provide ecosystem services, and thus

N. Gerber (*) • J. von Braun

Center for Development Research (ZEF), University of Bonn,
Walter Flex Strasse 3, 53113 Bonn, Germany
e-mail: [email protected]; [email protected]
E. Nkonya
International Food Policy Research Institute (IFPRI),
2033 K Street N.W., Washington, DC 20006, USA
e-mail: [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 181
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_12,
© The Author(s) 2014
182 N. Gerber et al.

increases environmental and social vulnerability. As used in this chapter, LD

captures comprehensively the terms ‘land and soil degradation’ as well as ‘desertifica-
tion’ and refers to the long-term decline of ecosystem function and productivity.
In this matter we followed the most recent definitions of LD (Nachtergaele et al.
2010) used by The Global Land Degradation Information System (GLADIS).
Humid areas of the world account for a higher share of the world’s total degraded
land (78 %) than previously thought (Bai et al. 2008a). Nonetheless LD of arid and
semi-arid areas (often referred to as desertification) remains a significant global
problem because they host 38 % of the world’s population (Reynolds et al. 2007)
and are typically less resilient to environmental disturbance than more humid areas.
According to the Global Land Degradation Assessment (GLADA) LD is increasing,
with almost one-quarter of the total global land area having been degraded between
1981 and 2003. Furthermore Nachtergaele et al. (2010) estimate that 1.5 billion people
depended on degraded areas (to variable degrees) for their livelihoods in 2007,
while in excess of 42 % of the world’s very poor population lived in degraded areas.
LD is influenced by natural and anthropogenic factors, including socioeconomic
conditions. Most human-induced LD occurs as a result of the interactions between
the land and its users (Vlek et al. 2010). LD is of importance to humans because it
negatively affects the provision of ecosystem services and the benefits they provide
to human well-being. Unfortunately and despite considerable advances in the
economic valuation of ecosystem services in recent years, there are still many gaps
in the valuation of such services, including the economic valuation of specific
ecosystem services and the geographic dimension of such valuation. LD can be
particularly detrimental to ecosystem function and human well-being in arid and
semi-arid areas, where the meaningful valuation of ecosystem services has been
under-represented.1
LD can affect any ecosystem, including anthropogenic ecosystems such as agro-
ecosystems, forestry plantations, rangelands, and urban areas (Ellis and Ramankutty
2008). Anthropogenic and natural ecosystems provide environmental support and
regulation services (e.g., nutrient and hydrological cycling, climate regulation, etc.).
Hence the impacts of LD on human well-being should be assessed across all terrestrial
ecosystems for an accurate valuation of these impacts. The links between human
well-being and LD are multiple and multi-directional due to the variety of human
benefits derived from ecosystem services. Furthermore human well-being and espe-
cially human ‘ill-being’ (i.e., poverty) can be both a cause and a consequence of LD.
LD impacts human well-being in several dimensions or components. The
Millennium Ecosystem Assessment (MA 2005) described the links between eco-
system services and human well-being using four broad dimensions: security, basic
living material, health, and social relations. Balmford et al. (2008) showed that the
value of ecosystem services is best accounted for when ecosystem services are
translated into benefits to humans (i.e., food, water, material, energy, property,

1
This is reflected in the relatively small number of arid and semi-arid area case studies reviewed
by TEEB (2009).
12 Land Degradation, Poverty and Marginality 183

health, psychological well-being). Ecosystem services as defined by MA (2005)

mostly refer to processes such as nutrient and hydrological cycling, and cannot be
directly valued without incurring the risk of over-estimating (e.g., land restoration
through afforestation can be valued for the environmental regulation services it
provides and also for material benefits such as fuel, construction, or pulp wood).
McGregor and Summer (2009) defined human well-being in the context of devel-
opment as being three dimensional: material, relative, and subjective. These
dimensions are linked to “the resources that a person is able to command…what
needs and goals they are able to meet [and] the meaning that they give to these
goals” (McGregor 2006, 4). Economic assessments of the impacts of LD on human
well-being generally concentrate on the first two points: how LD affects resource
availability and use, and how those effects impact the choices and goals of resource
consumers and society at large.
Several authors have modeled relationships between the state of the environment
and human well-being. One recurrent question has been whether lower levels of
well-being (e.g., measured as limited socioeconomic development or poverty rates)
lead to more or less destructive resource use and management practices. From the
literature it can be concluded that a clear consensus on the relationship between
poverty and LD does not exist. The levels of LD resulting from land use practices
involve complex and dynamic processes and multiple feedback and synergistic
effects that tend to be local, specific, and cannot be analyzed at the global scale.
An effective analysis would require close attention to local circumstances, in terms
of the environment and of socioeconomic characteristics, including institutional
frameworks. Theoretical models of development and resource degradation
help explain how these dynamics may play out in specific settings (Reardon and
Vosti 1995; Grepperud 1997; Bulte and van Soest 2001; Carter and Barrett 2006;
Barbier 2010).

12.2 Conceptual Framework

To investigate the relationships between poverty and LD it is useful to place them in

the broader context of an assessment of LD, including its causes, consequences, and
costs. Figure 12.1 presents a conceptual framework for this broader context. The
rounded boxes represent physical elements of the system under consideration that
have direct relationships, and the black arrows denote the directionality of impact
flow. The proximate causes of LD include: biophysical factors (climatic conditions,
topography, etc.) and unsustainable land management techniques (clearing, over
grazing, etc.). The underlying causes of LD include the policies and institutional or
other socioeconomic factors that determine land management practices. Proximate
and underlying causes may be related to each other, for instance through feedback
loops or synergistic processes, making it difficult to assess the influence of a single
factor. National, international, and local policies and strategies; access to markets;
infrastructure; the availability of appropriate agricultural extension services;
184 N. Gerber et al.

Fig. 12.1 A conceptual framework for the analysis of the economics of land degradation (Based
on Nkonya et al. 2011a)

population density; poverty; and land tenure conditions are all underlying causes
that have been empirically associated with LD. For example, poverty could contribute
to the failure to invest in sustainable land management practices. Similarly policies
that enhance investment in land management, such as payments for ecosystem
services resulting in reforestation of steep slopes can mitigate the proximate causes
of LD (soil erosion on denuded slopes). Increasing population density may lead
to increased investment in land management (Boserup 1965; von Braun et al. 1991;
Tiffen et al. 1994) or to LD (Grepperud 1996) depending on situational factors.
The level and type of LD determine its outcomes or effects—whether on-site or
off-site—on the provision of ecosystem services and the benefits humans derive
from those services. If actions to halt or mitigate LD are taken, the actors involved
should be determined by the causes of LD that need to be addressed, the level of
degradation, and by its effects. Actors can then take action to control or counter the
causes, levels, or effects of LD.
Many ecosystem services are not traded in markets, so the beneficiaries do not
directly pay for those services. The economic concept of externalities refers to the
costs and benefits arising from the production or consumption of goods (or ‘bads’)
and services for which no appropriate compensation is paid (e.g., on- and off-site
environmental effects such as sedimentation, and indirect societal effects such as
emigration and food insecurity). The values of such externalities are not necessarily
12 Land Degradation, Poverty and Marginality 185

considered in land-use decisions, often leading to the undervaluation of land and of

ecosystem services, and to levels of LD that are not socially optimal (the social
optimum is reached when the marginal social costs of LD are equal to the marginal
social costs of action to prevent or rectify it).
The dark square box at the base of Fig. 12.1 represents a cost/benefit analysis
of the human impacts of LD. The light arrows indicate the flow of information
necessary to perform such an analysis. A decrease in the provision of ecosystem
services and their benefits has direct economic costs to humans, such as decreased
food security and increased food prices. In addition LD has indirect negative effects
of economic importance to humans: LD affects land prices and the prices of the
goods produced on it. The impacts of LD on economic sectors directly linked to
land use have indirect, economy-wide effects that are passed to other sectors by
multiplier effects. Direct and indirect effects of LD can broadly affect poverty and
national income, and thus have far-reaching socioeconomic consequences. Ideally
all direct and indirect effects of LD would need to be accounted for in an assessment
of its social and environmental costs. The causal relationships represented in
Fig. 12.1, as well as their magnitudes, are partly determined by a broad set of
institutional arrangements—or the ‘rules of the game’ influencing how actors make
decisions. For example, excessive use of groundwater for irrigation has resulted
from a lack of direct financial costs based on the volume of water used.

12.2.1 The Dynamic Nature of LD

Degradation of an ecosystem may not translate directly or immediately into a

noticeable loss of ecosystem services and benefits. Generally speaking ecosystems
can sustain some level of degradation before reaching a threshold after which
ecosystem function declines (often rapidly) in one or more dimensions (TEEB
2009). The impacts of specific LD processes and of the actions used to mitigate
them become apparent over time in a way that is often nonlinear. For instance, the
impacts of afforestation on nitrogen cycling are clearly time dependent. Similarly,
erosion has a nonlinear impact on crop yields due to the fact that the initial stages of
topsoil depletion represent a greater loss of nutrients and productivity than subse-
quent erosion. Facing such dynamic and nonlinear processes and relationships, it
is important to analyze them accordingly. This requires analysis of the impacts of
changes in LD on the economy, as well as the impacts of changes in socio-economic
factors on LD levels.
In this context the choice of discount rates and the time horizon for the cost-
benefit analysis is crucial. High discount rates indicate a strong time preference for
the present (a 10 % yearly discount rate indicates that a $1 investment today must
provide a $1.1 return in 12 months to justify the initial investment). They tend to
discourage investments that generate long-term benefits and favor those that create
short-term benefits but also significant long-term costs. From a society’s perspective
186 N. Gerber et al.

the latter case is not desirable. From an individual’s perspective it might be a rational
decision to adopt practices that maximize short-term benefits even if they contribute
to LD in the long-term.
Particularly in a poor economy, the basis of land use decisions made by individuals
often reflects their preference for the immediate or short term benefits and is often
much different from what a social planner with a long-term perspective would
adopt. Private individuals’ preference for the short term, conceptually reflected in
high private discount rates, may be due to constrained land management choices
due to poverty, leading to LD rather than sustainable practices, or else due to high
levels of risk aversion in determining land-use practices. In some cases LD, poverty,
and short-term temporal preferences can be interrelated and synergistic. As poverty
increases land use practices are increasingly determined by immediate need, thereby
accelerating LD. The resulting increase in LD reduces ecosystem function and its
ability to provide the services and benefits upon which the land users (and others)
depend, aggravating poverty and thus increasing incentives for basing land-use
decisions on short-term outcomes.
At this stage it is important to stress that such synergy is not a universal or necessary
feature of the poverty-LD nexus. As discussed in this chapter, there is no consensus
in the literature on the general relationships between poverty and LD. Indeed these
deleterious spiral effects can be avoided. An essential dynamic to mitigate in
such a context would be the increasing poverty trend. There are several examples of
schemes that target this dynamic. For instance, guaranteeing secure land tenure may
help induce farmers to adopt long-term approaches to land management. Payments
for ecosystem services can alleviate the impacts of poverty on low investment in
sustainable land management. The success of any such measures and programs will
be dependent on their design and administration.2

12.3 Global Indicators of LD

A number of efforts have attempted to assess LD at the global scale (Nkonya

et al. 2011a). The Global Assessment of Human-induced Soil Degradation
(Oldeman et al. 1991) evaluated the extent and severity of soil degradation across
climate types based on expert judgment, which helped raise awareness about
global soil degradation. In response the World Overview of Conservation
Approaches and Technologies network (WOCAT) initiated efforts to help
prevent and reduce LD through sustainable land management (SLM) practices in
1992 by documenting and evaluating soil and water conservation measures
worldwide. The Land Degradation Assessment in Drylands (LADA) and the

2
See Requier-Desjardins et al. (2011) for a more detailed review of instruments for the management
of externalities that are suitable for preventing or mitigating LD and desertification.
12 Land Degradation, Poverty and Marginality 187

Fig. 12.2 Relationship between change in NDVI and population density (Based on NDVI data
from GLCF 2011 in Nkonya et al. 2011a, population data from CIESIN 2010 in Tucker et al. 2004,
and GDP data from IMF 2010)

Global Land Degradation Assessment (GLADA) projects of the FAO made use
of geographical information systems and remote-sensing data to map LD between
1981 and 2003. More recently the GLADIS effort (Nachtergaele et al. 2010)
assessed global LD using a combination of biophysical and socioeconomic
indicators (biomass, soil, water, biodiversity, economic, and socio-cultural).
Methodological concerns about the LADA, GLADA, and GLADIS efforts are
discussed in Nkonya et al. (2011a).
Most studies use land-cover changes as indicators to approximate LD. Land-cover
change is the most direct and pervasive anthropogenic effect used to determine
trends in land status (Vitousek 1994). Many studies have used the normalized
difference vegetation index (NDVI) and related indices as indicators of changes in
ecosystem productivity and LD. NDVI measures vegetation cover and thus analyses
of changes in vegetation cover based on NDVI may conceal some forms of LD. For
example, the encroachment of weeds and alien species or the effects of carbon
fertilization on degraded lands (Vlek et al. 2010) may be perceived as land improve-
ments on the basis of NDVI analyses. In addition, land-use changes such as the
conversion of agricultural areas into residential areas may be difficult to identify
using NDVI data based analyses. Despite these and other disadvantages (Nkonya
et al. 2011a) NDVI remains one of the most widely used analytical tools for providing
information about the condition of the aboveground biomass based on remote sensing
images, especially at the global scale.
The methods, assumptions, and extent of LD in each of these studies varied and
therefore are not directly comparable to one another (Nkonya et al. 2011a). For a
global assessment of LD and its links to poverty, analyses of remote-sensing images
and geo-referenced data are definitely needed. However, assessment results should
be validated on the ground before they can be considered reliable. In particular, the
empirical results presented in the meta-analysis suggest that some of the links
between population density and LD, and between economic growth and LD (Figs. 12.2
188 N. Gerber et al.

Fig. 12.3 Relationship between GDP and NDVI (Based on NDVI data from GLCF 2011 in
Nkonya et al. 2011a, population data from CIESIN 2010 in Tucker et al. 2004, and GDP data from
IMF 2010)

and 12.3),3 display complex regional differences. For instance, the maps illustrate
that a decrease in NDVI can be associated with either an increase or decrease in
population density in neighboring areas (Fig. 12.2, data from the western Sahara), and
also with either an increase or decrease in GDP (Fig. 12.3, same region). This clearly
supports the idea that targeting population density or poverty alone will not improve
LD. Population density and poverty can be underlying causes of LD, but need not
be depending on their combination with other causal factors and institutional frame-
works (such combinations are not captured in Figs. 12.2 and 12.3).
In order to support SLM globally, cost calculations of global LD are urgently
needed for a comparison with the costs of remedial efforts. Such cost assessments
need to rely on a clear theoretical framework that links the causes of LD and their
consequences in terms of their marginal impact on LD levels. The links between
local poverty, development, and food security; their indirect impacts on global soci-
ety; a clear investigation of the strong regional and national dimensions of LD; and
the extent and severity of LD are essential elements for building this framework.

12.4 The Links Between Poverty and LD

LD is important because of its direct effects on:

• the range of land use activities that people can undertake and the range of services
provided by the land—in other words LD restrains land use choices and options

3
To mitigate the impacts of abnormal years, the baseline and endline for the change in NDVI were
computed at an 8 × 8 km resolution for the periods 1982–1984 and 2003–2006 respectively from GLCF
(2011) and Tucker et al. (2004). The population data was collected at a 0.5° × 0.5° resolution from 1990
to 2005 (CIESIN 2010). The GDP data for each country was from 1982–1984 and 2003–2006.
12 Land Degradation, Poverty and Marginality 189

Fig. 12.4 Possible linkages between poverty, land management, land degradation, and agricultural
productivity, with associated factors. The nature of the linkages followed by (?) are undetermined.
(Nkonya et al. 2008)

• the productivity of land-use activities and ecosystem services, and the economic
returns they generate
• the intrinsic value of land
Degradation affects the economic value of land because such values are based on
the capacity to provide services and benefits. These benefits include not only physical
outputs (e.g., food and resources), but also other services that are beneficial to human
well-being (e.g., recreation). The ethical, philosophical, and cultural dimensions of
ecosystem valuation, as well as the intrinsic value of land, are not part of the scope
of our evaluation of the links between LD and human well-being.
Considering that LD negatively affects people and society both directly and
indirectly, why does it take place and why are investments in SLM not reversing, or
at least halting LD worldwide? Is poverty a major cause of this under-investment,
and if so, through what mechanisms? In this chapter we focused on the case of LD
in developing countries, as there is little doubt that socially sub-optimal LD in the
industrialized world is not an issue of poverty.
The lack of consensus on the relationships between LD and poverty can be
extended to the global context of poverty and natural resource depletion. Nkonya et al.
(2008, 8) attribute this fact to the “complexity and context dependence of the linkages
[and] to the lack of comparable empirical evidence on these issues.” Multiple examples
of the linkages between poverty and LD are illustrated in a figure that leaves no doubt
about the complexity of the issue (Fig. 12.4), and its uncertainty in terms of the
nature of some of the linkages or indeed their existence (Nkonya et al. 2008, 9–10).
190 N. Gerber et al.

The authors also reviewed established empirical links between poverty and LD with
an emphasis on the contradictory nature of the published results, and made a strong
case for the contextual dependency of such results. Furthermore, a concept as broad
as poverty is subject to different interpretations and definitions. The adoption of a
specific definition over another, or a narrow focus on a specific dimension of poverty
(e.g., income level) over another (e.g., access to markets) can lead to contrasting
empirical findings in terms of the links between poverty and LD. That meta-analysis
(Nkonya et al. 2011a) also found that while some studies support the association of
poverty with behavior based on short-term perspectives, other studies found that
such behaviors were not correlated to income levels. Some studies have found a
positive link between high discount rates and lower willingness to pay for conserva-
tion, while other studies have found that poor farmers facing negative shocks prefer
to reduce consumption in order to preserve their assets (including land).
In this section we discuss some of the theoretical treatment of the poverty/LD
interrelationships found in the literature. In economic studies it is now common
practice to use inclusive definitions of poverty that consider physical, human, natural,
and financial limitations, and even access to infrastructure and services, which
should be given priority over restrictive definitions linked to indicators of income or
consumption. Nkonya et al. (2008) adopted a comprehensive definition of poverty
that intended to encompass all matters that constrain the land management
decisions of farmers (access to different forms of capital, infrastructure, markets,
services, education, etc.). This broad definition of poverty is consistent with the
notion of ‘investment poverty’ developed by Reardon and Vosti (1995), and with
more recent work on poverty traps (Carter and Barrett 2006). The latter refers to the
‘assetless’ poor whose situation is characterized by self-perpetuating patterns of
poverty: poor farmers become locked in a dynamic of contributing to environmental
degradation due to a lack of alternatives available to them as a result of a limited
asset base.
Barbier (2010, 647–653) presented a simple poverty trap model in which rural
households hold only two types of capital, land and labor. The latter can be used for
production activities on the land or allocated to external paid labor. In this case a
poverty trap is created by labor market dynamics when wage rates for external
employment fall below the household’s reservation wage (i.e., the maximum wage
at which the household still chooses to allocate zero time for external paid labor).
For instance, when LD forces labor off farms on a large scale (due to decreased
agricultural labor productivity) it may cause excess labor supply for nonagricultural
work. At that stage households will reallocate their labor back to on-farm production
activities, further depleting their land resources and locking themselves into a
poverty-LD trap. The author postulated that such cases are likely when poor people
live on marginal land, which was partially supported by the results of a statistical
analysis (Barbier 2010).
Theoretically and in practice, causal linkages between resource degradation and
worsening poverty have not been established (Nkonya et al. 2008). In fact resource
degradation may be an optimal strategy from the private or household perspective if
other activities and assets can be invested in pursuits that yield higher returns. It is
12 Land Degradation, Poverty and Marginality 191

then a case of substituting different types of capital to maximize household welfare.

As the authors noted, under perfectly functioning market conditions resources are
allocated according to their most profitable uses. Under imperfect conditions such
as uncertainty, however, market and institutional failures may cause poverty and LD
to be closely associated. The form, intensity, and direction of this association may
depend on the nature of the market failure, poverty, or the type of LD.
A theoretical example of the relationship between poverty and environmental deg-
radation that has been much discussed in economics literature is the ‘environmental
Kuznets curve’ (EKC). The EKC postulates that the relationship between income
level and environmental degradation has the shape of an inverted U, that degradation
initially increases with income up to a maximum point and thereafter monotonically
decreases as income increases continue. The main assumption behind the EKC is that
accelerating wealth creation through economic growth encourages the technological
progress necessary to counter environmental degradation. The EKC has been dis-
cussed empirically and theoretically (Bulte and van Soest 2001; Hill and Magnani
2002), and proven empirically for specific types of degradation (e.g., air pollution in
Selden and Song 1994), disproven for others (e.g., deforestation in Koop and Tole
1999), and similarly across regions and countries (Harbaugh et al. 2002). A major
theoretical issue lies in the existence and determination of the ‘turning point,’ i.e., the
income level at which the sign of the relationship changes. Caviglia-Harris et al.
(2009) and Azomahou et al. (2006) further reviewed existing studies that tested the
EKC concept. Caviglia-Harris et al. (2009) also tested for the existence of an EKC
relationship in the context of a comprehensive indicator of environmental degrada-
tion, the ecological footprint, and found no empirical evidence to support it.
Bulte and van Soest (2001) examined the possibility of a (reverse) EKC in a
general theoretical setting that was well-suited to developing countries. Their
approach is interesting in that it did not rely on the usual trade-offs between produc-
tion (or consumption) and degradation at various income-levels (as most studies do).
Rather the resource management decisions of the households indirectly consider
degradation through their allocation of labor time. This set up is particularly suited to
the issue of LD, as farm households can effectively choose whether or not to allocate
time to specific activities in order to preserve their natural capital (land).4 The authors
assumed that households cannot trade labor, but fully allocate it across three activities:
agricultural production, preservation/regeneration of their stock of natural capital,
and leisure. The households maximize their utility under a budget constraint and
simultaneously make decisions on consumption and production (i.e., a non-separable
household model). Environmental degradation is accounted for by inter-temporal
variations in the stock of the natural resource or by the rate of resource extraction,
and changes in income are reflected by changes in the farm output prices. Their inter-
esting results show that under imperfect labor markets, the combined income and
substitution effects of rising prices can be either consistent with the EKC hypothesis
or not, depending on the choice of environmental degradation indicator.

4
Nkonya et al. (2011a) provides a summary of solutions to different types of LD.
192 N. Gerber et al.

Salvati and Carlucci (2010, 3–4) noted that most EKC studies focus on air
pollution and to a lesser extent on deforestation. They only listed a few studies that
examined EKCs in the context of other processes with potential links to LD such
as agricultural land use, land cover change, and farmland conversion. Their study
empirically tested the links between economic drivers (proxied by per capita value
added) and process-specific drivers (crop intensity, irrigation, share of industry in
total product, urban land use, and tourism concentration), and LD vulnerability at
the district level in Italy. Their LD vulnerability index (LVI) is composed of 14
indicators reflecting soil quality, climate, and land use. The authors tested several
specifications of the relationship between per capita value added and LVI, and
overall found a negative relationship between economic growth (increase in
district-level value added) and the LVI over the period 1990–2000. It must be
noted, however, that even though Italy has very significant differences in income
and growth levels across regions, these results cannot be taken as a universal illus-
tration of the existence of an EKC for LD. Given the highly disputed theoretical
underpinning of the EKC, it can only inform policymaking after it has been thor-
oughly tested at appropriate spatial scales.

12.5 Case Studies

We chose Niger and India as case studies because both countries have implemented
land management policies that have had significant positive impacts despite daunting
challenges. Niger is ranked among the poorest countries in the world, yet the country
has developed exemplary land management programs, demonstrating that even the
poorest countries can achieve SLM with appropriate policies. Niger’s GDP per capita
increased from an average of US$475 in 1981–1984 to US$610 in 2003–2006 (based
on purchasing power parity [PPP] (IMF 2011)), while deforestation rates in the
country decreased from 3.7 % in 1990–2000 to 1.7 % in 2000–2010 (FAO 2011a).
Reforested areas increased significantly during the same period and contributed a
large share of the reduction in overall deforestation in the country.
India—the second most populous country in the world—has implemented
community-based resource management efforts with notable successes. Over the
past 20 years India’s economy has been growing fast. Per capita GDP increased
from an average of US$529 in 1981–1984 to US$2,068 in 2003–2006 (based on
PPP and international dollars (IMF 2011)). Despite an increasing population
size and greater demand for forest products, forest cover in India increased by 0.2 %
in 1990–2000 and by 0.55 % in 2000–2010 (FAO 2011a).
As seen in Table 12.1, both countries are mostly arid or semi-arid, the amount of
arable land area per capita is less than 1 ha, and soil sodicity (i.e., a form of salinity
caused by excessive concentrations of sodium in the soil that results from the evapo-
ration of sodium bearing irrigation water and that negatively affects plant growth) is
a common issue. Soil erosion is a larger problem than sodicity in India, primarily
water-induced erosion. In Niger water induced erosion is limited because of less
12 Land Degradation, Poverty and Marginality 193

Table 12.1 Land resources Land resource and severity

and severity of land of degradation India Niger
degradation
Arable land per capita (ha) 0.18 0.44
Arid and semi-arid lands (% of total 72 94
land area)
Soil erosion hazard (% of total land area) 29 7
% of total land area affected by sodicity 1 1
FAO (2010)

Table 12.2 Adoption rates of Practice Nigera Kenyaa

land management practices in
Use of chemical and organic fertilizer 0 33.0
Niger and Kenya
Animal manure application 1.0 68.0
Improved fallow 0.6 4.9
Crop-residue incorporation 0.1 34.4
Mulching 6.4 35.2
Rotational grazing 0.4 7.5
Water harvesting 0.4 17.2
Nkonya et al. (2011a), based on FAO (2011b)
a
Percentage of households using practice

rainfall and relatively flat terrain, however, wind erosion is a major problem there
(Sterk 2003). Soil nutrient depletion, overgrazing, salinity in irrigated cropland, and
deforestation are also major problems in Niger.

12.5.1 Niger

Soil nutrient depletion in Niger is high. On average the country applies only about
0.28 kg of nitrogen per hectare, the second lowest rate in the region. In comparison,
the rate of NPK uptake per hectare for millet—the most common crop in Niger—is
56 kg (Henao and Baanante 2006), suggesting that the rate of soil nutrient mining
is high. As shown in Table 12.2, the use of organic soil fertility management prac-
tices is also limited. Six percent of households used mulching and only 1 % or less
adopted similar land management practices. Underscoring the strong association of
land degradation and poverty, Nkonya et al. (2011b) showed that the use of chemical
fertilizer and organic inputs was lower for poor farmers than for farmers who were
not poor. Similarly it is tempting to conclude that there is an association between
poverty and the lack of adopting certain land management practices with respect to
change in GDP and land management in Niger and Kenya. GDP per capita in Niger
(see section above) grew significantly less between 1981–1984 and 2003–2006
than in Kenya (US$764 to US$1,382 based on PPP and international dollars
194 N. Gerber et al.

(IMF 2011)), and this pattern reflects a much lower rate of adoption of improved
land management practices in Niger than Kenya. However, rural services, institutions,
marketing, and other factors also affect the adoption of certain land management
practices.
Livestock production is the major source of income in the arid and semi-arid
regions of the country and overgrazing is a major problem. The average stocking
rates in Niger (in terms of tropical livestock units [TLU]) increased from about
0.25 TLU/ha in 1992 to a little over 0.4 TLU/ha (Nkonya et al. 2011a).5 That study
also evaluated the effects of overgrazing in Niger and found that overgrazing
reduced forage yields by an average of 32 %, which has led to associated losses in
beef carcass weights and milk production.
Costs of action and inaction. Nkonya et al. (2011a) estimated the costs of action and
inaction at the farm level. The costs of action are the costs a farmer will incur by
engaging in efforts to address LD, whereas the costs of inaction are the losses the
farmer will incur due to the effects of LD if no remedial actions are taken. In the
case of soil nutrient depletion, the study evaluated sorghum, millet, and rice produc-
tion and estimated the loss of profits due to exclusive use of passive crop residues as
a soil fertility management practice. To determine the costs of inaction, sole reliance
on passive crop residue was compared with the use of: 40 kg of nitrogen per hectare,
1.67 tons of manure per hectare, and the active incorporation of 50 % of crop
residues. The results indicated annual national level profit losses due to exclusive
reliance on passive crop residues reaching US$6 million for rice, US$133 million
for millet, and US$157 million for sorghum. The availability of fertilizers and
appropriate training in fertilizer application are limiting factors in the case of poor
farmers. Well known systems exist to overcome such barriers (e.g., contract farming,
subsidies, extension services, etc.), their success in terms of using SLM practices
and poverty reduction being somewhat case specific.
In the case of sodicity the costs of action are the costs of the water and labor
required for leaching. The cost of inaction is the value lost due to sodicity. This cost
is estimated by determining the difference between the net present value (NPV) of
leaching practices and the NPV without desalinization. The study showed that the
costs of action were only about 10 % of the costs of inaction per hectare, indicating
the high costs that farmers suffer for not addressing sodicity problems, as well as
the importance of LD costs in terms of income and annual GDP loss (8 %) due to
salinity. Nonetheless poverty may prevent farmers from taking action, especially if
the water and labor costs for leaching are up-front costs. Without financial support
poor farmers are left to suffer decreasing yields and benefits, thus increasing the risk
of poverty and further LD.
Land management success stories in Niger. Niger is one of the few Sub-Saharan
African countries that has achieved remarkable land rehabilitation results. The

5
Tropical livestock units (TLU) are based on a standard animal live weight of 250 kg. The average
TLU conversion factor for common livestock are 0.9 TLU for individual adult cattle and 0.2 TLU
for goats and sheep (Defoer et al. 2000).
12 Land Degradation, Poverty and Marginality 195

government and development partners have invested in land management programs

because the majority of the population depends heavily on the land. The ‘Special
Program of the President,’ Projet de Gestion des Ressources Naturelles (Natural
Resources Management Program), and more than 50 other programs have been
implemented by the national government, NGOs, and donors since the early 1980s
(World Bank 2009). In addition to these efforts the government has also restructured
its institutions and passed a rural code in 1993 that gave traditional leaders more
power to manage land, and encouraged them to engage in reforestation efforts and
allows them to benefit from such efforts without government intervention. Allowing
the poor to benefit from conservation actions was a key incentive for active land
management. For instance, the forest policy gave landholders tenure rights to the
trees that they planted or protected (Yatich et al. 2008; World Bank 2009). In addition
the government promoted contract farming in state-controlled forests (Yatich et al.
2008). These policies created a sense of proprietorship and helped provide the
economic incentives needed to protect the forests. The sales and subsistence use of
forest products also helped farmers cope with agricultural production risks.
These policies and investments have led to the significant recovery of forest
cover in the Sahelian regions where they were implemented. For example, villages
where the Projet Intégré Keita; Projet de Développement Rural de Maradi operated
were found to be much greener than what could be explained by change in rainfall
only (Herrmann et al. 2005; Adam et al. 2006; Reij et al. 2008). In total, tree planting
and protection efforts have led to the restoration of forest cover on three million
hectares (Adam et al. 2006).
Other important factors contributed to this remarkable success. The droughts that
occurred during the 1970s and 1980s led to a loss of vegetation that increased the
perceived value of shade produced by trees. The loss of vegetation also increased
the difficulty of collecting firewood and reduced forage available to the livestock
sector, especially in northern Niger where trees provide fodder during the dry season.
Hence tree scarcity significantly affected livelihoods in rural communities, prompting
residents to shift from land-clearing to reforestation. The national government of
Niger also responded to this challenge. In the 1970s it started to aggressively pro-
mote reforestation and changed Independence Day to National Tree Day.
NGOs and religious organizations also helped significantly in building the capacity
of local institutions to manage natural resources and mobilized communities to plant
and protect trees. For example, the farmer-managed natural regeneration (FMNR)
program—in which communities protect or plant new trees in order to benefit from
fuel wood, fodder, nitrogen fixation from leguminous trees, reduced erosion due
to windbreaks, and other ecosystem benefits—was initiated by a religious organiza-
tion (Reij et al. 2008). The authors in that study estimated that villages working with
FMNR had 10–20 times more trees than before participating in FMNR. Consistent
with Bai et al. (2008a), higher tree densities were found in villages with higher
population densities (Reij et al. 2008).
The lessons that can be drawn from Niger are the institutional (rural code) and
grassroots (NGOs and religious organizations) linkages. These gave local communities
the mandate and the capacity to cultivate and manage natural resources. The rural
196 N. Gerber et al.

Table 12.3 Trends of total factor productivity growth of major crops in India (Kumar and Mittal
2006)
Annual TFP Annual TFP
Crop Period Declining TFP growth <1 % growth >1 %
Rice 1971–1986 30.5 25.9 43.6
1987–1900 15.0 32.8 52.2
Wheat 1971–1986 10.3 17.3 72.4
1987–2000 2.8 74.7 22.5
Coarse cereal 1971–1986 19.8 9.6 70.5
1987–2000 60.2 9.8 30.1
Pulse bean 1971–1986 42.8 36.6 20.5
1987–2000 69.2 26.6 4.2
Oilseed 1971–1986 35.6 18.3 46.1
1987–2000 28.3 10.6 61.1
Sugarcane 1971–1986 20.3 61.0 18.6
1987–2000 90.9 5.4 3.7
Fiber 1971–1986 53.8 7.2 39.0
1987–2000 32.5 1.4 66.1
Vegetables 1971–1986 0.0 27.5 72.5
1987–2000 27.5 0.0 72.5

code provision, which allowed communities to benefit from their reforestation

efforts also created strong incentives for farmers to invest their limited resources in
land management, which alleviated poverty and resource degradation.

12.5.2 India

The nature of land degradation in India is different from what is found in Sub-
Saharan African countries. India is among the countries that benefited from the
Green Revolution. Agricultural productivity in India has been generally improving
due to the increasing use of fertilizers and improved crop varieties. The total factor
productivity (TFP) of more than 50 % of the major crops increased from 1970 to
2000 due to the increasing use of fertilizer and other inputs (Table 12.3).
Overexploitation of groundwater is one of India’s major environmental problems.
Approximately 63 % of total cereal production is irrigated and groundwater
accounted for 45 % of the 567 km3 of irrigation water used in 2000 (Kumar et al.
2005; de Fraiture et al. 2008). Wheat and rice are the major irrigated cereals; the
other cereal crops are largely rain fed (Kumar et al. 2005). Soil salinity is also
becoming an increasing problem in areas where crops are irrigated, affecting an
estimated 2 % of the cropland in India (FAO 2010). Salinity can reduce crop rice
yields by up to 22 % (Nkonya et al. 2011a). The costs of action for mitigating LD in
this case include the costs of desalinization, which involves staggered leaching
of salts. The annual cost of irrigation water in India varies from US$0.0 to as high
12 Land Degradation, Poverty and Marginality 197

as US$470/ha in Gujarat (Cornish et al. 2004). The costs of desalinization in Gujarat

were estimated to be about US$127/ha (Nkonya et al. 2011a). As seen in Niger, the
costs of action are less than the costs of inaction, suggesting that inaction is not
due to the lack of an economic incentive. Shouldering these costs may simply not
be possible for many farmers, who may eventually be forced to abandon their
land and move to other (marginal) areas.
Three land management success stories in India. India has had mixed success with
community-based watershed management strategies. A study in Tamil Nadu evalu-
ated the impact of community-based watershed management through Panchayati Raj
(traditional governance) institutions, local user groups, and NGOs. Results showed that
community-based watershed management efforts in Tamil Nadu have raised water
tables, improved the reliability of wells, and increased the availability of water for
agriculture and domestic use (Kuppannan and Devarajulu 2009). These findings are
consistent with other studies that have found cases of successful community-based
natural resource management in India and elsewhere (Ostrom and Nagendra 2006;
Kerr 2007). The examples in India illustrate the importance of participatory and
bottom-up approaches that place natural-resource management responsibilities into
the hands of local institutions and communities. A review by Darghouth et al. (2008)
found that participatory watershed management was successful when the programs
were of common interest to the community, were flexible, and offered mechanisms
for capacity building and empowerment of local communities. Community-based
watershed management efforts have not been effective in managing larger watersheds
(Darghouth et al. 2008), or where culturally or economically diverse communities are
involved (Kerr 2007). These findings suggest the need for creating well-coordinated
linkages that address complex watershed management scenarios.
India is one of a few countries that have seen significant improvements in rain
fed agriculture performance (Bai et al. 2008b), due in part to successful agroforestry
and renewable energy programs. Such improvements are evidence of the potential
of SLM practices for combating LD and improving agricultural productivity. A
contributing factor to the increased NDVI values in rain fed agricultural areas of
India is the significant development of agroforestry, which is a traditional practice
(Pandey 2007). Agroforestry plantations cover approximately 17 million hectares in
India (Pandey 2007), equivalent to about 10 % of the country’s agricultural area
(FAO 2008). India is also one of the leading producers of jatropha, a multi-purpose
crop often used as a biofuel that can grow on highly-degraded soils and in arid areas.
Jatropha cultivation has helped reclaim 85,000 ha of degraded land in northern India
(ICRAF 2008). Jatropha production on highly degraded lands has helped alleviate
poverty, beneficiaries of a project in northern India earned an average of US$1,200
annually per producer from sales of jatropha seed only 3 years after initial investments.
Targeting degraded lands is one of the key features of this project, which has the
potential to reclaim up to 30 million hectares of severely degraded land in India
(ICRAF 2008).
India enacted the National Rural Employment Guarantee Act in 2006. Under this
social protection act participants are guaranteed employment for at least 100 days
198 N. Gerber et al.

per year (UNEP 2011). About 84 % of the public works executed under this program
have addressed water conservation, irrigation, and land management investments.
It is estimated that the program has provided three billion workdays and benefited
58 million households (UNEP 2011). Even though there have been challenges with
such programs in India and elsewhere (Deshingkar et al. 2005) they have shown to
be win-win public investments, creating employment, reducing poverty, and enhancing
the land and water resources (UNEP 2011).

12.6 Conclusion

To close this chapter we want to re-emphasize the complexity and plurality of the
types and magnitudes of the causal relationships between poverty and environmental
degradation. The wide array of relationships between LD and poverty were confirmed
by literature review, where it is difficult to find an empirical based consensus on the
nature of the links between the two. The breadth of the nature of poverty–LD linkages
can be explained by several factors. First, there are several feedback and synergistic
effects between the different causal factors of LD, therefore the true impact of poverty
on LD cannot be captured in a bivariate analysis. Second, there are feedback effects
between poverty and LD that make it harder to determine the direction of the causality.
Finally, the institutional framework within which land users operate and make their
land management decisions can supersede the impacts of poverty on LD. This was
clearly exemplified by case studies in Niger and India. All of these empirical findings
were shown to be independent of the cost effectiveness of SLM practices (Nkonya
et al. 2011a), and are echoed in the theoretical literature on poverty and growth and
their links to environmental degradation. The existence of an EKC has been debated
at length. Bulte and van Soest (2001) showed that under imperfect labor markets,
the combined income and substitution effects of rising resource prices can lead to
an allocation of farm labor that is either consistent with the EKC hypothesis or not,
depending on the choice of environmental degradation indicator.
Poverty as well as LD can be accounted for using a wide variety of indicators, the
choice of which will influence the understanding gained about any particular linkage,
which can differ vastly across regions and countries. A global and inclusive definition
of poverty is inevitably linked to a general concept of well-being and welfare. There
are currently no established indicators at the global-level that fully capture the
variability in human well-being. Well-being is partially attributable to the state of
the environment. This can be captured to some extent by global indicators such as
genuine savings (Aglietta 2010), which adjusts the savings rate for the depletion of
natural resources in particular. However, as noted in Nkonya et al. (2008), LD is a
form of ‘squandering’ that is not addressed by any current frameworks and indicators
of well-being, not even in measures of genuine savings. This needs to change, because
productive land is a global public good that is increasingly under pressure to deliver
the required ecosystem benefits to support a growing world population, sparking
increasing competition over its ownership (Deininger and Byerlee 2011). A first
12 Land Degradation, Poverty and Marginality 199

step towards the inclusion of LD into general measures of well-being that truly
reflect the state of the environment is a systematic and science-based assessment of
the extent, severity and costs of LD worldwide (Nkonya et al. 2011a).

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Adam T, Reij C, Abdoulaye T, Larwanou M, Tappan G (2006) Impacts des Investissements dans la
Gestion des Resources Naturalles (GRN) au Niger: Rapport de Synthese. Niamey. Centre
Régional d’Enseignement Specialisé en Agriculture (CRESA)/The Freedom University of
Amsterdam, Niamey/Amsterdam
Aglietta M (2010) Sustainable growth: do we really measure the challenge? Paper presented at
the eighth Agence Française de Développement, European Development Research Network
conference, Paris, 1 December 2010
Azomahou T, Laisney F, Van Nguyen P (2006) Economic development and CO2 emissions: a non-
parametric panel approach. J Public Econ 90(6–7):1347–1363
Bai ZG, Dent DL, Olsson L, Schaepman ME (2008a) Proxy global assessment of land degradation.
Soil Use Manag 24(3):223–234
Bai ZG, Dent DL, Olsson L, Schaepman ME (2008b) Global assessment of land degradation and
improvement. 1. Identification by remote sensing. GLADA report 5 (November). Report 2008/01.
International Soil Reference and Information Center, University of Wageningen, Wageningen
Balmford A, Rodrigues A, Walpole M, ten Brink P, Kettunen M, Braat L (2008) The economics of
biodiversity and ecosystems: scoping the science. European Commission, Cambridge
Barbier E (2010) Poverty, development, and environment. Environ Dev Econ 15:635–660
Boserup E (1965) The conditions of agricultural growth: the economics of agrarian change under
population pressure. Aldine Press, New York
Bulte EH, van Soest DP (2001) Environmental degradation in developing countries: households
and the (reverse) environmental Kuznets curve. J Dev Econ 65:225–235
Carter MR, Barrett CB (2006) The economics of poverty traps and persistent poverty: an asset-based
approach. J Dev Stud 42(2):178–199
Caviglia-Harris JL, Chambers D, Kahn JR (2009) Taking the “U” out of Kuznets: a comprehensive
analysis of the EKC and environmental degradation. Ecol Econ 68(4):1149–1159
CIESIN Center for International Earth Science Information Network (2010) Global distribution of
poverty. Center for International Earth Science Information Network, New York. http://sedac.
ciesin.columbia.edu/povmap/ds_info.jsp. Accessed 3 Aug 2012
Cornish G, Bosworth B, Perry C, Burke J (2004) Water charging in irrigated agriculture. An analysis
of international experience, FAO water report 28. Food and Agriculture Organization of the
United Nations, Rome
Darghouth S, Ward C, Gambarelli G, Styger E, Roux J (2008) Watershed management approaches,
policies, and operations: lessons for scaling up, Water sector board discussion paper 1. World
Bank, Washington, DC
De Fraiture C, Giordano M, Liao Y (2008) Biofuels and implications for agricultural water use:
blue impacts of green energy. Water Policy 10(1):67–81
Defoer T, Budelman A, Toulmin C, Carter S (2000) Building common knowledge: participatory
learning and action research. In: Defoer T, Budelman A (eds) Managing soil fertility in the
tropics. A resource guide for participatory learning and action research. Royal Tropical
Institute, Amsterdam, pp 23–35
200 N. Gerber et al.

Deininger K, Byerlee D (2011) Rising global interest in farmland: can it yield sustainable and
equitable benefits? World Bank, Washington, DC
Deshingkar P, Johnson C, Farrington J (2005) State transfers to the poor and back: the case of the
food-for-work program in India. World Dev 33(4):575–591
Ellis EC, Ramankutty N (2008) Putting people in the map: anthropogenic biomes of the world.
Front Ecol Environ 6:439–447
FAO (2008) FAOSTAT Online agricultural data. Food and Agriculture Organization of the United
Nations, Rome. http://faostat.fao.org. Accessed 25 June 2012
FAO (2010) Land resource potential and constraints statistics at country and regional level
(TERRASTAT). Food and Agriculture Organization of the United Nations, Rome. http://www.
fao.org/ag/agl/agll/terrastat. Accessed 25 June 2012
FAO (2011a) State of the world’s forests. Food and Agriculture Organization of the United
Nations, Rome
FAO (2011b) FAOSTAT Online agricultural data. Food and Agriculture Organization of the United
Nations, Rome. http://faostat.fao.org. Accessed 25 June 2012
GLCF (2011) Global inventory modeling and mapping studies (GIMMS). Global land cover facility,
University of Maryland, College Park. http://glcf.umiacs.umd.edu/data/gimms. Accessed 26
June 2012
Grepperud S (1996) Population pressure and land degradation: the case of Ethiopia. J Environ Econ
Manag 30(1):18–33
Grepperud S (1997) Poverty, land degradation and climatic uncertainty. Oxf Econ Pap 49(4):586–608
Harbaugh WT, Levinson A, Wilson DM (2002) Rethinking the empirical evidence for an environ-
mental Kuznets curve. Rev Econ Stat 84(3):541–551
Henao J, Baanante C (2006) Agricultural production and soil nutrient mining in Africa: implications
for resource conservation and policy development. International Fertilizer Development
Center, Muscle Shoal
Herrmann S, Anyamba A, Tucker CJ (2005) Recent trends in vegetation dynamics in the African
Sahel and their relationship to climate. Glob Environ Chang 15(4):394–404
Hill RJ, Magnani E (2002) An exploration of the conceptual and empirical basis of the environmental
Kuznets curve. Aust Econ Pap 41:239–254
ICRAF (2008) Agroforestry for food security and healthy ecosystems. World agroforestry center
annual report. International Center for Research in Agroforestry, World Agroforestry
Center, Nairobi
IMF (2010) World economic outlook data. International Monetary Fund, Washington, DC. http://
www.imf.org/external/pubs/ft/weo/2010/02. Accessed 27 June 2012
IMF (2011) World economic outlook data. International Monetary Fund, Washington, DC. http://
www.imf.org/external/pubs/ft/weo/2011/01/weodata/index.aspx. Accessed 19 June 2012
Kerr J (2007) Watershed management: lessons from common property theory. Int J Common
1(1):89–110
Koop G, Tole L (1999) Is there an environmental Kuznets curve for deforestation? J Dev Econ
58:231–244
Kumar P, Mittal S (2006) Agricultural productivity trends in India: sustainability issues. Agric
Econ Res Rev 19:71–88
Kumar R, Singh RD, Sharma KD (2005) Water resources of India. Curr Sci 89(5):794–811
Kuppannan P, Devarajulu SK (2009) Impacts of watershed development programmes: experiences
and evidences from Tamil Nadu. Agricultural economics research review 22 (conference
number), pp 387–396. http://mpra.ub.uni-muenchen.de/18653. Accessed 22 June 2012
MA (2005) Ecosystems and human well-being—a framework for assessment. Millennium
ecosystem assessment. Island Press, Washington, DC
McGregor JA (2006) Researching wellbeing: from concepts to methodology. WeD research paper
20. ESRC Research Group on Wellbeing in Developing Countries, Bath. http://www.welldev.
org.uk/research/workingpaperpdf/wed20.pdf. Accessed 25 July 2012
12 Land Degradation, Poverty and Marginality 201

McGregor JA, Summer H (2009) After 2015: ‘3D human wellbeing’, Focus policy briefing 9.
Institute for Development Studies, Brighton
Nachtergaele F, Petri M, Biancalani R, van Lynden G, van Velthuizen H (2010) Global land
degradation information system (GLADIS), Beta version, an information database for land
degradation assessment at the global level, land degradation assessment in drylands technical
report 17. Food and Agriculture Organization of the United Nations, Rome
Nkonya E, Pender J, Kaizzi KC, Kato E, Mugarura S, Ssali H, Muwonge J (2008) Linkages between
land management, land degradation, and poverty in Sub-Saharan Africa: the case of Uganda,
IFPRI research report 159. International Food Policy Research Institute, Washington, DC
Nkonya I, Gerber N, Baumgärtner P, von Braun J, De Pinto A, Graw V, Kato E, Kloos J, Walter T (2011a)
The economics of land degradation: toward an integrated global assessment. Peter Lang, Essen
Nkonya E, Place F, Pender J, Mwanjololo M, Okhimamhe A, Kato E, Crespo S, Ndjeunga J, Traore
S (2011b) Climate risk management through sustainable land management in Sub-Saharan
Africa, IFPRI discussion paper 01126. International Food Policy Research Institute,
Washington, DC
Oldeman LR, Hakkeling RT, Sombroek G (1991) Status of human-induced soil degradation, 2nd edn.
International Soil Reference and Information Center, University of Wageningen, Wageningen
Ostrom E, Nagendra H (2006) Insights on linking forests, trees, and people from the air, on the
ground, and in the laboratory. Proc Natl Acad Sci USA 103(51):19224–19231
Pandey D (2007) Multifunctional agroforestry systems in India. Curr Sci 92(4):455–463
Reardon T, Vosti S (1995) Links between rural poverty and the environment in developing countries:
asset categories and investment poverty. World Dev 23(9):1495–1506
Reij C, Tappan G, Smale M (2008) Re-greening the Sahel: farmer-led innovation in Burkina Faso
and Niger. In: Spielman D, Pandya-Lorch R (eds) Millions fed. Proven successes in agricultural
development. International Food Policy Research Institute, Washington, DC, pp 53–58
Requier-Desjardins M, Adhikari B, Sperlich S (2011) Some notes on the economic assessment of
land degradation. Land Degrad Dev 22:285–298
Reynolds J, Stafford Smith M, Lambin E, Turner BL II, Mortimore M, Batterbury SPJ, Downing
TE, Dowlatabadi H, Fernández RJ, Herrick JE, Huber-Sannwald E, Jiang H, Leemans R, Lynam
T, Maestre FT, Ayarza M, Walker B (2007) Global desertification: building a science for
drylands development. Science 316:847–851
Salvati L, Carlucci M (2010) Economic development as a determinant of desertification risk:
exploring a local-EKC hypothesis. Paper prepared at the 31st general conference of The
International Association for Research in Income and Wealth, St. Gallen, Switzerland, 5–11
August 2010. http://www.iariw.org/papers/2010/poster1Salvat.pdf. Accessed 30 July 2012
Selden TM, Song D (1994) Environmental quality and development: is there an environmental
Kuznets curve for air pollution emissions? J Environ Econ Manag 27:147–162
Sterk G (2003) Causes, consequences and control of wind erosion in Sahellan Africa: a review.
Land Degrad Dev 14:95–108
TEEB (2009) The economics of ecosystems and biodiversity. An interim report. A Banson
Production, Cambridge. http://www.teebweb.org/Portals/25/Documents/TEEB-InterimReport-
English.pdf. Accessed 30 July 2012
Tiffen M, Mortimore M, Gichuki F (1994) More people, less erosion: environmental recovery in
Kenya. Wiley, London
Tucker CJ, Pinzon JE, Brown ME (2004) Global inventory modeling and mapping studies
(GIMMS). Global land cover facility. University of Maryland, College Park
UNEP (2011) Green economy. Pathways to sustainable development and poverty eradication.
A synthesis for policy makers. United Nations Environment Programme, Nairobi
Vitousek PM (1994) Beyond global warming: ecology and global change. Ecology 75:1861–1876
Vlek PL, Le QB, Tamene L (2010) Assessment of land degradation, its possible causes, and threat
to food security in Sub-Saharan Africa. In: Lal R, Stewart BA (eds) Food security and soil quality.
CRC Press, Boca Raton, pp 57–86
202 N. Gerber et al.

von Braun J, de Haen H, Blanken J (1991) Commercialization of agriculture under population

pressure: effects on production, consumption, and nutrition in Rwanda, Research report 85.
International Food Policy Research Institute, Washington, DC
World Bank (2009) Impacts of sustainable land management programs on land management and
poverty in Niger, World Bank country report 48230-NE. World Bank, Washington, DC
Yatich T, Kalinganire A, Alinon K, Weber JC, Dakouo JM, Samaké O, Sangaré S (2008) Moving
beyond forestry laws in Sahelian countries. Policy brief. World Agroforestry Center, Nairobi
 Part IV
Experiencing Marginality
in Africa and Asia
Chapter 13
Tackling Social Exclusion and Marginality
for Poverty Reduction: Indian Experiences

Sukhadeo Thorat

Abstract This chapter examines changes in poverty in combination with changes in

income and the character of income growth for multiple socio-religious groups in
India. The extent to which income growth has been pro-poor was also evaluated.
Income growth was compared between agricultural and non-agricultural livelihoods,
between rural and urban areas, and across ethnic, caste, and religious groups. The
analysis found that poverty was reduced at a lower rate for Scheduled Tribes,
Scheduled Castes, and Muslims, who suffer from social exclusion and discrimina-
tion, than for the rest of the society. These groups have a history of high levels of
poverty in India, and compared to mainstream society members these groups typically
own less agricultural land, have less access to private non-agricultural economic
activities, and are more dependent on wage employment.

Keywords Poverty • Social exclusion • Marginality • Pro-poor growth • India

13.1 Introduction

Empirical evidence on changes in poverty has indicated that “poverty reduction has
most benefitted people living close to the poverty line rather than those at the very
bottom of income distribution” (von Braun et al. 2009, 5). The most extreme and
persistently poor generally have common features, such as a lack of assets and
education, but they also often belong to certain social groups—typically groups
distinguished by race, color, social origin (caste), religion, or geographical location.
They suffer from chronic poverty that is often passed on to consecutive generations.

S. Thorat (*)
Centre for the Study of Regional Development, School of Social Sciences,
Jawaharlal Nehru University, New Delhi, India
e-mail: [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 205
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_13,
© The Author(s) 2014
206 S. Thorat

While poverty is closely associated with the economic condition of the poor, the
persistence of poverty is a feature particularly linked with marginality.
We have a fairly good basis of understanding the economic factors associated
with poverty and hunger (von Braun et al. 2009). Extreme poverty is caused in
part by limited ownership or access to agricultural land or business opportunities,
the potential to develop education and skills, and stable employment. Poverty is
also caused by the lack of participation in governmental control over poverty
alleviation efforts. Therefore, the poor are frequently landless or are small-produc-
ers or farmers in marginal areas for agriculture, or are wage-labor dependent, or
have limited skills and education, or are unemployed or else are engaged in low-
income occupations. Within the general category of the poor are people belonging
to certain social groups that generally suffer more from intense poverty than their
societal counterparts, indicating that in addition to economic factors there are
often constraints associated with social or cultural identity that contribute to their
condition (Thorat 2010).
Presumably there are specific reasons for higher rates of poverty among certain
social groups. Although it may appear that the factors that cause poverty are similar
for both the poor in general and certain social groups, the “channels of causation”
that determine higher poverty rates for certain social groups are different in some
respects. Studies have found that in these cases poverty is often closely linked with
social exclusion—the social processes through which some groups are denied equal
access to sources of income, employment, education, and participation in decision-
making processes. Groups that are socially excluded are not like the rest of the poor.
They are also disadvantaged by unavoidable circumstances and as a result are
deprived of many of the benefits of socio-economic development. Social exclusion
deprives people of choices and opportunities to escape from poverty and denies
them a voice to claim their rights (Thorat et al. 2005). Poverty outcomes are greatly
affected by the exclusion of women, ethnic, religious, indigenous, and caste groups
based on social norms, and values and customs within families, communities, or
markets. Sen (2000, 28–29) argued that “unfair exclusion and/or unfair inclusion”
(access based on discriminatory terms and conditions) with respect to opportunities
reduces entitlement and the capability to enhance personal well-being, thus aggra-
vating poverty among the socially excluded groups.
Drawing from the Indian experience, in this chapter I discuss changes in poverty
among multiple social groups. I have attempted to bring some insight on the possible
linkages between social exclusion and marginalization among certain ethnicities,
castes, and religious groups. India is in many ways a unique case with a high diversity
of social groups identified by ethnicity, caste, or religion for which the government
has developed specific policies, making it highly suitable for an analysis of social
marginalization. I examined changes in poverty in combination with changes in
income and the character of income growth for multiple socio-religious groups.
Specifically, I examined to what extent income growth has been “pro-poor.” The
degree of pro-poor income growth was compared between agricultural and non-
agricultural livelihoods, between rural and urban areas, and across ethnic, caste, and
religious groups.
13 Tackling Social Exclusion and Marginality for Poverty Reduction… 207

13.2 Some Insights from Literature on Poverty,

Social Exclusion, and Growth

13.2.1 Growth, Inequality, and Poverty Linkages

The first systematic treatment of the relationship between growth and poverty was
made by Kuznets (1955), who argued that long-term secular behaviors of inequality
follow an inverted U-shaped pattern, with inequality increasing during the early
stages of growth in developing countries and then decreasing after some time.
Empirical studies that followed Kuznets’ pioneering work found evidence that in
addition to growth in per capita income, income distributions are also determined by
other socio-economic factors such as: population growth rates, income (or worker)
shares in non-agricultural economic sectors, urbanization levels, education, and
government interventions (Adelman and Morris 1973; Paukert 1973; Ahluwalia
1976a, b; Papanek 1978; Tsakloglou 1988). Tsakloglou (1988) found that the rate
of population growth was positively related to income inequality, whereas educa-
tional levels, the extent of government activity, and rates of GDP growth per capita
had negative relationships with inequality. Other studies have found that while the
share of non-agricultural sectors in overall economies is important, increases in
employment in more productive sectors and the productivity of traditional sectors
are critical for reducing poverty (Paukert 1973; Ahluwalia 1976a, b; Kraay 2004;
Hull 2009).
Research by Ravallion (2001, 2009; Ravallion and Datt 2002; Ravallion and
Chen 2003) brought further insight into the relationships between economic
growth, inequality, and poverty, but with distinct results. For 80 countries dur-
ing most of the 1980–2000 period, Ravallion (2009) reported that there was
little or no correlation between changes in inequality and rates of economic
growth, and concluded that growth tended to be roughly distribution neutral.
The authors also found that the rates of poverty decline tended to be less pro-
poor in countries where initial inequality was higher compared to those where it
was lower.
In Ravallion’s view (2009), certain inequalities not only generate higher poverty in
the immediate term, but also impede future growth and poverty reduction, including
social exclusion, discrimination, travel restrictions, constraints on human develop-
ment, lack of access to financial and insurance support, and corruption. All of these
inequalities perpetuate poverty by limiting the prospects for economic advancement
among certain segments of the population. “More rapid poverty reduction would
require more growth, a more pro-poor pattern of growth, and success in reducing the
antecedent inequalities that limit poor people’s access to economic opportunities”
(Ravallion and Chen 2003, 185). These insights indicate that participation in growth
and poverty reduction are dependent on a number of factors that include not only
growth in per capita income, but also a host of socio-economic inequalities that limit
economic opportunities for the poor.
208 S. Thorat

13.2.2 Pro-poor Growth

The concept of pro-poor growth has been a subject of discussion in the context of
“inclusive growth.” Ravallion and Chen (2003, 94) argued that “pro-poor growth is
any growth in the mean income that benefits poor.” Kakwani argued that pro-poor
growth has limitations as “it would encompass the vast majority of growth episodes
so long as poverty declines” (Kakwani 2000, 68). It is therefore argued that for
growth to be pro-poor, it should benefit the poor proportionately more than the
non-poor.
This brings the focus on the distribution of outcomes of growth among the poor
and non-poor. In this context a distinction is often drawn between strongly pro-poor
and weakly pro-poor impacts of growth, the former relates to situations in which
incomes rise proportionally faster for the poor than for the non-poor, and the latter
for a situation in which growth benefits the poor considerably less than the non-
poor (Lipton 1991). Moreover, the tradeoffs between changes in absolute poverty
and inequality between rich and poor are meaningful (Grinspun 2009). What
happens to inequality in income distribution and the distribution of income around
the poverty line in the process of economic growth are important determinants of
the relative “pro-poor” qualities of growth (and not the distribution of income
among the upper classes).
Osmani (2005) took the definition a step further. He argued that the true test of
being pro-poor is the existence of policies biased in favor of the poor with reference
to the country’s past record of poverty reduction, and defined pro-poor growth as
growth processes that reduce poverty more than past policies. The benchmarks in
this case will be country specific.
To overcome the limitation of Osmani’s criterion, three alternative criteria of
“pro-poorness” have been proposed: (1) that the share of income growth for the
poor exceeds their existing share, (2) that the poor’s share in incremental income
surpasses their share in the population, and (3) that the share of the poor in incre-
mental growth exceeds some international norm (Grinspun 2009). Thus, pro-poor
growth is achieved when the poorest of the poor benefit from the growth and not just
the poor nearer to established poverty lines. To the extent that the pro-poor qualities
of growth involve not only the reduction of absolute poverty, but also its rate of
change for poor and the share of the poor in incremental income gains, it really
focuses on the tradeoffs between changes in poverty and inequality in income
distribution in the process of growth. A set of comprehensive criteria for “pro-poor”
growth include:
• achieves absolute reduction in poverty
• benefits the poor proportionately more than the non-poor
• raises income proportionally faster for the poor than the non-poor
• reduces poverty beyond some country specific benchmark
• benefits the poorest and not just the poor near poverty lines
• increases the income of the poor in excess of their share of the population
• increases the income of the poor beyond some international norm
13 Tackling Social Exclusion and Marginality for Poverty Reduction… 209

13.2.3 Social Exclusion and Persistent Poverty

Literature on poverty and socially excluded groups has identified that factors such
as social exclusion and discrimination perpetuate persistent poverty. There is
extremely limited empirical work on how social exclusion and discrimination
contribute to poverty among excluded groups. We have limited understanding about
the processes involved in “exclusion induced poverty.” Social exclusion in general,
and particularly socioeconomic exclusion, are the processes through which groups
are denied access to rights or economic and social participation in society. Exclusion
may occur directly (unfavorable exclusion), or through deliberate government
policies (active exclusion), and through unintended circumstances (passive exclusion),
or through the inability of some social groups to respect the rights of others (consti-
tutive relevance) (Sen 2000, 28–29). Social exclusion aggravates poverty directly by
denying access to opportunities channeled through market and non-market transac-
tions, and indirectly by adversely affecting economic growth. In so far as exclusion
and discrimination are involved in the denial of access to resources, employment,
education, and common facilities certainly impoverish the lives of individuals
belonging to excluded groups. Addressing “exclusion induced poverty” will require
policies that provide safeguards against market and non-market discrimination of
excluded groups. To base policies on empirical insights, it is necessary to better
understand how market and non-market discrimination aggravate poverty both
directly and indirectly.

13.3 Poverty Changes in India by Socio-religious

Groups 1983–2005

The incidence of household poverty and changes therein between rural and urban
areas were examined for multiple socio-religious groups during the period from
1983–1984 to 2004–2005 using a simple measure of poverty called the Head Count
Ratio (HCR). For each group the poverty level was also examined by livelihood
source. In rural areas these included: “self-employed in agriculture” (SEAG) or
farmers, “self-employed in non-agriculture” (SENA), “agricultural wage labor”
(AGLA), and “non-agricultural wage labor” (OTLA). For urban areas the livelihood
groups included: “self-employed” (SEMP), “wage/salaried” (RWSE), “casual
labor” (CALA), and “other” (OTHER).
The socio-religious groups in the analysis included four social groups: the
“Scheduled Tribes” (ST), “Scheduled Castes” (SC), “Other Backward Classes”
(OBC), and “other castes” (non-ST/SC/OBC), and three religious groups, the
“Muslim minority” (MM), “Hindus,” and “other religious groups” (ORM). The
ST are indigenous ethnic groups that are often physically and socially isolated.
The SC traditionally suffered from social exclusion and discrimination associated
with the caste system. The government treats Muslims as a targeted minority for
210 S. Thorat

special policies. Data on the OBC were not separable from the “other castes” in
the 1983–1984 database and therefore they were both included together with
“other castes” for some analyses.

13.3.1 Socio-religious Groups

First, general changes in poverty were examined and afterwards the data were
disaggregated by livelihood, socio-religious group, and between rural and urban
areas. Between 1983 and 2005 overall rural poverty declined at an average rate of
1.9 % per annum, which was equivalent to an 18 % point decline over the entire
period. Across social groups the average annual rate of decline in poverty was great-
est for the non-ST/SC group (2.1 %), followed by the SC (1.8 %), and ST (1.4 %)
groups. In regard to religious groups, rural poverty declined at a slightly higher
average annual rate for the Hindus and ORM (both 1.9 %), than for the MM group
(1.7 %), despite the fact that the latter group has been targeted by government
poverty alleviation policies.
During 2004–2005 the HCR poverty levels were persistently higher among the
ST households (47 %), followed by the SC (37 %), OBC (26 %), and non-ST/SC/
OBC (17 %). Although the MM had a higher poverty level (33 %) than the Hindus
and ORM groups, they had lower poverty levels than the SC and ST groups. This
pattern prevailed during the mid-2000s.

13.3.2 Livelihood Categories and Socio-religious Groups

I examined the changes in poverty among livelihood categories by socio-religious

group. Among rural livelihood categories the wage labor households were poorer
than the self-employed. In 2004–2005 approximately 44 % of the AGLA liveli-
hood households and 33 % of the OTLA households were poor. By comparison
the poverty levels for both the SEAG and SENA livelihood households were
about 20 %.
Rural poverty also declined at higher annual rates for the self-employed liveli-
hoods (2.3 % for SEAG and 2.1 % for SENA) than for wage-labor (approximately
1.6 % for both AGLA and OTLA). The decline was slightly less for the AGLA
livelihoods than for OTLA. The poverty rates among the SEAG livelihood house-
holds differed by socio-religious group. Poverty declined for the SEAG livelihood
households among all groups over the study period, but at lower rates for the SC and
ST households than the other social groups, and at a lower rate for the MM com-
pared to the Hindus among the religious groups.
During the overall period the SENA livelihood households also experienced
declines in poverty among all socio-religious groups, however, the ST and SC SENA
households declined at lower average annual rates (about 1.9 %) compared to all other
13 Tackling Social Exclusion and Marginality for Poverty Reduction… 211

groups (2.5 %). Among the religious groups the MM declined at a lower average
annual rate (1.7 %) than the Hindus and ORM households (both about 3.6 %).
Among the socio-religious groups, poverty declined at the lowest rates for the ST
and SC wage-labor livelihood households. In the case of AGLA livelihoods among
social groups, poverty declined at an average annual rate of 1.1 % for the ST, 1.5 %
for the SC, and 1.7 % for the non-ST/SC group. In the case of OTLA livelihoods
among social groups, poverty declined at an average annual rate of 1.2 % for the ST
and SC groups, compared to 2.0 % among non-ST/SC households. Thus, for both
rural wage labor livelihood categories the ST and SC groups exhibited less poverty
reduction than the other social group. In the case of religious groups there were no
significant differences in the rates of poverty decline among rural wage labor liveli-
hood households, however, the ORM OTLA households did slightly better than
their Hindu and MM counterparts.

13.3.3 Changes in Income—Monthly Per Capita Expenditures

as a Proxy Variable

I examined changes in household income using “monthly per-capita expenditures”

(MPCE) as a proxy value. The National Sample Survey Organisation defines the
means of livelihood on the basis of majority income contribution source at the
household level. For example, a household is classified as an AGLA household if
50 % or more of its total annual income is derived from agricultural labor. By this
criterion the OTHER category of household livelihoods are those whose income
comes from two or more sources such that income from each source is less than
50 % of total household income. During the study period the overall MPCE
increased at an average rate of 1.2 % per annum. The average per annum percent-
age increases were nearly the same for all socio-religious groups (1.1–1.3 %). The
SENA livelihood households exhibited the highest average annual increases in
MPCE (1.6 %) and the rest of the livelihood groups increased at lower rates
(around 1.0 %). Urban OTHER livelihood households experienced the highest
average per annum increase at 2.0 %.
Among the social groups the ST and non-ST/SC had relatively higher MPCE
increases than the SC group. Among the religious groups the Hindus and ORM
groups had relatively higher rates of MPCE increase than the MM. The SC and MM
SENA households lagged behind all other socio-religious groups in the rates of
MPCE increase.
There was not much difference in the rates of MPCE increase of SEAG liveli-
hood households among the socio-religious groups. Among the rural wage labor
livelihoods the AGLA households increased at an average per annum rate of
1.1 % compared to 0.8 % for OTLA households. With the exception of the MM
households, whose MPCE increased at slightly higher rate (1.6 %), there were
no substantial differences in the average annual MPCE rates among the socio-
religious groups.
212 S. Thorat

13.3.4 Changes in Poverty, Income, and the Elasticity

of Poverty Reduction

During the 21-year period between 1983 and 2005, rural poverty declined at an
average rate of 1.9 % per annum, while the MPCE grew by an average rate of 1.2 %
per annum. This suggests that increases in consumption expenditures were corre-
lated with reduced poverty. The growth in consumption expenditures was lower, just
as poverty was reduced at a moderate rate. The overall elasticity value of poverty
reduction with respect to MPCE growth was 2.3 in 2004–2005.
The fact that the average per annum growth in MPCE was more or less the same
across socio-religious groups, and that the ST, SC, and MM groups that were ini-
tially the poorest also lagged behind to the other groups in poverty reduction, indi-
cate that economic growth has not been pro-poor. This assertion was supported by
the different elasticity values of poverty reduction with respect to MPCE during the
1983–2005 period. The growth elasticity with respect to poverty reduction values
were 2.6 for the non-ST/SC group, 2.0 for SC, 1.6 for ST, and 1.7 for the MM group.
Therefore, economic growth brought less reduction in poverty for groups suffering
from greater poverty compared with those experiencing less poverty.

13.3.5 Livelihood by Socio-religious Groups

13.3.5.1 Rural Self-Employed Households

The results indicated that per unit growth in income among self-employed livelihoods
brought greater declines in poverty for agricultural (SEAG) than non-agricultural
(SENA) livelihoods. This was clearly supported by much higher growth reduction
elasticity (3.4) for SEAG households compared to SENA (2.2) households.
The differential decline also suggests that rural agricultural growth has been
more pro-poor than non-agricultural growth. Given the facts that the bulk of the
poor live in rural areas and that growth has the potential to benefit poor farmers,
more rural agricultural investment with a clear focus on marginal and small
scale farmers is likely to alleviate poverty. To improve the pro-poor performance of
growth in the non-agricultural sector requires policy measures that will change
policy orientation towards small producers and businesses in such a manner that
future growth will increase the income of poor self-employed producers and business
households.

13.3.5.2 Rural Wage-Labor Households

The patterns are somewhat different with respect to the pro-poor character of agri-
culture and non-agricultural sector growth for wage laborers, who constituted the
13 Tackling Social Exclusion and Marginality for Poverty Reduction… 213

bulk of the poor in rural areas. The distributions of gains from increased farm
income were not shared equally by SEAG and AGLA livelihood households. AGLA
livelihoods experienced the lowest poverty decline, had the highest HCR poverty
level, and the lowest rate of MPCE increase of any livelihood category. Although
growth in small-scale agriculture is likely to reduce poverty for SEAG households,
it may not perform for wage laborers as many SEAG households provide their own
labor. The growth of labor intensive agriculture may result in greater employment
or higher-wages and thus reduce poverty for wage laborers.

13.3.5.3 Socio-religious Groups

This pattern of greater improvement among self-employed versus wage-labor

agricultural livelihoods was observed across all socio-religious groups. Despite
similar increases in MPCE poverty declined at differential rates, indicating differen-
tial poverty reducing impacts of growth on MPCE across the various groups. Poverty
among the ST, SC, and MM groups declined at lower per annum rates compared to
the other groups. What explains this differential reduction in poverty? Increases in
MPCE corresponded with lower declines in poverty for the ST, SC, and MM house-
holds compared to the other socio-religious groups. This was borne out by the lower
poverty reduction elasticity with respect to MPCE for the ST, SC, and MM house-
holds, which were 1.6, 2.0, and 1.8 respectively. The elasticity values were 2.6 for
the non-ST/SC group, 2.4 for Hindus, and 2.5 for ORM.
The pro-poor character of growth in MPCE also differed for self-employed and
wage labor livelihoods across socio-religious groups. In the case of SEAG liveli-
hoods, there was little difference in the rate of MPCE change across social groups.
Therefore, the variable elasticity of poverty reduction was associated with unequal
outcomes in poverty reduction across socio-religious groups. Growth elasticity
values were lower for the ST and SC households among social groups, and for
Hindus among religious groups. Similar to SEAG livelihoods, poverty among
SENA livelihood households also declined at lower rates for the ST, SC, and MM
socio-religious groups compared to the rest. Lower declines in poverty were closely
associated with relatively lower increases in MPCE for these groups. The average
per annum poverty decline was particularly lower for the SC group. The elasticity
value of poverty reduction was also lower for the ST and MM groups. In the case of
the SC group, although MPCE increased only marginally they seem to have gone in
favor of the poor households in so far as the elasticity of poverty reduction value
was the highest of any socio-religious group. This may also mean that the SC non-
agricultural production and business activities have higher potential for poverty
reduction in rural areas. In the case of the ST and MM groups, lower increases in
MPCE as well as lower poverty reduction elasticity with respect to MPCE were
linked to lower poverty declines among SENA livelihood households. In the case of
the non-ST/SC, Hindus, and ORM groups, both higher average per annum increases
and higher elasticity of the MPCE values were linked to higher poverty reduction
among SENA livelihood households.
214 S. Thorat

For wage labor livelihoods, poverty declined among AGLA households at a

lower rate for the ST group, followed by the SC, MM, and non-ST/SC groups.
Except for a somewhat higher increase for MM households, the per annum increase
in MPCE was similar for all groups. Therefore, the lower poverty declines of the ST,
SC, and MM AGLA livelihood households were likely linked to the lower poverty
reduction elasticity values. The ST, SC, and MM groups seemed to face more con-
straints in AGLA livelihoods compared to the rest of the groups. The higher poverty
declines among AGLA livelihoods for the ORMs and Hindus occurred along with
greater increases in both MPCE and poverty reduction elasticity values. In the case
of OTLA livelihoods, poverty declined at much lower rates for the ST and SC
groups compared to the rest of the socio-religious groups. The lower poverty reduc-
tion rates for the ST and SC OTLA livelihoods were accompanied by lower increases
in MPCE, as the poverty reduction elasticity values did not differ between them. In
the cases of the other social groups, both increased MPCE and relatively higher
elasticity values were accompanied by reductions in poverty. In the case of OTLA
livelihood MM households the growth in MPCE was an important factor.

13.4 Urban Poverty

13.4.1 Changes in Poverty by Social Groups

The overall poverty level in urban areas was 26 % in 2004–2005, only about 2 %
lower than in rural areas; however, there were notable differences among the socio-
religious groups. Unlike in rural areas where poverty incidence was the highest
among the ST group followed by the SC, MM, and OBC, in urban areas the SC and
MM groups had the highest incidences of poverty. The ST and OBC groups had
similar poverty levels, whereas non-ST/SC/OBC among the social groups and
ORMs among religious groups had the lowest poverty incidences. For the ST, SC,
and MM groups the highest poverty levels were for CALA livelihood households at
58 %. The groups with the highest poverty levels among CALA households were
the ST and SC, with HCRs of 74 and 67 % respectively. Poverty was also high
(40 %) among the SEMP (self-employed) livelihoods, particularly for the SC, ST,
and MM groups. The poverty level was also high (34 %) among RWSE (wage/sala-
ried) livelihood households of the MM group.
Several interesting features emerged regarding the direction of change across
socio-religious groups. During the 1983–2005 period overall poverty levels declined
at an average rate of 1.9 % per annum. For the overall period, the lowest average
annual poverty reduction rates were observed among the ST (1.1 %) and the MM
(1.4 %), and the highest among the non-ST/SC (2.4 %) and SC (2.0 %) groups.
Among religious groups the highest rates of decline were for the ORM (3.1 %)
group, followed by Hindus (2.1 %), and MM (1.4 %). With regard to poverty reduc-
tion rates in the urban sector by socio-religious group, the data were only compa-
rable for the period from 1993–1994 to 2004–2005, because the group classification
13 Tackling Social Exclusion and Marginality for Poverty Reduction… 215

in 1983 was different. Poverty among SEMP livelihood households declined at

relatively lower rates for the SC and MM groups compared to the rest. In the case of
RWSE livelihoods, poverty declined at much higher average annual rates for the ST
(5.4 %) and ORM (4.1 %) groups, followed by the SC, non-ST/SC, and Hindus. The
declines were lowest for the MM group (0.6 %). While the SC group performed
badly in poverty reduction for SEMP livelihoods, the rates were better for RWSE
livelihoods. The MM group exhibited low poverty reduction among both the SEMP
and RWSE livelihood categories during the 1993–2005 period. In the case of CALA
livelihood households, the SC and ST groups lagged behind the others. The MM
and non-ST/SC groups experienced greater poverty reduction among CALA liveli-
hood households than the rest.
Growth in urban income. During the entire 1983–2005 period urban MPCE grew by
an average of 1.8 % per annum for all households, which was significantly higher
than in rural areas. Among social groups the highest growth rate was experienced by
the non-ST/SC group (1.9 %), while the SC group had the lowest (1.3 %). The ST
group was closer to the non-ST/SC group with an average rate of 1.8 % per annum.
Growth, poverty, and elasticity linkages. The overall poverty reduction elasticity
value was 1.5. Among socio-religious groups elasticity values were relatively higher
for the ORM (2.04) group, non-ST/SC (1.64), and Hindus (1.61); and lowest for the
MM (1.26), SC (1.31), and ST (1.48) groups. Thus, the urban economic growth was
less pro-poor with respect to impacts on the MM, SC, and ST groups. The SEMP
livelihoods for which comparable data were available exhibited similar patterns,
except that the rank order was the ST, followed by MM and SC.

13.5 Implications for Socially Inclusive Policy

These results have implications for poverty reduction policies for all socio-religious
groups. Before discussing policy issues in India, we need to examine insights from
theoretical and empirical literature. In particular it is important to consider literature
on the relationships between economic development, poverty, and inequality in
income distribution, including recent discussions on inclusive growth and to what
extent it addresses the issues of excluded groups.

13.5.1 Improving Pro-poor Performance of Agricultural

and Non-agricultural Growth

In light of the changes in poverty among social groups during the 1983–2005 period
and insights from earlier studies on pro-poor growth, I propose group-specific
policy measures. The findings point towards strengthening some existing policies
and also developing some new ones.
216 S. Thorat

13.5.1.1 Agricultural Livelihoods

On the positive side, growth in agriculture has been far more pro-poor than
non-agricultural growth. In fact among all livelihood categories the elasticity of
poverty reduction with respect to MPCE was highest for SEAG livelihoods, imply-
ing that agricultural sector growth has been more effective at reducing poverty
among small-scale and marginal producers. Since agricultural sector growth has
poverty reducing potential, there is justification for strengthening the observed pat-
tern of growth. There are, however, negative aspects of agricultural growth which
would likely benefit from policy change. Agricultural growth did not reduce poverty
among AGLA (wage labor) livelihoods in equal measure. The elasticity of poverty
reduction with respect to MPCE was much less for AGLA (1.4) livelihoods com-
pared to SEAG (2.3). This is consistent with the facts that AGLA livelihood poverty
decreased at a lower rate than any other livelihood category, had the highest HCR
value, and had the lowest increase in MPCE.
AGLA livelihood households will benefit more from growth in agriculture only
if their employment opportunities or wages increase. Growth in labor intensive
medium and large scale agriculture may result in greater wage employment and thus
reduce poverty for AGLA livelihood households. Policies that encourage large- and
medium-scale producers to cultivate labor-intensive crops and utilize labor-intensive
practices have poverty reducing potential. Thus, an overall agricultural growth strat-
egy that addresses both labor intensive production among large and medium-scale
producers and continued improvements for small-scale and marginal producer will
be more effective at reducing poverty.
Another negative feature of the observed agricultural growth was the limited
poverty improvement among the ST, SC, and MM socio-religious groups. Growth
elasticity values with respect to poverty were 2.5 for the non-ST/SC and Hindus, 2.0
for the SC and MM groups, and only 1.4 for the ST. Strengthening the position of
small-scale producers in general, particularly among the ST, SC, and MM groups is
a necessary element of inclusive agricultural sector growth. While employment-
oriented policies are necessary for improving AGLA livelihoods in general, special
efforts are necessary for the ST, SC, and MM groups, who often suffer from dis-
crimination in rural labor markets (Thorat and Newman 2010).
Based on experiences since the early 1980s, inclusive policy for the agricultural
sector would include:
• More labor-intensive agriculture among medium- and large-scale producers
• Focus on small-scale producers
• Special assistance for the ST, SC, and MM small-scale producers
• Special attention for the ST, SC, and MM, AGLA livelihoods

13.5.1.2 Rural Non-agricultural Livelihoods

During the 1983–2005 period rural non-agricultural growth was more pro-poor for
wage labor than for self-employed livelihoods. The elasticity value of poverty
13 Tackling Social Exclusion and Marginality for Poverty Reduction… 217

reduction for OTLA livelihoods was 2.0 as opposed to 1.4 for AGLA livelihoods.
Limited poverty reduction for the ST and SC OTLA households was associated with
lower increases in MPCE. In the case of the non-ST/SC group, increases in both
MPCE and elasticity were relevant to poverty reduction.
The non-agricultural sector policy implications are clear. Since the elasticity of
poverty reduction was positive and higher for wage labor, and the slow decline in
poverty was linked to limited increases in MPCE, measures are necessary to increase
rural non-agricultural sector growth. The pro-poor character of growth also needs to
be strengthened, because elasticity began to decrease in the 1990s. The pro-poor
character of production activities needs to be enhanced, particularly for self-
employed livelihoods, through support and incentives for increasing the profitability
of small producers and businesses.

13.5.2 Group-Specific Policy Measures

Poverty was reduced at a lower rate for the ST, SC, and MM socio-religious groups,
which suffer from social exclusion and discrimination in India. Members of these
groups typically own less agricultural land, have less access private non-agricultural
economic activities, and are more dependent on wage employment. These groups
have higher historical levels of poverty, which has been found to limit poverty
reduction. The socio-religious groups with better access to assets and lower histori-
cal poverty have done better. This implies that the groups with higher poverty levels
and less access to assets and sustainable livelihoods need specific and affirmative
action policies.
Among the social groups the ST performed the worst in poverty reduction for
both self-employed and wage labor households. The lack of access to agricultural
land is generally less of a problem for the ST, about 40 % of ST households were
farmers and approximately 40 % of these households were classified as poor. The
poverty level among the ST AGLA households was formidably high at 56 % in
2005. Poverty levels associated with tribal agriculture are persistent and higher due
to low productivity and income. How to make agriculture more productive in tribal
areas is a challenge that has not been addressed effectively.
It appears that a core need is to develop appropriate crops and methods for tribal
areas. This will require renewed research efforts on the special needs of agriculture
in tribal areas, including the potential for developing tree, flower, fruit, and livestock
or dairy production on grazing land. Tribal agriculture has pro-poor qualities, both
for SEAG and AGLA livelihoods, as the elasticity values were positive and tended
to be higher in good years.
Unlike the ST, the SC group has much less access to agricultural land, and as a result
the bulk of the SC households (about 56 %) were wage labor reliant. Poverty reduction
among the SC will depend on employment enhancement. Increased employment
opportunities and improved skill and education development will raise employment
rates households in both the agricultural and non-agricultural sectors (Dev 2005).
218 S. Thorat

In the case of the MM group, access to income earning assets was better than for
the SC group. More than half of the MM households were engaged in productive
enterprises and/or businesses. Policy measures that focus on strengthening the
incentives and supporting small-scale enterprises and businesses have potential to
reduce poverty for this group.

13.5.3 Urban Development

In the case of SEMP livelihoods, all socio-religious groups except the ST exhibited
poverty declines with moderate increases or stagnation in MPCE. In the case of the
MM group, both income growth and elasticity with respect to poverty were quite
low. Approximately 52 % of the urban MM households were engaged in SEMP
livelihoods, but MPCE growth was lower and poverty was higher. Approximately
one-third of the SC households were engaged in SEMP livelihoods, but this group’s
MPCE increase was the lowest and its poverty levels highest among the socio-
religious groups. In order to accelerate poverty reduction, growth in urban SEMP
livelihoods is necessary. A policy focus on small producers and businesses would
make growth more pro-poor and inclusive. In the case of RWSE livelihoods, for
which we only have data for the 1990s, poverty showed significant declines for the
ST and ORM socio-religious groups, followed by the SC, Hindus, and MM. So
while the expansion of employment opportunities in urban areas should continue,
there is a need to make it more pro-poor, which requires enhancement in education
and skill development opportunities among the SC and MM groups. Lastly, the
CALA livelihoods, which were the poorest among the three urban livelihood cate-
gories, experienced the lowest poverty declines. This was particularly the case for
the SC and ST CALA households. Increasing hands-on skill development and the
expansion of employment opportunity are the only alternatives for this chronically
poor livelihood category.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Adelman I, Morris CT (1973) Economic growth and social equity in developing countries. Stanford
University Press, Stanford
Ahluwalia MS (1976a) Income distribution and development: some stylized facts. Am Econ Rev
66(May):128–135
Ahluwalia MS (1976b) Inequality, poverty and development. J Dev Econ 3:307–342
Dev MS (2005) Pro-poor growth in India: what do we know about the employment effects of
growth 1980–2000? Working paper no 161. Overseas Development Institute, London
Grinspun A (2009) Pro-poor growth: finding the Holy Grail, IPC-IC collection of one pagers no 6.
International Poverty Centre, United Nations Development Programme, Brasilia
13 Tackling Social Exclusion and Marginality for Poverty Reduction… 219

Hull K (2009) Understanding the relationship between economic growth, employment and poverty
reduction. Organisation for Economic Co-operation and Development, Paris
Kakwani N (2000) On measuring growth and inequality components of poverty with application
to Thailand. J Quant Econ 16(1):67–79
Kraay A (2004) When is growth pro-poor? Cross-country evidence, World Bank policy research
working paper no 3225. World Bank, Washington, DC
Kuznets S (1955) Economic growth and income inequality. Am Econ Rev 45(1):1–28
Lipton M (1991) A note on poverty and sustainability. IDS Bull 22(4):12–16
Osmani S (2005) Defining pro-poor growth. In: IPC-IC collection of one pagers no 9. International
Poverty Centre, United Nations Development Programme, Brasilia
Papanek G (1978) Economic growth, income distribution and political processes in LDCs. In:
Griliches Z, Krelle W, Krupp H-J, Kyn O (eds) Income distribution and in economic inequality.
Campus Verlag/Halstead Press, Frankfurt/New York
Paukert F (1973) Income distribution at different levels of development. Int Labour Rev
108:97–125
Ravallion M (2001) Growth, inequality and poverty: looking beyond averages. World Dev
29(11):1803–1815
Ravallion M (2009) Economic growth and poverty reduction: do poor countries need to worry
about inequality? In: von Braun J, Hill RV, Pandya-Lorch R (eds) The poorest and hungry:
assessments, analyses, and actions. International Food Policy Research Institute, Washington,
DC, pp 179–186
Ravallion M, Chen S (2003) Measuring pro-poor growth. Econ Lett 78:93–99
Ravallion M, Datt G (2002) Why has economic growth been more pro-poor in some states of India
than others? J Dev Econ 68(2):381–400
Sen A (2000) Social exclusion: concept, application and scrutiny, Social development paper no 1.
Office of Environment and Social Development, Asian Development Bank, Manila
Thorat A (2010) Ethnicity, caste, and religion—implications for poverty outcomes. Econ Polit
Wkly XLV(52):18–24
Thorat S, Newman K (eds) (2010) Block by caste: economic discrimination modern India. Oxford
University Press, New Delhi
Thorat S, Vanketsan S, Mahamallick M (2005) Human poverty, social exclusion and marginalized
groups, Working paper. United Nations Development Programme, New York
Tsakloglou P (1988) Development and inequality revisited. Appl Econ 20(4):509–531
von Braun J, Hill RV, Pandya-Lorch R (eds) (2009) The poorest and hungry: assessments, analyses,
and actions. International Food Policy Research Institute, Washington, DC
Chapter 14
Consumption Behavior of the Poorest
and Policy Implications in Indonesia

Evita Hanie Pangaribowo

Abstract Improving household food consumption involves a multitude of issues.

The effort becomes more complicated when it addresses the poorest and most
vulnerable. This chapter examines these issues and in particular the characteris-
tics of the poorest households that are interconnected with their food consumption
behavior. The responses of the poorest and forgotten households in Indonesia to
changes fundamental economic variables, income and prices, were examined. It
was found that aspects of the poorest households’ consumption behavior would
have impacts on the effectiveness of food subsidy efforts, and should therefore
be taken into account when such programs are designed to avoid unintended or
detrimental effects.

Keywords Consumption behavior • Poverty • Food subsidies • Indonesia

14.1 Introduction

Food is a basic and fundamental human need. Adequate access to food as a human
right was established in the United Nations Universal Declaration of Human Rights,
which states that “everyone has the rights to a standard of living adequate for the
health and well-being of himself and his family, including food” (UN 1948). This
concept was further codified by the UN in the assertion that “every man, woman,
and child has the inalienable right to be free from hunger and malnutrition in order
to develop fully and maintain their physical and mental faculties” (UN 1974). At the

E.H. Pangaribowo (*)

Center for Development Research (ZEF), University of Bonn,
Walter Flex Strasse 3, 53113 Bonn, Germany
Department of Environmental Geography, Gadjah Mada University,
Yogyakarta, Indonesia
e-mail: [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 221
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_14,
© The Author(s) 2014
222 E.H. Pangaribowo

global level the first Millennium Development Goals to halve poverty and hunger
parallel the mandate of the Universal Declaration of Human Rights. To assure the
right to food for everyone, each country needs to achieve food security. Food secu-
rity exists when all people have physical and economic access to sufficient food to
meet their dietary needs for a productive and healthy life at all times (FAO 1983,
2003; Timmer 2000). Accordingly, food security has become one of the main national
agendas in most developing countries (Maxwell 1990, 1996; von Braun et al. 1992).
The behavior of poor households should be of policy interest for governments
(von Braun et al. 2009), particularly when it comes to the basic need for food. In
Indonesia, as in many other developing countries, food is considered one of the
most important issues in the economy, as well as in socio-political debate. The
national government of Indonesia has been devoted to large-scale endeavors to
address food insecurity through the Food Security Council. The Indonesian govern-
ment also provides food subsidies to poor households for the purpose of maintaining
access to basic food items and to achieve a wide range of other socio-economic effects.
Improving household food consumption involves a multitude of issues (von
Braun et al. 1992). The effort becomes more complicated when it addresses the
poorest and most vulnerable. My aim in this study was to examine these issues and
in particular to observe the behavioural characteristics of the poorest households
that are interconnected with their food consumption behavior. I also analyzed the
responses of the poorest households to changes in fundamental economic variables:
income and price. Examining the impacts of current food policy and interventions
for the poorest households is vital for understanding how effectively these policies
are benefiting the poorest.

14.2 Characteristics of the Poorest

Understanding the characteristics of the poorest is an appropriate starting point for

designing policies and programs that improve their welfare. Following Ahmed et al.
(2007), the poorest households were defined as the bottom 20 % of the household
income range, whose members were living on less than US$0.50/day. I focused on
the characteristics of the poorest households that highlight several aspects of the
relationships between food policy and household food consumption behavior,
including expenditure patterns of food consumption, geographical aspects, and
social conditions. Understanding the diversity of food consumption is an important
key for addressing the food and nutritional problems of the poor. The poorest house-
holds are often found in relatively remote locations that have the least access to
public infrastructure and facilities such as roads and markets, which disadvantages
them by limiting food variety and increasing food prices. This geographical associa-
tion with poverty also hinders access to other basic needs, such as health care and
education. With respect to human and social capital, members of the poorest house-
holds are less likely to be educated and more likely to be systematically excluded
from certain opportunities.
14 Consumption Behavior of the Poorest and Policy Implications in Indonesia 223

14.2.1 To Eat: Spending on Food

Household consumption patterns are considered to be among the most reliable

indicators of the economic development and public welfare of a country. As income
rises households tend to diversify their diet and increase the consumption of foods
with greater nutritional value. The Indonesian Bureau of Statistics (2009) reported
changes in household consumption patterns over the last decade. In early 2000 the
mean share of Indonesian household budgets spent on food was 58 %, which
decreased to around 50 % in 2009. Nonfood consumption increased from 40 to 50 %
during the same period. While these consumption changes have been evident at the
national level, less change has occurred among the poorest households. The mean
monthly income per capita of these households was around 120,000 rupiahs (US$13).
Based on data from the Indonesian Family Life Survey (IFLS),1 the poorest house-
holds spent more than half of their budget on food (63 %) in 1997, and this figure
changed only slightly in 2007 to 61 %, indicating that the poorest spent a relatively
higher proportion of their household budget on food than the average household.
Within food expenditures, the “staple foods” category dominated the budgets of
the poorest households (Fig. 14.1). The second greatest food expenditure share
among the poorest households was for the “meat and fish” category. Dairy products
and other foods from animal sources generally have higher nutrient content and bet-
ter taste, but cost more than alternative foods, therefore poorer households are less
likely to be able to afford them.
The expenditure share of the “alcohol and tobacco products” category was higher
than for “dairy products” among the poorest households. The Indonesian Consumer
Foundation estimated that 70 % of the smokers in Indonesia are poor, pointing out
that the poor are the main contributors to government revenue from tobacco taxes
(Suara Media 2010). While expenditures on the “vegetables and fruit,” “meat and
fish,” and “dairy products” categories have remained stagnant, the increased con-
sumption of “snack and dried foods” and “alcohol and tobacco products” categories
indicates that the poorest households spend money on non-nutritious food and other
unhealthy alternatives.
Dairy product consumption represented the smallest share of the budget among
the poorest households. Dairy products are important sources of micronutrients,
especially calcium, and regular consumption of dairy products reduces the risk of
malnutrition. The low level of dairy product consumption is likely associated with
the low purchasing power of the poorest households, and it is widely known that
consumption choices of the poor are greatly influenced by price. The typically high
prices for dairy products in Indonesia are associated with the need to import

1
IFLS is a longitudinal socioeconomic and health survey that has been conducted in 1993, 1997,
2000, and 2007 (RAND 2010). IFLS collects data on individual respondents, their families, their
communities, and the health and education facilities they use. In 1997 the IFLS sample represented
approximately 83 % of the Indonesian population living in 13 provinces.
224 E.H. Pangaribowo

35.00

30.00
% of Total Food Expenditures

25.00

20.00

15.00

10.00

5.00

0.00
Staple Veg. and Meat and Dairy Oil Alcohol Snack and Other
Fruit Fish and Dried food
Tobacco
1997 2000 2007

Fig. 14.1 Consumption by expenditure category among the poorest households in Indonesia
(Based on IFLS data)

30.00
Expenditure Share on Dairy Products

25.00

20.00

15.00

10.00

5.00

0.00
1 2 3 4
# Children Under-Five
Poorest Mean Wealthiest

Fig. 14.2 Household expenditures on dairy products in relation to the number of children under
the age of five among the highest and lowest household income groups in Indonesia (Based on
IFLS data)

ingredients for their production. Figure 14.2 presents expenditures on dairy products
in relation to the number of household children under the age of five. In contrast to
wealthier and the middle-income households that consumed more dairy products
as the number of young children increased, the poorest households exhibited the
reverse pattern.
14 Consumption Behavior of the Poorest and Policy Implications in Indonesia 225

60

% of the Poorest Households

50

40

30

20

10

0
Java Non-Java Urban Rural

Fig. 14.3 Relative distributions of the poorest households in Indonesia: 2007 (Based on IFLS data)

14.2.2 Where Do the Poorest Live?

Similar to patterns of distribution in other developing countries, the majority of the

poorest households in Indonesia are located in rural areas and on the outer islands
(Fig. 14.3).2 The rural poor households are typically dependent on agricultural
livelihoods and have limited access to land. According to related studies the poorest
households in urban areas are typically engaged in informal market sector liveli-
hoods such as being self-employed or operating micro-businesses (Ahmed et al.
2009). The level of development in Indonesia is highly imbalanced, being particu-
larly greater in urban relative to rural areas and concentrated on the island of Java
relative to the rest of the country (Hill 1992; Akita 2003). Access to markets
(products and labor) is also greater in urban areas and on Java. As the capital and
government administration centers are located in Java, the distribution of social
safety-net program activities are also frequently subject to spatial biases favoring
Java relative to the rest of the country. For example the Rice for the Poor program
was implemented first in Java and the program’s distribution was imbalanced in
favor of the island until the mid-2000s (Suryahadi and Sumarto 2003), although in
that study it was recognized that the higher safety-net coverage might be associated
with greater effects of the 1997 financial crisis in urban areas of Java relative to the
rest of the country. Nevertheless the imbalanced execution of food security policies
prevents benefits from reaching the nation’s poorest households.
Household food insecurity and malnutrition are not only associated with food
intake problems, but are also linked with environmental conditions. The quality of

2
Java is one of the principal islands of Indonesia and the most densely populated. It is where the
capital city is located and the economic and governmental center of the country.
226 E.H. Pangaribowo

30.00

% of Poorest Households with Access 25.00

to Potable Water
20.00

15.00

10.00

5.00

0.00
Total Urban Rural

Fig. 14.4 Potable water access among the poorest households in Indonesia: 2007 (Based on
IFLS data)

water and sanitation services can significantly affect household health. Sanitary
environmental conditions are strongly associated with unhindered childhood growth
and a lower prevalence of diarrhea (Packnawin-Mock et al. 2000; Cameron and Shah
2010). Figure 14.4 presents the percentages of the poorest households with direct
access to drinking water. The poorest households have the least access to improved
sanitation and drinking water (Cameron and Shah 2010). Urban poor households had
better access to water and sanitation infrastructure relative to the rural poor.

14.2.3 Education

The education level of the household head had a positive and significant association
with the consumption of “vegetables and fruit,” “meat and fish,” and “dairy products,”
which implies that more education contributes to the consumption of more nutritious
foods. In contrast household head education had a negative and significant associa-
tion with the consumption of “staple foods” and “alcohol and tobacco products.”
Based on the IFLS data, the heads of the poorest households were the least educated
among all households, nearly all of them had not completed elementary education.
Compared to the other households, the poorest and least educated households
consumed greater shares of “staple foods,” “alcohol and tobacco products,” and
“snack and dried foods” (the latter often with image-bearing packaging to enhance
marketing), but consumed lesser shares of “dairy products.”
The fact that the poorest households were also the least educated may affect
consumption behavior in several ways. Limited education may be associated with
limited understanding of dietary nutrition. The poorest households also appear to be
responsive to food marketing efforts, which tends to promote an idealized image of
14 Consumption Behavior of the Poorest and Policy Implications in Indonesia 227

40.00

35.00
% of Total Food Expenditure

30.00

25.00

20.00

15.00

10.00

5.00

0.00
No Schooling 2 years 3 years 4 years 5 years Completed
Elementary
Staple Dairy Products
Alcohol and Tobacco Snack and Dried food

Fig. 14.5 Percentages of household expenditures among selected good categoris by education
level of the household head in Indonesia (Based on IFLS data)

consumers that is consistent with popular cultural fashion. As a result this group
might be less attracted to alternative foods that may be superior in nutrient content.
Anecdotally, the poorest and least educated households tend to be more culturally
traditional, which might also favor consumption of staple foods (Fig. 14.5).

14.2.4 Social Networks: Participation in Community Activities

In many studies the poorest demographics are characterized as excluded from and
less participative in society. Are the poorest households socially excluded in
Indonesia? Community activities in Indonesia are an excellent laboratory to observe
the social behavior of the poor. Previous studies in Indonesia have found that selec-
tivity exists in community participation, and that that marital status and education
were strong determinants of social engagement (Grootaert 1999; Beard 2003; Beard
and Cartmill 2007). Another study found that participation at the community level
increased the likelihood of receiving government aid during economic crises and of
employment for the household head (Perdana 2006).
There are various types of community participation at the village level in
Indonesia, many of which are specific to gender and age. Anti-poverty development
programs such as the Social Safety Net Program, the Kecamatan Development
Program, the Urban Poverty Program, the Rice for the Poor program, the unconditional
cash transfer program, and other programs are discussed in community meetings.
These meetings provide forums for the discussion of activities such as program
228 E.H. Pangaribowo

Participation in Community Activities

100
90

(% of Poorest Households)
80
70
60
50
40
30
20
10
0
Voluntary Labor Community Religious Rotating savings
Meeting Activities
Yes No

Fig. 14.6 Participation of the poorest households in community activities in Indonesia (Based on
IFLS data)

socialization and planning, proposal preparation, funding decisions, and program

implementation. Local meetings are attended by several elements of the communities,
such as local legislative bodies, local government staff, and representatives of women
and youth groups. Among the various community activities is a traditional reciprocal
form of social capital known as gotong royong (voluntary labor) (Sullivan 1992;
Beard 2005). This activity is originally from Java and originated from the coopera-
tive action required for wet rice production (Beard 2005). Based on the IFLS data the
poorest households were not entirely socially excluded (Fig. 14.6). About 50 % of
poorest households participated in religious activities and gotong royong in their
villages. Nevertheless when it came to monetary types of participation or more
formal activities such as rotating savings and community meetings, the poorest
households were less represented than households in the rest of the income range.
Given the fact that the poorest households participated in formal communal
activities, socialization related to access to development programs and empowering
the poorest could be facilitated by working through these types of community activ-
ities. Although relatively less formal, participation in religious and voluntary labor
activities facilitates the sharing of information among villagers and thus creates
social capital. This social capital in turn creates spillover benefits through improved
access to nonfinancial resources for the poorest households.

14.3 The Responsiveness of the Poorest Households

to Income and Price Changes

As reported above, the poorest households consumed greater shares of good from
the “staple foods” and “tobacco or alcohol products” categories, and lesser shares of
nutrient-rich foods from the “meat and fish” or “dairy products” cateories than other
14 Consumption Behavior of the Poorest and Policy Implications in Indonesia 229

Table 14.1 Expenditure elasticity estimates for the poorest households across annual surveys in
Indonesia (Based on IFLS data)
Expenditure categories Total 1997 2000 2007
Staple foods 1.0152 1.1232 0.9129 0.9677
(0.0424) (0.0601) (0.0714) (0.4189)
Vegetables and fruit 0.9967 0.9089 1.0640 0.9804
(0.0639) (0.0983) (0.1061) (0.3990)
Meat and fish 1.0358 0.9333 1.1085 1.2387
(0.0549) (0.0844) (0.0855) (0.6125)
Dairy products 1.2800 1.4963 1.0663 1.2389
(0.1240) (0.2005) (0.1948) (0.2612)
Oils 0.7960 0.7757 0.8645 0.7814
(0.0712) (0.1044) (0.1264) (0.1671)
Alcohol and tobacco products 1.3568 1.1909 1.5788 1.4227
(0.1327) (0.2168) (0.1669) (1.2463)
Snack and dried foods 0.9091 0.8115 0.9752 0.6652
(0.1021) (0.1743) (0.1595) (0.2463)
Other foods 0.9138 0.9118 0.8666 1.0573
(0.0465) (0.0646) (0.0779) (0.2760)
Standard errors are shown in parentheses

households. An important aspect of understanding the behavior of the poorest

households for policy making purposes is their responsiveness to changes in income
and prices, which can be observed using empirical data from the surveys. How the
poor respond to changes in prices and income should help direct food policy with
respect to policy emphasis on either income or price, or on a combination of the two
(Sadoulet and de Janvry 1995). Particularly when policies are needed to leverage
nutrition and health, side interventions should be considered. Policy to improve
household food consumption should therefore be grounded on sound knowledge of
household responses to these key determinants of food consumption.
Changes in consumption among the poorest households in response to changes
in income can be evaluated through expenditure elasticity. Elasticity estimates for
expenditures of the poorest households on eight consumer goods categories are pre-
sented in Table 14.1. All of the expenditure elasticity estimates were calculated
based on the IFLS data using the Quadratic Almost Ideal Demand System
(QUAIDS), specific demand system model. For the “vegetables and fruit,” “oils,”
and “snack and dried foods” categories, the expenditure elasticity estimates were
close to unity and therefore these goods are deemed as necessities, whereas “staple
foods,” “meat and fish,” “dairy products,” and “alcohol and tobacco products” cat-
egories were all classified as luxuries. Expenditure elasticity for the alcohol and
tobacco products category was the highest. For this category the model indicated
that a 10 % increase in total food expenditures would lead to a 13 % share increase
in consumption. Comparing “staple foods” expenditures across annual surveys,
these appeared to be elastic for the poorest households in 1997, but exhibited less
elasticity in subsequent annual surveys. This finding was consistent with Bennet’s
law, which states that households switch consumption from cheaper to more
230 E.H. Pangaribowo

Table 14.2 Own-price and cross-price elasticity value estimates for the poorest households in
Indonesia by expenditure category (Based on IFLS data)
Alcohol
and
Expenditure Staple Vegetables Meat Dairy tobacco Snack and
categories foods and fruit and fish products Oils products dried foods
Staple foods −0.9473 −0.0641 0.1013 −0.0383 −0.0126 −0.0678 −0.0061
(0.1081) (0.0344) (0.0554) (0.0229) (0.0269) (0.0476) (0.0419)
Vegetables −0.1920 −1.0703 0.2446 0.0324 −0.0474 −0.0586 −0.0117
and fruit (0.1103) (0.0654) (0.0703) (0.0293) (0.0323) (0.0571) (0.0552)
Meat and fish 0.2344 0.1731 −1.1402 −0.0286 0.0017 −0.0651 −0.1367
(0.1294) (0.0536) (0.1510) (0.0495) (0.0530) (0.0498) (0.0538)
Dairy −0.5949 0.0392 −0.2008 −1.2391 0.2777 0.1606 −0.0153
products (0.3098) (0.1446) (0.2825) (0.1786) (0.1382) (0.1843) (0.1234)
Oils −0.0121 −0.0213 0.0408 0.1399 −0.7757 −0.0878 0.0348
(0.1673) (0.0868) (0.1385) (0.0636) (0.0981) (0.0917) (0.0634)
Alcohol and −0.2983 −0.2327 0.3988 0.0608 −0.1042 −1.2450 0.0135
tobacco (0.2416) (0.1323) (0.1724) (0.0748) (0.0811) (0.1868) (0.1094)
products
Snack and 0.0159 0.0227 −0.1317 0.0057 0.0094 0.0300 −0.7680
dried (0.1148) (0.0646) (0.0619) (0.0241) (0.0279) (0.0523) (0.0954)
foods
Standard errors are shown in parentheses

expensive calorie sources as income rises (Timmer et al. 1983; Fuglie 2004). For the
poorest households, the “dairy products” category remained expenditure elastic for
over a decade. As expenditures on tobacco and alcohol products increased over the
last decade and considering the elastic nature of expenditures in this category, cau-
tion should be taken before drawing conclusions due to the possible crowding-out
effect of alcohol and tobacco product expenditures on the consumption of more
nutritious food items.
Price elasticity estimates reflect the responsiveness of consumption to changes in
prices. Table 14.2 presents the own- (with respect to changes in the price of goods
within each category) and cross-price (with respect to changes in the prices of goods
from other categories) elasticity estimates for the poorest households based on a
QUAIDS analysis of the IFLS data. Similar to expenditure elasticity, price elasticity
for the “alcohol and tobacco products” category showed the most substantial
response to price changes. The model estimated an elasticity value of 1.25 in abso-
lute magnitude, therefore predicting that if the prices for alcohol and tobacco prod-
ucts increase by 10 %, then consumption would fall by 12.5 %. The poorest
households also exhibited price elasticity for the “dairy products” and “meat and
fish” categories. The own-price elasticity of the dairy products category was similar
to that for “alcohol and tobacco products,” indicating that price changes for “dairy
products” affected consumption less substantially than price changes for “alcohol
and tobacco products.” The price elasticity value for the “staple foods” category was
nearly close to unity among the poorest households, indicating that the demand for
staple foods is relatively responsive to price changes.
14 Consumption Behavior of the Poorest and Policy Implications in Indonesia 231

Examination of the relationships among these consumption categories through

the cross-price elasticity estimates revealed a mixture of complementary and substi-
tutive relationships. Cross-price elasticity values were generally much lower than
own-price elasticity values, implying that consumption of each category among the
poorest households was more responsive to changes in the price of goods within
each category. Though all cross-price elasticity estimates were found to be inelastic,
the notion of substitutes and complements plays a significant role in understanding
consumption patterns. The poorest households somewhat freely substituted between
the “meat and fish” and “vegetables and fruit” categories, suggesting that they seem
to have similar roles in their diet. The comparison of the “vegetables and fruit” category
prices to those of “meat and fish” yielded a demand elasticity value of 0.24, which
predicts that a 10 % increase of vegetables and fruit prices would lead to a 2.4 %
increase in the consumption of meat and fish.
A similar relationship was also found between the “meat and fish” and “staple
foods” categories. The model predicted that the poorest would increase their staple
foods consumption by 2.3 % if the price of meat and fish increased by 10 %. This
finding suggests that poor households would cut expenditures on nutrient-rich foods
such as green leafy vegetables or meat and fish, and shift to cheaper calorie goods
such as staple foods. During the 1997 financial crisis there were dramatic price
increases for most consumer goods, particularly for nutrient rich foods belonging to
the “meats and fish,” “dairy products,” and “vegetables and fruits” categories, which
had negative consequences on the nutritional welfare of Indonesian households.
Another striking substitutive effect, though inelastic, was found in the relationship
between the demand for goods from the “alcohol and tobacco products” and “meat
and fish” categories. The model predicted that if alcohol and tobacco product prices
were to rise by 10 % then meat and fish demand would increase by 3.0 %. A substi-
tutive relationship between “alcohol and tobacco products” with “meat and fish” is
challenging for national food policy. What is important to poor households with
respect to alcohol and tobacco products demand is the price. Only when the price of
alcohol and tobacco products significantly increased, was a shift to greater meat and
fish consumption among poor households predicted. Nevertheless given the large
expenditure share of alcohol and tobacco consumption and its elasticity, it should be
cautiously considered. The loss in terms of nutrition could be considerable due to
the possible crowding-out effect of alcohol and tobacco products expenditures on
those for nutrient rich foods such as eggs, meats, dairy products, and therefore
household well-being would also likely be affected by decreases in nonfood expen-
ditures such as education and medical expenses (BAPPENAS 2006; Mukherjee 2006).

14.4 The Poorest and the Food Subsidy Program

The Indonesian government provides food subsidies to poor households for the
purpose of maintaining access to basic food items. Indonesia operates a food security
program called RASKIN, which is an abbreviation of beras miskin, literally meaning
232 E.H. Pangaribowo

Table 14.3 Food security program targeting performance by income and

region in Indonesia (Based on IFLS data)
Percentage of households
Per capita expenditure quintiles Java Outside Java All
1st (lowest) 28.58 32.59 29.56
2nd 25.33 26.91 25.71
3rd 19.99 21.52 20.37
4th 16.55 12.56 15.57
5th (highest) 9.56 6.43 8.79
Total 100 100 100

“rice for the poor.” Historically the program was part of the Social Safety Net
introduced by the government in 1998 just after the financial crisis hit Indonesia in
mid-1997. The purpose of this program was to ensure that poor households (both
the chronically poor and those newly poor due to the crisis) were able to access
basic food at affordable prices (Sumarto 2006). Eligible households were selected
by the National Family Planning Agency (Badan Koordinasi Keluarga Berencana
Nasional [BKKBN]).
Tabor and Sawit (2001) pointed out that the program authorities were aware that
the BKKBN welfare criteria were not designed to identify food insecure house-
holds. The BKKBN evaluated households based on the following indicators:
whether all household members regularly consumed at least two meals a day,
whether household members had different sets of clothes for different activities
(home, work, school, and public), whether houses had dirt-floors, whether house-
holds were able to bring their children to health centers to receive medical treatment
when they were sick, whether the households used family planning methods, and
whether household members were able to practice their religious observances. The
program provided highly subsidized rice at a price of Rp 1,000/kg, compared to
average market prices of Rp 3,000/kg. The amount of rice that could be purchased
by beneficiary households was originally 20 kg per month, but this was reduced to
between 10 and 20 kg in 2000 (Tabor and Sawit 2005; SMERU 2008).
Since the program implementation was rushed and the eligibility criteria were
not designed properly (Tabor and Sawit 2001), leakage has been a significant problem
with the Rice for the Poor program. Table 14.3 presents indicators of the effective-
ness of the food security program targeting. The percentage of households that
received benefits from the program fell as household per capita expenditures rose in
both Java and the rest of the country. Only about 30 % of the Rice for the Poor
program beneficiary households were among the poorest 20 % of society. The
targeting performance was slightly better outside of Java where almost 60 % of the
program beneficiaries were among the poorest 40 % of households.
Strikingly almost 20 % of the program beneficiaries were from the highest
income quintile households, even though the program was intended to help the poor.
The misadministration of the program was also evident in Java, where almost 10 %
of the wealthiest households benefitted from the program. This indicated a failure
of the food security program operators to provide oversight of the selection of
14 Consumption Behavior of the Poorest and Policy Implications in Indonesia 233

beneficiaries. Improved targeting would be an effective means for the Indonesian

government to improve program benefits for the poor, but the costs and feasibility
of more narrow targeting need to be assessed in order to identify economically
optimal targeting methods.
The results presented in Table 14.3 show that even a targeted social protection
effort for the poor can be subject to errors of inclusion. The errors of inclusion in the
implementation of the Rice for the Poor program have several implications. Based
on the latest IFLS community data, the program was the most well-known among
the poorest and was considered the most beneficial social protection program.
However, it was also the second most criticized program due to the lack of program
socialization and a suitable mechanism for filing complaints.3 There were other
consequences related to the limited program impact on the poorest households. In a
similar case of a conditional transfer program in Honduras, there was no positive
impact on household nutritional status due to implementation problems (Adato and
Hoddinott 2009). In the case of the Rice for the Poor program in Indonesia, target-
ing seems to be a bottleneck that limited the program’s impacts.
The evidence that the Rice for the Poor program had limited impacts on the poor-
est is presented in Table 14.4. The program’s impacts on food consumption were
evaluated at the household level after program implementation and the resulting
changes in consumption are presented. The program benefitted the poorest house-
holds by increasing meat consumption by 50 % at the time of post-exposure pro-
gram disbursement. “Dairy products” category expenditures increased by about
90 % among the poorest households. Though animal source foods such as meat and
dairy products contain more nutrients, the increased consumption of these catego-
ries by the poorest households was attributed to a shift to “better-tasting” food
(Jensen and Miller 2008; Banerjee and Duflo 2011).
As previously mentioned, the food security program seems to have had many loop-
holes. By design the Rice for the Poor program provided generous support to the
households that were suffering most from the crisis. Although it was found that the
program reached the intended households, the mistargeting of households was also
clear and lead to unintended program effects. Since the poorest households have lim-
ited understanding of dietary nutrition, accompaniment of the food subsidy interven-
tion seems to be a prudent option. Changing food consumption behavior of the poor
might require time and education rather than be an immediate response to improved
food availability. In addition, the fact that extra resources from cash transfers are often
used to purchase unhealthy goods that are consumed by adult males such as cigarettes
is another indication that government intervention to improve household food con-
sumption should consider an accompanying educational component.

3
According to the IFLS community data from 2007 the unconditional cash transfer and the Rice
for the Poor program received the most complaints from communities. Among 321 enumeration
areas, complaints about the unconditional cash transfer and the Rice for the Poor programs were
made by 69 and 16 % of the areas respectively. Despite this level of complaints, more than 60 %
of the areas reported that the Rice for the Poor program was the most accepted program because it
indeed helped the poor.
234 E.H. Pangaribowo

Table 14.4 Average treatment effects on food and nonfood consumption by the poorest households
in Indonesia (Based on IFLS data)
Outcomes Current consumption Consumption change
Total food expenditures 0.008 0.029
(0.032) (0.047)
Rice 0.181 0.055
(0.231) (0.300)
Other staple foods 0.097 −0.198
(0.167) (0.254)
Dairy products 0.283 0.889***
(0.368) (0.388)
Meat 0.649*** 0.278
(0.212) (0.220)
Fish 0.298 0.228
(0.292) (0.320)
Alcohol and tobacco products 0.475 0.204
(0.358) (0.375)
Nonfood expenditures −0.003 0.049
(0.037) (0.055)
Medical 0.269 0.532
(0.281) (0.362)
Education −0.272 −0.025
(0.372) (0.353)
Standard errors are shown in parentheses, ***denotes statistical significance at 1.0 % level

Although the Rice for the Poor program beneficiaries were mostly determined by
BKKBN welfare criteria, local village authorities were responsible for the distribu-
tion of benefits at the village level. Addressing the political and bureaucratic causes
of exclusion might be an important pathway towards more effective targeting (von
Braun et al. 2009). The Rice for the Poor program had little bottom-up process in
the implementation, particularly the involvement of households as the ultimate
point of distribution. Analysis of the IFLS community data revealed that the village
head and associated staff, along with the BKKBN cadre and the head of neighbor-
hood associations determined which households were eligible for the Social Security
Net programs, including Rice for the Poor. The data analysis results further revealed
that only 50 % of the villages had mechanisms in which a household could have
applied to be a program recipient if they felt that they qualified for receiving subsi-
dized rice but were not selected. The long bureaucratic distribution process com-
bined with local and cultural contexts might have restricted the eligibility of some
community members, particularly the ones with the lowest local socioeconomic
status, from being involved in determining the program’s application.
Many studies have highlighted the occurrence of elite capture in development
programs and have recommended community-based development program approaches
to reduce this problem (Mansuri and Rao 2004; Beard and Dasgupta 2006; Fritzen
2007). This problem appears to have affected the Rice for the Poor program. Making
the program more reflective of people’s needs and involving bottom-up participation
14 Consumption Behavior of the Poorest and Policy Implications in Indonesia 235

are potential methods of improving the design of food security programs, just as they
have benefitted other community-driven development programs (Rao and Ibanez
2005). An inclusive and transparent process of recipient eligibility identification is
one of the operational aspects that would help mitigate this problem.

14.5 Conclusion

Understanding the food consumption behavior of the poorest and their responsiveness
to price and income changes are vital for the improvement of their nutritional welfare. It
was found that the poorest households’ consumption behavior will have impacts on
the effectiveness of food subsidy efforts, and should therefore be taken into account
when such programs are designed to avoid unintended or detrimental effects. In order
to maximize the nutritional benefits of food subsidy programs, side-interventions
such as nutrition education should be considered. Food subsidy programs like Rice for
the Poor might also benefit through a focus on gender dimensions. Some studies high-
light the growing recognition that public welfare efforts that specifically target women
are more likely to perform better in terms of household welfare outcomes (Quisumbing
and de la Briere 2000; Doss 2005; Quisumbing and McClafferty 2006).
Food policies in Indonesia are still evolving as food issues respond to the complex
dynamics and uncertainties of multiple factors, not only economic aspects, but also
social and political dimensions. In terms of social inclusion of the poorest households,
applying conditionality on food program recipients might be another effective policy
approach to improve the performance of public assistance efforts. Understanding the
fact the poorest households are involved in less formal communal activities, side inter-
ventions for the poorest household could consider a beneficiary selection approach
featuring these activities. Target households of food assistance programs could be
required to attend community-based nutritional education classes in order to receive
assistance to broaden the welfare benefits of government interventions. Simultaneously,
food policies that involve local-level participation and that accommodate local con-
texts might provide support not only in terms of monetary value, but also in nonmon-
etary terms such as stimulating collective action and public empowerment.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Adato M, Hoddinott J (2009) Conditional cash transfer programs: a ‘magic bullet’ for reducing
poverty? In: von Braun J, Hill RV, Pandya-Lorch R (eds) The poorest and hungry: assessments,
analyses, and actions. International Food Policy Research Institute, Washington, DC,
pp 299–306
236 E.H. Pangaribowo

Ahmed AU, Hill RV, Smith LC, Wiesmann DM, Frankenberger T (2007) The world’s most
deprived: characteristics and causes of extreme poverty and hunger, 2020 discussion paper 43.
International Food Policy Research Institute, Washington, DC
Ahmed AU, Quisumbing AR, Nasreen M, Hoddinott J, Bryan E (2009) Comparing food and cash
transfers to the ultra poor in Bangladesh, Research monograph no 163. International Food
Policy Research Institute, Washington, DC
Akita T (2003) Decomposing regional income inequality in China and Indonesia using two-stage
nested Theil decomposition method. Ann Reg Sci 31(7):55–77
Banerjee A, Duflo E (2011) Poor economics: a radical rethinking of the way to fight global poverty.
Public Affairs, New York
BAPPENAS (2006) Rancangan Pangan dan Gizi 2006–2011 (Food and nutrition platform
2006–2011). Badan Perancanaan Pembangunan Nasional (The National Development
Planning Agency), Jakarta
Beard VA (2003) Learning radical planning: the power of collective action. Plan Theory 2(1):13–35
Beard VA (2005) Individual determinants of participation in community development in Indonesia.
Environ Plan C Gov Policy 23(1):21–39
Beard VA, Cartmill RS (2007) Gender, collective action and participatory development in
Indonesia. Int Dev Plan Rev 29(2):185–213
Beard VA, Dasgupta A (2006) Collective action and community-driven development in rural and
urban Indonesia. Urban Stud 43(9):1451–1468
Cameron L, Shah M (2010) Scaling up rural sanitation: findings from the impact evaluation
baseline survey in Indonesia. Technical paper, water and sanitation program, World Bank,
Washington, DC
Doss C (2005) The effects of intrahousehold property ownership on expenditure patterns in Ghana.
J Afr Econ 15(1):149–180
FAO (1983) World food security: a reappraisal of the concepts and approaches. Food and
Agriculture Organization of the United Nations, Rome
FAO (2003) Trade reform and food security: conceptualizing the linkages. Food and Agriculture
Organization of the United Nations, Rome. ftp://ftp.fao.org/docrep/fao/005/y4671e/y4671e00.
pdf. Accessed 15 Feb 2012
Fritzen SA (2007) Can the design of community-driven development reduce the risk of elite
capture? Evidence from Indonesia. World Dev 35(8):1359–1375
Fuglie KO (2004) Challenging Bennett’s law: the new economics of starchy staples in Asia. Food
Policy 29:187–202
Grootaert C (1999) Social capital, household welfare and poverty in Indonesia, Policy research
working paper no 2148. World Bank, Washington, DC
Hill H (1992) Regional development in a boom and bust petroleum economy: Indonesia since
1970. Econ Dev Cultur Change 40(2):351–379
Indonesian Bureau of Statistics (2009) Statistical year book of Indonesia 2009. Indonesian Bureau
of Statistics, Jakarta
Jensen RT, Miller NH (2008) Giffen behavior and subsistence consumption. Am Econ Rev
98(4):1553–1577
Mansuri G, Rao V (2004) Community-based and driven development: a critical review. World
Bank Res Obs 19(1):1–39
Maxwell S (1990) Food security in developing countries: issues and options for the 1990s. IDS
Bull 21(3):2–13
Maxwell S (1996) Food security: a post-modern perspective. Food Policy 21(1):155–170
Mukherjee N (2006) Voices of the poor: making the services work for the poor in Indonesia. World
Bank, Washington, DC
Packnawin-Mock J, Jarvis L, Jahari AB, Husaini MA, Pollit E (2000) Community-level determi-
nants of child growth in an Indonesian tea plantation. Eur J Clin Nutr 54:28–42
Perdana AA (2006) Better together or not? Community participation, consumption smoothing and
household head employment in Indonesia, Development economics working paper no 777.
East Asian Bureau Economic Research, Canberra
14 Consumption Behavior of the Poorest and Policy Implications in Indonesia 237

Quisumbing AR, de la Briere B (2000) Women’s assets and intrahousehold allocation in rural
Bangladesh: testing measures of bargaining power, FCND discussion paper 86. International
Food Policy Research Institute, Washington, DC
Quisumbing AR, McClafferty B (2006) Food security in practice: using gender research in develop-
ment. International Food Policy Research Institute, Washington, DC
RAND (2010) The Indonesia Family Life Survey – IFLS4 public release. RAND Corporation,
Santa Monica. http://www.rand.org/labor/FLS/IFLS.html. Accessed 22 Dec 2010
Rao V, Ibanez AM (2005) The social impacts of social funds in Jamaica: a “participatory econometric”
analysis of targeting, collective action, and participation in community-driven development.
J Dev Stud 41(5):788–838
Sadoulet E, de Janvry A (1995) Quantitative development policy analysis. John Hopkins University
Press, Baltimore/London
SMERU (2008) The effectiveness of the Raskin program. Social Monitoring and Early
Response Unit, Jakarta. http://www.smeru.or.id/report/research/raskin2007/raskin2007_eng.pdf.
Accessed 23 Mar 2008
Suara Media (2010) Pengguna Rokok Mayoritas Orang Miskin – smokers are mostly the
poor. Suara Media, Jakarta. http://www.suaramedia.com/berita-nasional/19121-lm3-heran-
pengguna-rokok-mayoritas-orang-miskin.html. Accessed 10 Sept 2010
Sullivan J (1992) Local government and community in Java: an urban case study. Oxford University
Press, New York
Sumarto S (2006) Social safety net: Indonesia. Policy brief 6, Overseas Development Institute,
London. http://www.odi.org.uk/resources/download/1071.pdf. Accessed 23 Mar 2008
Suryahadi A, Sumarto S (2003) Evolution of poverty during the crisis in Indonesia. Asian Econ J
17(3):221–241
Tabor SR, Sawit MH (2001) Social protection via rice: the OPK rice subsidy program in Indonesia.
Dev Econ XXXIX(3):267–294
Tabor SR, Sawit MH (2005) RASKIN: a macro-program assessment. BULOG, Jakarta. http://
www.bulog.co.id/old_website/data/doc/20070321RASKIN_Executive_Summary_for_
Distribution.pdf. Accessed 15 Feb 2009
Timmer PC (2000) The macro dimensions of food security: economic growth, equitable distribution,
and food price stability. Food Policy 25:283–295
Timmer PC, Falcon WP, Pearson SR (1983) Food policy analysis. John Hopkins University Press,
Baltimore
United Nations (1974) Universal declaration on the eradication of hunger and malnutrition. United
Nations High Commissioner for Human Rights, Geneva. http://www2.ohchr.org/english/law/
malnutrition.htm. Accessed 20 Jan 2012
United Nations General Assembly (1948) Universal declaration of human rights. United Nations,
New York. http://www.unhcr.org/refworld/docid/3ae6b3712c.html. Accessed 20 Jan 2012
von Braun J, Bouis H, Kumar S, Pandya-Lorch R (1992) Improving food security of the poor:
concept, policy, and programs. International Food Policy Research Institute, Washington, DC
von Braun J, Hill RV, Pandya-Lorch R (2009) The poorest and the hungry: a synthesis of analyses
and actions. In: von Braun J, Hill RV, Pandya-Lorch R (eds) The poorest and hungry: assess-
ments, analyses, and actions. International Food Policy Research Institute, Washington, DC,
pp 1–61
Chapter 15
Addressing Extreme Poverty and Marginality:
Experiences in Rural China

Ling Zhu

Abstract China’s experiences with addressing extreme poverty and marginality

are presented and analyzed. The evolution of effective and efficient policies, especially
in rural areas is found closely connected to China’s economic growth in the past
three decades, as well as to regional inequality. The “Di Bao” (minimum livelihood
guarantee system) in different areas is reviewed. In developed regions, the “Di Bao”
social assistance program, social insurance and public services have effectively
mitigated the difficulties of the extremely poor. In poor areas, however, addressing
extreme poverty and marginality remains a challenge. Increasing the effectiveness
of poverty reduction efforts in poor areas is, and will continue to be central to the
eradication of extreme poverty in China.

Keywords Poverty • Marginality • Social protection • Social insurance • China

15.1 Introduction

In the past 30 years China has made significant progress in poverty reduction.
According to the poverty line set by the World Bank (US$1.25/day per capita
consumption or income), China’s urban poverty incidence dropped from 44.5 % in
1981 to 0.9 % in 2008, whereas the incidence of rural poverty dropped from 94.2 to
22.3 % during the same period (Chen 2010). According to China’s official poverty
line, rural poverty was reduced from 250 million people affected in 1978 to 26.88
million people affected in 2010, with the overall incidence of poverty dropping
from 30.7 to 2.8 %. Though data vary considerably based on different poverty lines,
they all reflect a trend of substantial declines in poverty in China.

L. Zhu (*)
Institute of Economics, Chinese Academy of Social Sciences, Beijing, China
e-mail: [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 239
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_15,
© The Author(s) 2014
240 L. Zhu

China’s poverty line has remained lower than the international standard. For
instance in 2010, the rural poverty line was defined in terms of per capita annual net
income at 1,274 yuan (US$193), which was equivalent to 21.5 % of the national
average rural income of 5,919 yuan (US$896.82) (NBS 2011a). If converted into
$US at the year-end exchange rate (US$1.0 = 6.6 yuan), the amount was about
US$0.53/day per capita, which is equivalent to 42.3 % of the international poverty
standard. The comparison shows that those rural residents who still live below
China’s poverty line fall within the category of the extremely poor. The previous
antipoverty policies in China have always targeted absolute poverty.
During the period of the 12th 5-year national socioeconomic development plan
(2011–2015) and over the next decade the Chinese government will continue priori-
tizing the elimination of absolute poverty in its antipoverty strategies. The extremely
poor areas that are adjacent to each other (especially areas inhabited by ethnic
minorities, border areas, and ecologically delicate areas) will be the key regions for
carrying out China’s antipoverty program through increasing poverty relief funds
and gradually raising the poverty line (Fan 2011a). However, western China has
experienced a slower pace in poverty reduction than eastern China, despite regional
targeting antipoverty strategies that have been implemented since the 1980s. Among
the current rural poor, nearly two-thirds are scattered across the western areas of the
country (Fan 2011b), which indicates that in order to improve the effectiveness of
antipoverty efforts the Chinese government should come up with more innovative
poverty relief means and targeting mechanisms by taking into consideration the
specific hardships faced by the extremely poor in western China.
In this paper, I focus on the following questions. First, in the last 30 years, especially
during the period 2000–2010, what measures have been taken to alleviate poverty
in rural China? Second, what new policies are being carried out in response to the
most imperative living and development needs of extremely poor households
and individuals? Third, what other crucial measures should be taken to help the
extremely poor escape marginality and poverty? The main primary sources used
in this chapter include policy documents from the central and local governments,
rural poverty monitoring reports released by the National Bureau of Statistics
(NBS) of China, existing research papers, and the results of field work that I con-
ducted personally.

15.2 From Poverty Reduction Programs

to Social Protection System

In its transition from a planned economy to a market economy, China has witnessed
an accelerated process of industrialization and urbanization. In addition, due to the
strict implementation of family planning (the one-child policy in particular), the
problem of an aging population has come to the surface. Meanwhile the government
and the public have developed a deepening understanding of the causes of poverty
and adopted new ideas about social protection in the era of globalization (García
15 Addressing Extreme Poverty and Marginality: Experiences in Rural China 241

and Gruat 2003). The goal of antipoverty policies has extended from guaranteeing
the basic livelihood of the poor to reducing causes of poverty, with an emphasis
on helping the underprivileged achieve their potential and take advantage of oppor-
tunities for self advancement. Such changes have not only prompted the adjustment
of antipoverty strategies and policies according to socioeconomic development, but
also facilitated the establishment of a social security system.
In this period of transition and development, social relief has become the main
source of assistance to the poorest households and individuals. Under the planned
economy during the period of the 1950–1970s, rural society was segregated from
the urban sphere and lacked a social security system. The major programs responsible
for social protection were the “social and disaster relief” and “five-guarantee” policies,
the former providing temporary resources for the poor and the latter providing
long-term resources and services.
The target groups of the five-guarantee policy include: the elderly, the disabled,
and rural residents under 16 years of age who have no ability to work, no income to
support their livelihood, and no legal guardians to take care of them. So far the
categorization has not changed substantially in amendment of the policy (SCC 2006).
The “five guarantees” refer to the provision of food and fuels, clothing and daily
necessities, housing, medical care, and burial services for the target groups (the
policy also guarantees educational expenses for target group children). The resources
of the five-guarantees program, provided by the People’s Commune and production
teams, could only support the most basic needs of poor households when poverty
prevailed in the communes. Since the abolition of the People’s Commune in 1985,
local governments have become the main administrators of the resources provided
through the five-guarantees program, whereas village committees have remained
responsible for caring for the households that enjoy program benefits. In 2010 the
number of rural residents who received relief from the five-guarantee program
reached 5.549 million beneficiaries (NBS 2011a).
The rural economic reform during 1978–1985 strongly promoted agricultural
growth and thus enabled a majority of rural households to attain food security.
However, in areas with insufficient resources and weak infrastructure, a consid-
erable number of rural households did not benefit fully from economic growth
and nearly 102 million people suffered from food shortages. To address this
problem the central government initiated a nationwide antipoverty strategy in
1986 and clearly defined the goal of providing food and clothing for the poor
through promoting development in poor areas (SCC 1989). By 1994 the goal of
food security was virtually achieved. In the subsequent antipoverty schemes the
central goal has been the elimination of income poverty. The key targets have
expanded gradually from poor counties to poor towns and villages, and the main
antipoverty measures so far have been carrying out socioeconomic and human
resource development projects (CDRF 2007). However, these projects are economic
growth oriented and those who have priority in acquiring investment opportunities
are mostly the residents of predominantly poor areas who are not poor or else
belong to the borderline poor. Therefore those who are able to get out of poverty
first are usually not the poorest.
242 L. Zhu

For the extremely poor, even those who have the ability to work, it is difficult to
be granted the opportunities of the commodity production projects in the antipoverty
schemes. For example in 2009, less than 3.0 % of the poor households that had
members capable of working were granted antipoverty funds (NBS 2011b). Two
major reasons have contributed to this inequality. One has to do with the existing
social exclusion, in that those who are not poor have a greater voice than the poor
in the distribution of antipoverty resources. The other reason has to do with the expec-
tations for the projects of the administrative institutions. Given that the extremely
poor are in an unfavorable position with regard to aspects of their living locations,
access to information, open mindedness, health conditions, education levels, working
skills, and management abilities; even if they were granted market oriented investment
projects, they would suffer a much higher risk of failure than other groups. Therefore
the limited participation of the extremely poor in antipoverty projects actually
reflects their marginalized position.
Though the extremely poor find it hard to benefit directly from antipoverty projects,
they have enjoyed an increasing amount of social relief. By 2008 China’s social
relief system was made up of the following policy tools; “relief of basic living,”
“special relief,” “preferential policies,” and “temporary relief.” “Relief of basic living”
largely supports recipients with food and clothing. “Special relief” solves specific
difficulties the recipients encounter in medical care, housing, education, and disasters.
“Preferential policies” refer to the provision of public services free of charge or at
reduced prices. “Temporary relief” is mainly responsible for hardships resulting
from emergencies (MCA 2008). Among them the Minimum Livelihood Guarantee
Scheme (Di Bao in Chinese) provides a social safety net for the extremely poor.
A pilot program was launched in a few provinces and cities in 1996 and was later
extended to rural areas nationwide in 2007. In principle the local governments
provide subsidies with local funds for rural resident households whose per capita
income is below the minimum living standard (the Di Bao line), in order to ensure
that their basic needs are met (SCC 2007). In 2009 the Di Bao program beneficiaries
received an average cash benefit of 816 yuan (US$123.64) per person annually
(Table 15.1). Since 2011 urban and rural Di Bao standards have been linked to a price
index to reduce the impacts of inflation on extremely poor households (CN 2011).
Several major factors have contributed to the enhanced role of social relief. First,
the risks of poverty are increasing in an era featuring a market economy, globalization,
climate change, and the subsequent food, energy, and financial crises. As a result
there has been an increase of newly impoverished residents in areas that have lagged
behind in developed areas, even though aggregate poverty rates have exhibited a
declining trend over time. Moreover the newly poor households are spatially scattered
and it is hard for antipoverty projects to reach all of them. Second, there are increasingly
diverse causes of poverty, such as unexpected disasters, severe illnesses, disabilities,
and the deaths of household income earners, each of which could drive an entire
household into poverty. When there was an absence of a social insurance system or
low payment of insurance benefits, social assistance proved to be a timely means of
protection to relieve income shock on households.
In fact, the Chinese government has actively promoted the establishment and
extension of the “New Type of Rural Cooperative Medical System” and the “New
15 Addressing Extreme Poverty and Marginality: Experiences in Rural China 243

Table 15.1 A national statistical profile of the Di Bao program in China during the period 2001–
2009 (NBS 2011c)
Average
Total Poverty Di Bao cash
number line Average Di Number of participants/ benefits
of rural (Yuan/ Population Bao line Di Bao total rural (Yuan/
people person/ in poverty (Yuan/ participants population person/
Year (millions) year) (millions) person/year) (millions) (%) year)
2001 933.829 869 90.30 – 3.05 0.33 –
2002 935.025 872 86.45 – 4.08 0.44 –
2003 937.506 882 85.17 – 3.67 0.39 –
2004 942.537 924 75.87 – 4.88 0.52 –
2005 949.075 944 64.32 – 8.25 0.87 –
2006 933.913 958 56.98 – 15.93 1.71 –
2007 939.130 1,067 43.20 840.0 35.66 3.80 465.6
2008* 954.048 1,196 40.07 987.6 43.06 4.51 588.0
2009* 946.579 1,196 35.97 1,209.6 47.60 5.03 816.0
2010* 960.000 1,247 26.88 1,316.4 52.28 5.45 –
* The total number of rural people refers to those with permanent rural residential registration. The
totals in 2008, 2009, and 2010 were estimated based on the incidence of poverty. Data were
unavailable for boxes without values

Type of Rural Old-age Security Scheme” initiatives during the 11th 5-year plan
period (2006–2010) in order to address the problem of remarkable increases in pov-
erty associated with illness and aging. By 2010 the Rural Cooperative Medical
System covered 96.3 % of rural residents and in pilot regions (10 % of the counties
nationwide) of the Rural Old-age Security Scheme the number of insured has
reached 102.77 million. Among them, those over 60 years of age receive an allow-
ance of 55 yuan (US$8.3) per month allocated from specific funds. There is no
doubt that both policies help to reduce the risks of poverty.
With regard to the vision for 2020, the Chinese government has developed a plan to
build a social protection system that would cover both urban and rural residents, with
emphasis on the basic pension program, basic medical care, and the Di Bao (CPC
2006, 2007). It can be expected that these three schemes and the ongoing antipoverty
policies will provide more opportunities for the poor who currently have jobs, who are
unable to work, and other vulnerable groups to escape poverty and marginality.

15.3 Coordination Among Di Bao, Public Services,

and Social Insurance Policies: Institutional
Innovations in Developed Areas

According to the Di Bao program “Notification on the Establishment of Nationwide

Rural Minimum Livelihood Guarantee Scheme” enacted by the State Council in
2007, “the target group of the Di Bao program includes rural residents whose net
annual household per capita income is below the local Di Bao line, mainly rural
244 L. Zhu

residents who have been in poverty for many years due to illness, handicap, old
age, physical weakness, the loss of working ability, and adverse living conditions.”
This means that people covered by the Di Bao program are likely those living in
extreme poverty. Thus it can be deduced that the Di Bao line should not be higher
than the poverty line and that the number of Di Bao participants should be at most
equivalent to size of the population living in poverty. However, Table 15.1 pres-
ents data that suggest that since 2008 the number of Di Bao participants has been
larger than the size of the poor population. Since 2009 the Di Bao line has
been higher than the official poverty line. In 2010 the number of Di Bao participants
was larger than the population in poverty by 94.5 %, and the Di Bao line has been
5.6 % higher than the official poverty line. How can this puzzle be explained?
First, the statistics for each group come from different sources. The official poverty
line is defined by China’s central government. Based on the official poverty line, the
NBS estimates the size of the population living in poverty using a national rural
survey sample. The Di Bao line is defined by local governments above the county-level,
with a similar formula to the one used for the calculation of the official poverty
line, but that can be adjusted to reflect local circumstances of socioeconomic devel-
opment. According to the latest guidance from the Ministry of Civil Affairs (MCA),
the Di Bao line should include essential food and nonfood spending of rural residents
for subsistence purposes for a year (clothing, water, electricity, coal or gas, public
transport, daily necessities, etc.) (MCA 2011). Due to significant regional disparities
the Di Bao line is considerably higher in developed regions, so that the average level
of regional Di Bao lines is leveraged over the national poverty line and the number
of the Di Bao participants is larger than that of the poor.
Secondly, the formation of the Di Bao line and the implementation of the Di Bao
policy are highly decentralized. Local governments, village committees, and villagers
all have a certain level of discretion regarding the identification of recipients. As a
result the actual coverage of the Di Bao program may not satisfy both conditions
at the same time: family income lower than the Di Bao line and special personal
difficulties. For instance, although for some families their income is higher than the
national poverty line or the Di Bao line, their family financial status is insolvent
due to spending on the costs of childhood education or serious illness. Commonly
villagers and village committees decide that Di Bao benefits be granted to this type
of family. Statistically, the increase of Di Bao participants outpaced the decrease
of the poor.
In the actual implementation of the Di Bao policy, the various localities observe
the same procedure defined by the State Council, particularly the requirement
for village committees and villagers to take part in decision making. According to
the procedure villagers first submit a Di Bao application to a village committee to
organize either a villagers’ representative meeting or a formal meeting to assess the
financial status of the applicant. After assessment the applications are submitted to
the local township government. Township civil officers are responsible for verifying
the household property, income, work ability, and the actual living conditions of the
applicants, and then submit the results to county-level civil affairs bureaus for
approval. In order to ensure a fair and transparent decision-making process, village
15 Addressing Extreme Poverty and Marginality: Experiences in Rural China 245

committees, township governments, and county civil affairs bureaus make the
application, assessment, review, and approval information public through local
media and community notice boards.
This decision-making process leaves room for local governments and village
committees to engage in institutional innovation according to local conditions.
In developed regions with rapid rates of industrialization and urbanization, local
authorities developed an integrated urban and rural Di Bao system. Urban and rural
residents must follow the same procedures to apply for and obtain Di Bao benefits,
and the only difference lies in the Di Bao line and amount of the allowance. All
families eligible for Di Bao benefits are covered by the Di Bao system. However,
the Di Bao program targets individuals who are in the greatest difficulty, such as the
aged, handicapped, and seriously ill. Recipients at different levels of poverty severity
are entitled to different financial benefit amounts. In identifying the level of assistance
according to the socioeconomic and demographic features of the beneficiaries, local
governments do not have a standardized income threshold for Di Bao. Instead some
parameters have been included in a threshold determination. For instance, the provincial
government in Jiangsu has issued Di Bao guidance lines for the prefectures within
that province with disparate levels of development. The provincial finance provides
part of the Di Bao fund to the prefectures below the average level of provincial
development. In practice, this adds a regional parameter to the Di Bao line.
In the implementation of assistance at the municipal and county government
levels, other parameters related to household size and individual characteristics
have been included. For instance, the Zhangjiagang municipal government delivers
allowances that are equivalent to 140 % of the Di Bao line to the five-guarantee
households (or “Wu Bao Hu,” the rural poor who are infirm, widowed, childless,
orphaned, or have lost the capacity to work). An additional 10 % of the full Di Bao
allowance will be delivered to the following four target groups: (1) the elderly aged
70 years or older, (2) the ill with malignant tumors, kidney transplants, uremia, and
leukemia, (3) the handicapped with a certificate of disability, (4) the single member
households. Moreover, an additional 20 % of the full Di Bao allowance will be
delivered to the two types of target groups other than the five-guarantee households:
recipients without any income source, work capacity, legal support, foster parent, or
support provider, and the blind and/or seriously handicapped with a certificate of
disability (Hong et al. 2006).
Since its inception the Di Bao system has been supported by public service and
social insurance policies. Aside from monthly allowances, the target groups can
also access related policy preferences. The allowances and other social benefits have
been raised along with the increasing local finance revenue and average resident
income. In January 2011 I visited a Di Bao recipient in the village of Xibang, in
Tongxiang county of the province of Zhejiang. There were four members from three
generations in the household. Mr. Jia, the household head, was deaf and his wife
was congenitally retarded. Ten years ago Mr. Jia received a monthly allowance of
150 yuan (US$22.72). Since 2009 his monthly allowance had been increased to
630 yuan (US$95.45). Printed on his Di Bao certificate were other preferential
treatments offered by the county government. Usually the cash value represented by
246 L. Zhu

these assistance policies is far higher than the amount of Di Bao allowance received
by target group recipients. Benefits received by Mr. Jia included:
• A monthly allowance
• Free access to 9-year compulsory education and entitlement to an educational
coupon for high school and vocational school
• Free access to a cooperative medical program and exemption of hospital regis-
tration fees
• Access to financial assistance for those affected by serious illness
• Free access to vocational training and employment placement services
• Exemption of commercial administrative fees for a family-operated business
• Entitlement to legal assistance.
In order to prevent extreme poverty among households whose income is only
slightly higher than Di Bao line, some local governments have identified a special
group of “Di Bao fringe households” (the marginal poor), which includes house-
holds with per capita income of up to twice the level of the Di Bao line and that have
seriously ill or handicapped family members. These households can also access
social protection that combines relief with public services. In January 2010 I visited
a Di Bao household in the town of Shajiabang near the city of Changshu in the
Jiangsu province. Both daughters of the family were married. The parents had
worked in a township enterprise and were both pension recipients. Ms. Xu (58 years
old) suffered kidney disease and needed regular dialysis. Due to her serious illness
she was identified as belonging to one of the marginal target groups of Di Bao.
In 2008 she received a monthly allowance of 116.2 yuan (US$17.61). In 2009
the allowance was increased by to 126.2 yuan (US$19.12) per month. According to
the medical bills shown by Ms. Xu, her dialysis treatments cost 1,600–2,000 yuan
(US$242.42–303.03) per visit. By the end of 2009 her cumulative annual medical
bills amounted to 41,981.78 yuan (US$6,360.76), of which 47.7 % (20,025.92 yuan
or US$3,034.23) were paid by a cooperative medical fund, and 31.2 %
(13,111.29 yuan or US$1,986.56) were paid by the medical relief fund for those
affected by serious illness. These medical relief payments were equivalent to
8.6 times the total of her annual Di Bao allowances (1,514.4 yuan or US$229.45). In
2009 the premium for participating in the cooperative medical care scheme in the
city of Changshu amounted to 300 yuan (US$45.45) per capita, of which 80 yuan
(US$12.12) comes from individual contributions, while the rest was divided among
municipal/township and village finances. Under this funding level the average reim-
bursement ratio for the hospitalized was 45 %, but Ms. Xu’s actual reimbursement
ratio was almost 80 %, reflecting the significant benefits from the social safety net.
Local governments conduct a re-identification of Di Bao recipients once a year.
The purpose of this effort is to include newly impoverished households into protec-
tion, adjust assistance levels, and exclude those who have escaped poverty, although
in reality very few people are excluded from protection. This suggests that the
majority of those who are covered by the Di Bao program are chronically poor
(Ravallion 2010). At the local level an assistance policy has been practiced to
encourage recipients who are of working age and have the ability to work their way
15 Addressing Extreme Poverty and Marginality: Experiences in Rural China 247

out of poverty through increased employment. If a recipient declines a job

introduction for public employment in a service institution three times, the amount
of their allowance will be reduced and ultimately canceled. For households whose
per capita income has stayed above the Di Bao line due to increased employment,
a Di Bao allowance of 3–6 months will be retained before withdrawal from the
program (Xinhuanet 2008).
At the county level or above, social relief, public service, and social insurance
are the responsibilities of different governmental departments. Coordination of
multiple programs for recipients is achieved through information sharing among
the departments, the extension of financial services to townships and villages, and
computer networks. After recipients are approved by the civil affairs bureau, the
names and family information will be transmitted to the other civil affairs bureaus
for finance, education, health care, employment, and social insurance. Benefits in
cash or services that these departments are authorized to deliver will be allocated
according to the names listed for each household. All cash transactions are supported
by a banking service, permitting allowances to be distributed directly to the recipients
account through monthly bank transfers. This has greatly benefited recipient families
and made financial supervision easier.
It should be noted that the above mentioned institutional arrangements in the
developed regions only target residents who have local permanent residential
registration status (hukou). The nonlocal residents (such as rural migrant workers)
and their families have not had the same access to public services and social protection
as permanent local residents. This phenomenon of the social exclusion of migratory
populations is a pressing problem of urbanization that needs to be resolved.

15.4 Linking Di Bao with Nongovernmental

Assistance and Poverty Reduction Programs:
Challenges in Poor Areas

After having been initially confined to a few developed regions, the rural Di Bao
program was expanded throughout China in 2007. So far this program still faces
great implementation challenges in poor areas of China’s central and western
regions. The biggest problems are the large size of the poor population and limited
local finances. Despite special financial support from the central government, there
is still a considerable part of the poor population that is not covered by the Di Bao
program. In addition, Di Bao allowances are low and play a very limited role in
improving the well-being and development of extremely poor recipients.
First, local financial capability has become the dominant criterion for initiating
the Di Bao target identification process. According to the amount of available
finances, provincial governments first identify the proportion of coverable Di Bao
target group recipients of the total rural population (Di Bao ratio) and then break
down the value of this indicator to different administrative areas. These breakdown
quantities are further divided by county and township governments and are finally
248 L. Zhu

converted to individual quotas for villages. In this respect Di Bao coverage is largely
dependent on local government allocation of financial resources, rather than the
actual scope of the population below the Di Bao line. For instance in 2007 the
Qinghai provincial government identified a rural Di Bao ratio equivalent to 6–7 %
of the population. In the Guoluo prefecture, which is located on a plateau at 4,200 m
above sea level, poverty incidence was as high as 12.5 %, but the approved Di Bao
ratio was only equivalent to 10 % of the population (Ding 2010).
Second, with given Di Bao quotas, village committees usually identify the
poorest families based on their own ranking of poverty according to family property
and labor status. Then they present a public notice with a list of Di Bao candidate’s
names, including the seriously ill, handicapped, or aged. This practice apparently
departs from official protocol, but has been popular among villagers because the
official Di Bao line is only based on per capita income at certain points in time. The
villagers’ perception of poverty is based on the long-term earning capacity of farm
households. Moreover, for village committees the costs of conducting an income
survey of all individual farm households precludes such an alternative. Taking
advantage of societal familiarity and participatory appraisal is undoubtedly an
option of socioeconomic rationality. When the number of local poor households
exceeds Di Bao recipient quotas, village committees prefer to identify individuals
that are the neediest rather than entire households. This actually increases Di Bao
coverage at the village level by reducing household allowances. Moreover, evenly
sharing resources is a tradition of rural grassroots society and therefore the deter-
mination of recipients through village committees is easily recognized and accepted
among villagers.
Against this backdrop the assessment of Di Bao targeting accuracy based on the
present sample survey statistics could deviate from reality, but this approach
helps researchers assess the socioeconomic impacts of the program through under-
standing the characteristics of Di Bao recipients. The national rural poverty monitoring
sample surveys conducted by the NBS are just such an authoritative avenue of
procuring information. This long-term annual survey effort has been conducted in
592 key poor counties across central and western China. In 2009 the survey sample
consisted of 5,400 villages, 53,270 farm households, and 225,298 people. Survey
results indicated that 5.3 % of the total sample population had an income level
below county level Di Bao lines, while 9.6 % received Di Bao allowances that year.
It seems that the Di Bao program can cover all qualified candidates, but in reality
only 12.2 % of the eligible candidates received Di Bao allowances. It should
be noted that among sample households that had Di Bao recipients, about 7.5 %
were households with relatively large businesses or farming operations and 5.9 % had
family members that were village cadres (Wu 2011), indicating that even within the
narrow coverage of Di Bao there was welfare leakage.
As for demographics, most Di Bao recipients are the aged, handicapped, or
under-aged. Among Di Bao recipients in the key poor counties, the above-mentioned
three groups accounted for a total of 55.6 % of the program beneficiaries. According
to MCA statistics this ratio was 56.4 % for China’s rural Di Bao recipients (Fig. 15.1).
In 2011 a Beijing Normal University task force conducted a five-province rural
15 Addressing Extreme Poverty and Marginality: Experiences in Rural China 249

43.64% 34.45%

The Elderly

Age<16

The Disabled

Other

12.53%
9.38%

Fig. 15.1 Composition of Di Bao recipients in rural China at the end of 2009 (MCA 2010)

survey in Anhui, Fujian, Jiangxi, Henan, and Shaanxi that found that the disabled
accounted for about 43.8 % of Di Bao recipients (Zhang 2011). Despite different
standards and methods across regions regarding the selection of Di Bao recipients,
the results were similar. This means that when village grassroots cadres and villagers
made adjustments to the Di Bao poverty line, they focused on the causes of household
and individual difficulty. Hence these adjustments are helpful for targeting those
that face real need. In fact when local civil affairs bureaus approve a Di Bao recipient’s
eligibility the most common benchmark is justification due to poverty. In Suichuan
county in the Jiangxi province (one of the key poor counties) natural disasters are
also contributors to transient poverty (Fig. 15.2).
In 2009 the average annual net income of rural Chinese households reached
5,153 yuan (US$780.76) per capita. With this reference the statistics presented in
Table 15.2 can be explained as follows. First, the per capita net income of rural
households in the key poor counties amounted to roughly 2,842 yuan (US$430.61),
which is not yet above the poverty line set by the World Bank, but represents 55.2 %
of the national average income. The per capita income levels of registered poor, Di
Bao, and lowest-income households were 50.5, 46.8, and 20.8 % of the national
average respectively. Second, since 2004 public transfer funds for farmers have
increased rapidly. In the calculations for the values presented in Table 15.2, ten types
of public transfer funds were included, such as agricultural production subsidies,
Di Bao allowance, subsidies for returning farmland to forest, the disaster relief and
assistance fund, etc. For farm households in the key poor counties public transfer
funds have become their third largest source of income. They account for 11.6–17.6 %
of the net per capita income of low-income groups. Third, among all types of farm
households Di Bao households get the largest share of public transfer funds. Public
transfer funds for Di Bao households were equivalent to 1.77 times the average level
of the total sample and 3.44 times the lowest income group. The reason for this is
that both agricultural and rural development policies provide preferential treatments
to Di Bao households. Di Bao eligibility has already become a symbol that attracts
social attention and public support.
Public support for the extremely poor in poor areas is far from sufficient. The
statistics presented in Table 15.2 indicate that farm households receive meager
amounts of Di Bao funds. Even Di Bao households only receive a per capita monthly
250 L. Zhu

23% 44% Catastrophic Disease Affected

Orphans and Widows

3%
Disabled

Disasters Affected
17%
13% Other

Fig. 15.2 Composition of Di Bao recipients in Suichuan County, Jiangxi Province of China in
2008 (Wu 2011)

Table 15.2 Composition of per capita annual net income of rural households in the state designated
key poor counties in 2009 (aggregate data provided by the NBS rural poverty monitoring sample
survey conducted in 2009)
Di Bao
Registered poor participating Bottom decile
Total sample households households income group
n = 53,270 n = 10,015 n = 5,859 n = 5,327
Amount Share Amount Share Amount Share Amount Share
Income source (Yuan) (%) (Yuan) (%) (Yuan) (%) (Yuan) (%)
Per capita annual 2,842.1 100 2,600.0 100 2,408.5 100 1,068.7 100
net income
Wages 1,011.2 35.58 833.8 32.07 630.2 26.17 325.0 30.41
Family-run 1,522.4 53.57 1,392.2 53.55 1,291.9 53.64 602.0 56.33
economy
Returns to assets 40.4 1.42 38.4 1.48 32.6 1.35 9.1 0.85
Private transfer 29.3 1.03 27.6 1.06 29.8 1.24 9.1 0.86
Reimburse to 10.5 0.37 11.8 0.45 10.7 0.44 1.5 0.14
medical
expense
Di Bao cash 19.3 0.68 53.4 2.05 162.1 6.73 24.1 2.26
benefits
Other public 209.0 7.35 242.8 9.34 251.2 10.43 97.9 9.15
transfer
The registered poor households accounted for 18.8 % of the total sample. Among the registered
poor households Di Bao households made up 58.5 %, while the bottom decile income group made
up 26.7 %. Moreover 8.0 % of the households in the bottom decile income group received Di Bao
allowance. Possibly this group contains many transient poor

allowance of 13.5 yuan (US$2.05). Private transfer income accounts for around
1.0 % of the net income of farm households. The relatives and friends of the poor
are usually poor as well, and mutual cash donations and in-kind assistance are
both limited, although labor assistance from relatives and neighbors are not included
in the statistics. During the field work conducted in the Qinghai-Tibetan plateau in
2007, I observed that most of the Di Bao recipients spent their allowance on food.
They relied partly upon medical assistance to resolve difficulties regarding access to
15 Addressing Extreme Poverty and Marginality: Experiences in Rural China 251

health care. Community members took part in assisting disabled Di Bao recipients
(including five-guarantee households) to maintain daily activities and social interac-
tions. Assistance from monasteries makes sure that the locally recognized poor
not only receive material aid, but psychological comfort as well. Private assistance
and social assistance went hand-in-hand for the same target groups; however, this
type of private assistance is not an institutionalized and regular activity. Despite
contributing to the subsistence of the poor, it cannot eliminate developmental barriers
for the recipients and help them escape poverty and marginality.
China’s poverty reduction policies have always featured poverty elimination
through development. In key poor counties a special poverty reduction fund appropri-
ated by the central government is a major source of income for county governments.
Hence in the process of expanding the Di Bao program from the urban to the rural
sector, the Chinese government has attempted to link Di Bao with poverty reduction
policies. In 2008 and 2010 the State Council clearly identified the target groups of
the Di Bao program and poverty reduction policies in the pilot work documents.
The target group of the poverty reduction policies was defined as “rural residents
whose annual per capita family net income is below the rural poverty line and who
have working capacity or the willingness to work, including rural Di Bao recipients”
(SCC 2010). The problem is that the combination of the two programs has not
been achieved because there were no substantial changes to the regional targeting
mechanisms of the poverty reduction program and the government departments in
charge of the different programs do not use the same information platform. It does
not matter whether beneficiaries are former Di Bao recipients who have the capacity
to work or poor households that withdrew from the Di Bao program, neither have
received the expected poverty reduction support (Wu 2011).
Drawing upon policy coordination experiences of developed regions, the following
approaches can be attempted in poor areas:
• First, treat the extremely poor, general poor, and marginal poor people as an
“axis” of connection among the various public support programs.
• Second, establish an information sharing system. In developed provinces the
databases of poor households and individuals in the civil affairs bureaus system
provide an information platform for all government departments and public
service institutions. Using this platform various agencies exchange data and
share information. This system certainly saves organizational costs and enhances
poverty reduction intensity. In poor areas the existing database of registered
poor households could be used as a foundation to further build an adequate
information platform.
• Third, expand existing poverty reduction programs to individual households.
For the extremely poor households and individuals, access to socioeconomic
development is deprived due to difficulties with access to road networks, electrical
power, telecommunications, drinking water, housing, sanitary equipment,
education, medical care, technological extension, and financial and information
services. There used to be poverty reduction projects in all of these sectors, but
they often targeted service providers rather than households and individuals.
252 L. Zhu

Even for programs that target households, most investment avenues require
matching funds from the households. The extremely poor cannot provide matching
funds and therefore they are usually excluded (Cao et al. 2011). These problems
are not addressed in the implementation of the Di Bao program. Hence the key
to improving the social benefits of poverty reduction programs and linking poverty
reduction programs with the Di Bao system lies with changing the targeting
mechanisms and investment approaches of poverty reduction programs.
• Fourth, expand the intervention areas of poverty reduction policies. The Di Bao
program only provides subsistence guarantees to the extremely poor. The food
security objective of poverty reduction policy emphasizes freeing the poor from
hunger. In order to prevent the intergenerational transmission of poverty it is
important to start with nutritional interventions for women during pregnancy and
nursing infants, as well as for children. Some nongovernmental organizations
have already achieved successful outcomes with childhood nutritional interventions
(CDRF 2011). The government needs to establish long-term intervention programs
in order to extend these experiences to all poor regions.
• Fifth, increase project management funding. China’s central government has
launched a series of new long-term programs for social relief, social insurance,
public service, and poverty reduction. Project funds have also been allocated to
key poor counties, however, these special transfer funds generally do not include
any support for program management. The local budgets of all of the key poor
counties run on deficits. In order to maintain project operations the managerial
agencies either appropriate funds from other sources or charge the recipients
for assistance rather than providing it for free. In addition, it is common to see
inadequate project management because better management would cost more
(Zhu 2006). Successful experiences in developed regions show that intensive
management is needed for projects in order to target the specific difficulties of
extremely poor households and individuals. In order to improve the efficiency
of social assistance it is imperative for the central and provincial governments to
allocate increased project management funds to the poor counties.

15.5 Summary

China has developed its rural social protection system during a process of socio-
economic transformation. Relevant policies stress the following key points. First,
they intend to ensure the subsistence of the poor. Second, they intend to mitigate
poverty-inducing risks. Third, they intend to help poor households develop their
own labor potential, take advantage of development opportunities, and get their
families out of poverty.
In developed regions the Di Bao social assistance program, social insurance, and
public services appear to have effectively mitigated the difficulties of the extremely
poor. In relation to this, children from poor and marginal poor families have been
guaranteed access to education and health care. Family members with the capacity
15 Addressing Extreme Poverty and Marginality: Experiences in Rural China 253

to work receive free vocational training and employment support. These comprehensive
policy interventions and intensive social investment programs tend to help recipient
families escape poverty traps and prevent intergenerational poverty transmission.
Sound targeting mechanisms and delicate project management are prerequisites for
the interventions to achieve significant effects. It should be noted that rural migrant
workers and their families have failed to receive the same policy treatment as local
permanent residents.
In poor areas there are significant difficulties in fighting extreme poverty and
marginality. First, the size of the extremely poor population is large, but locally
available financial resources are very limited, and as a result the intensity is inadequate
despite the implementation of comprehensive poverty interventions. Second, the
issue of linkage building between the Di Bao program and other poverty reduction
programs has been put on the agenda, but this policy objective has not yet been
achieved. The reason for this is the lack of coordination between program administration
institutions, different targeting mechanisms, and different information platforms.
Third, insufficient coverage and welfare leakage exist in both programs.
Increasing the effectiveness of poverty reduction efforts in poor areas is and will
continue to be the key to the eradication of extreme poverty in China, but will require
a number of changes. First, increasing the transfer of the Di Bao fund from the
central government to local governments in poor areas would help provide subsis-
tence security to all extremely poor households and individuals. Second, while
implementing comprehensive interventions on economic risks encountered by the
extremely poor, the government and public should also take actions to eliminate
social exclusion. This may not bring any significant changes in poverty incidence in
the short-term, but is an essential step to eradicate poverty and promote inclusive
development over the long term. For this reason it is necessary to make adjustments
to the indicators used to assess local government efforts. A shift should be made from
over-attention on changes in the incidence of poverty to an increasing consideration
of how much the poor are benefited from multi-dimensional intervention policies.

Acknowledgements I appreciate the commentary and information contributions from Zhongyi

Jiang, Guobao Wu, and Xiulan Zhang, as well as for the illustrations by Chengwu Jin and Wei He.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Cao H, Wang X et al (2011) Multi-dimensional poverty measurement and intervention for special
types of poor areas—case studies on Aba Tibetan and Qiang autonomous prefecture in Sichuan
Province. Agricultural Press of China, Beijing
CDRF (2007) China development report 2007: eliminating poverty through development. China
Development Publishing House, Beijing
254 L. Zhu

CDRF (2011) Boarding primary school nutrition improvement plan. China Development Research
Foundation, Beijing. http://211.144.32.61/project/project.php?cid=18&subcid=869&aid=2299.
Accessed 12 June 2011
Chen S (2010) Monitoring China’s progress against poverty and evaluating its anti-poverty policies.
Paper presented at the CASS Forum, Social Protection and Human Development in the Era of
Post-Financial Crisis, Beijing, 22–24 October 2010
China News (2011) China established dynamic urban and rural minimum livelihood guarantee
scheme linked to price index. http://www.chinanews.com/gn/2011/05-18/3049913.shtml.
Accessed 12 May 2011
CPC (2006) Communiqué of the Sixth Plenum of the 16th CPC Central Committee. http://news.
xinhuanet.com/politics/2006-10/11/content_5190605.htm. Accessed 12 May 2010
CPC (2007) Report to the seventeenth National Congress of the Communist Party of China. http://
news.xinhuanet.com/newscenter/2007-10/24/content_6938568_7.htm. Accessed 15 Oct 2007
Ding S (2010) Social relief and non-governmental assistance in rural Tibetan inhabited areas. In:
Wang L, Zhu L (eds) How to get out of the poverty trap. Economic Management Press, Beijing
Fan X (2011a) China strives to eliminate absolute poverty in ten years. Hong Kong Commercial
Daily. http://www.cpad.gov.cn/data/2011/0121/article_343820.htm. Accessed 13 Apr 2011
Fan X (2011b) Make contiguous extremely poor areas the main battlefield of anti-poverty efforts.
http://www.cpad.gov.cn/data/2011/0307/article_343991.htm. Accessed 13 Apr 2011
García AB, Gruat JV (2003) Social protection: a life cycle continuum investment for social justice,
poverty reduction and development. Social Protection Sector, International Labour Organization,
Geneva. http://www.ilo.org/public/english/protection/download/lifecycl/lifecycle.pdf. Accessed
10 Nov 2009
Hong D, Wang H, Li M, Wan H (2006) Guidance and establishment of rural Dibao program compatible
with economic development level—report on rural Di Bao survey in Jiangsu Province. Minimum
Livelihood Security Department, Ministry of Civil Affairs, Asia Development Bank, Beijing
MCA (2008) General research report of social relief system. Department of Social Relief, Ministry
of Civil Affairs, Beijing
MCA (2010) The composition of Di Bao participants in rural China in December 2009. Ministry
of Civil Affairs, Beijing. http://files.mca.gov.cn/cws/201001/20100128094132409.htm.
Accessed 10 May 2011
MCA (2011) Notification on further regulation of urban and rural minimum livelihood standards,
MCA issue 80. Ministry of Civil Affairs, Beijing. http//www.mca.gov.cn/article/zwgk/fvfg/
zdshbz/201105/20110500154356.shtml. Accessed May 2011
NBS (2011a) Statistical Communiqué of the People’s Republic of China on the 2010 national
economic and social development. National Bureau of Statistics, Beijing. http//www.stats.gov.
cn/tjgb/ndtjgb/qgndtjgb/t20110228_402705692.htm. Accessed10 May 2011
NBS (2011b) The poverty monitoring report of rural China 2010. NBS division of rural socioeco-
nomic survey. China Statistics Press, Beijing
NBS (2011c) Unpublished data provided at a meeting about the poverty dynamics organized by the
Poverty Reduction Office under the State Council, Beijing, 26 November 2010
Ravallion M (2010) A guaranteed minimum income? Theory and evidence on China’s Di Bao
program. Paper presented at the CASS Forum, Social Protection and Human Development in
the Era of Post-Financial Crisis, Beijing, 22–24 October 2010
SCC (1989) Poverty reduction office under the state council, a compilation of documents on
economic development in rural areas. State Council of China, Beijing
SCC (2006) The regulations for rural five-guarantee work, order 456 of the State Council of the
People’s Republic of China. State Council of China, Beijing. http//www.gov.cn/zwgk/2006-
01/26/content_172438.htm. Accessed 10 May 2011
SCC (2007) Notification on the establishment of nationwide rural minimum livelihood guarantee
scheme. State Council of China, Beijing. http//www.gov.cn/zwgk/2007-08/14/content_716621.
htm. Accessed 10 May 2011
15 Addressing Extreme Poverty and Marginality: Experiences in Rural China 255

SCC (2010) Notification on scaling-up the experiment on linking rural Di Bao program with
poverty reduction developmental policies. State Council General Office issue 31, State Council
of China, Beijing. http://www.gov.cn/zwgk/2010-05/11/content_1603979.htm. Accessed 10
May 2011
Wu G (2011) A study on the linkage between rural Di Bao program and developmental poverty
reduction policies, a research paper submitted for the State Council Poverty Reduction Office.
State Council of China, Beijing (unpublished)
Xinhuanet (2008) Xinhuanet.com Jiangsu channel: Suzhou’s social relief expansion picks up
speed, cancellation of assistance to those no longer in need. http://www.js.xinhuanet.com/
suzhou/2008-04/17/content_13000765.htm. Accessed 12 June 2011
Zhang X (2011) Strengthening the antipoverty effects of social assistance in China. Presentation
prepared for Asian Development Bank (unpublished)
Zhu L (2006) Management cost and efficiency of Rural Medical Assistance Program. China
Demogr Sci 4:16–27
Chapter 16
Experiences in Targeting the Poorest: A Case
Study from Bangladesh

Syed Masud Ahmed

Abstract This chapter examines the program “Challenging the Frontiers of Poverty
Reduction” (CFPR) implemented by the Bangladeshi non-governmental organiza-
tion BRAC. The program aimed to lift participants out of extreme poverty within
2 years and facilitate their entry into mainstream development programs. To achieve
this, the program combined interventions specifically tailored for the ultra-poor
with interventions to create an enabling environment for the ultra-poor. Experiences
show that a participatory process involving the local community, and accommodat-
ing local knowledge and wisdom, is the most pragmatic way of identifying the
poorest households in a community. However, meticulous implementation by a
motivated workforce was also key to the success of the effort.

Keywords Bangladesh • Poverty reduction • Targeting • Ultra-poverty

16.1 Introduction

Bangladesh, born out of a devastating civil war and one of the world’s ugliest
modern genocides in 1971, is a land of immense beauty and potential (Haq 1979).
It is a densely populated country with more than 1,100 people per km2. Despite the
regular occurrence of disasters both natural (e.g., floods, cyclones, and tidal waves)
and man-made (e.g., price hikes, food insecurity, political conflicts, military rule
etc.), poor governance, and the limited quality of human capital, Bangladesh has
succeeded in maintaining an annual economic growth rate of around 6 % in recent

S.M. Ahmed (*)

BRAC Research and Evaluation Division, Dhaka, Bangladesh
e-mail: [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 257
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_16,
© The Author(s) 2014
258 S.M. Ahmed

Table 16.1 Differences between poverty groups in Bangladesh with respect to key characteristics
Extremely poor Moderately poor Non-poor
(<1,805 kcal (1,805–2,122 kcal (≥2,122 kcal Significance
Characteristics per capita/day) per capita/day) per capital/day) (%)
No land (% of households) 9.8 8.2 5.4 <1
10 decimals of land 50.3 45.0 33.3 <1
(1 acre = 100 decimals)
(% of households)
Average agricultural day 0.5 0.3 0.1 <1
labor in each household
Average non-agricultural 0.4 0.4 0.2 <1
day labor in each
household
Female-headed 10.8 8.1 10.7 <1
households (%)
Literate (% per household, 29.7 39.4 60.4 <1
≥6 years)
Adapted from BBS (2007)

years. The country has recorded a steady decline in poverty (corresponding to an

upper poverty line of <2,122 kcal/person/day) from 53 % of the population in 1995
to 32 % in 2010, calculated on the basis of the “cost of basic needs” approach (BBS
2011).
One of the distinguishing features of the current poverty alleviation scenario in
Bangladesh is the participation of nongovernmental organizations (NGOs) as part-
ners in development—supplementing and complementing government efforts
(Lovell 1992; Fernando and Meyer 2002; Mondal 2004; Grameen Bank 2005), and
the use of microcredit as “a critical anti-poverty tool for the poorest, especially
women” (Micro-credit Summit 1997). These latter programs extend small loans to
poor people (mainly women) for income generating self-employment activities and
thereby diminish seasonal vulnerability through diversification of income earning
sources, building assets, and strengthening mechanisms for coping with crises
(Rahman 1995; Mustafa et al. 1996; Hussain 1998; MkNelly and Dunford 1998;
Panjaitan-Drioadisuriyo and Cloud 1999; Zaman 2000). Such efforts enhance wom-
en’s income earning potential, improve their role in nontraditional activities in the
informal sector, and facilitate empowerment (Hashemi et al. 1996). Micro-credit
programs have been associated with increased enrollment of poor children in
schools who had either been drop-outs or had never enrolled (Khandker 1998;
Barnes et al. 2001; Chowdhury et al. 2002). The importance of microcredit pro-
grams as a health intervention tool (Schuler and Hashemi 1994; Nanda 1999; Barnes
et al. 2001; Pitt et al. 2003) and as an efficient and equitable tool for directing
resources to women (Kabeer 2001) have also been emphasized in the literature.
The poor in Bangladesh are not homogeneous (Table 16.1). The proportion of
the population falling below the lower poverty-line (corresponding to the consump-
tion of 1,805 kcal per capita/day or less) is variously termed the “extremely poor,”
“hardcore poor,” or “ultra-poor,” and comprise around 25 % of the population
16 Experiences in Targeting the Poorest: A Case Study from Bangladesh 259

(BBS 2007). These households commonly have ≤10 decimals (0.04 ha) of land, no
productive assets, no income-earning male household member, and depend upon
earnings from the wage-labor of female household members for survival. This
group is unable to participate fully in social and economic activities (including
governmental decision making) that have an impact on their daily lives. This social
exclusion denies them access to essential goods and services, such as health care,
which are available to other segments of the population (Santana 2002; Nayar
2007). Experiences have shown that regular microcredit/microfinance programs are
not suited to their livelihood styles for a number of structural reasons. Included
among these are some supply-side factors such as the harsh discipline of microcredit/
microfinance institutions, the absence of a safety net in case of default of loan pay-
ments, and demand-side factors such as aversion to risk-taking and fear of making
cash transactions (Halder and Mosley 2004; Sulaiman and Gulesci 2008).

16.2 CFPR: The Customized Program

for the Ultra-poor by BRAC

Given the fact that regular microcredit/microfinance programs were found unsuitable
for addressing the needs of the ultra-poor, a number of options were attempted
during the development of a program by BRAC. During the late 1980s BRAC
collaborated with the United Nations World Food Programme to develop a program
specifically for assisting destitute women. The program, named “Income Generation
for Vulnerable Group Development” (IGVGD), successfully combined skills
training and microcredit/microfinance with food aid to make participating women
self-sustainable by the end of the 24-month aid cycle, and was later scaled up
nationally. However, evidence showed that the program was unable to provide a
sustainable livelihood for all participants (around 40 % of the women dropped out
of each cycle) and that further re-designing was needed to be able to assist some
households (Matin and Hulme 2003).
Experiences of the IGVGD program and BRAC’s own research (Halder and
Mosley 2004) led to the design of a customized grants based, asset transfer and
development program for the ultra-poor. This second program named “Challenging
the Frontiers of Poverty Reduction” (CFPR), aimed to lift participants out of
extreme poverty within a period of 24-months and to facilitate their entry into
mainstream development programs. To achieve this, the program combined “pushing
down” (interventions specifically tailored for the ultra-poor) and “pushing out”
(interventions to create an enabling environment for the ultra-poor) strategies. The
different components of the CFPR program with associated rationale are shown in
Table 16.2. The particular nature of the asset transfers is decided through a consulta-
tive process between selected ultra-poor women and the local BRAC staff, and
includes livestock, poultry, leased land, wage labor, street vending, etc. (Alarakhia
and Barua 2005).
260 S.M. Ahmed

Table 16.2 The challenging the frontiers of poverty reduction/TUP program components and
their rationale (Matin et al. 2008)
Component Rationale
Integrated targeting methodologies Effective targeting of the
extremely poor
Income-generating asset transfer (Range: Tk 3,000–9,000 Build an economic asset base
[US$50–150])
Income generation skill training and regular refreshers, Ensure a good return from
e.g., poultry/livestock rearing, vegetable cultivation, asset transferred
shoe making etc.
Technical follow-up of enterprise operations Ensure a good return from
asset transferred
Provision of all support inputs for the enterprise Ensure a good return from
asset transferred
Monthly stipends for subsistence [Tk 10/day(US$0.17/day) Reduce opportunity cost of
for 12–15 months] asset operations
Social development, e.g., social awareness and confidence Knowledge and awareness
building, legal awareness, social action on early marriage/ of rights and justice
dowry etc.
Mobilization of local elite for support (pro-poor advocacy Create an enabling
through seminar, workshop, and popular theatre) environment
Health support Reduce costly morbidity

The health component of the program that was tailored specifically for the
ultra-poor consisted of “essential health care” services (maternal health, family
planning, communicable disease control, children’s health, home based TB treatment,
and basic curative care); counseling and consumer information on health and health
care services (to overcome information barriers); free installation of latrines and
tube wells (to overcome disease transmission); identity cards for facilitated access
to health care facilities (to overcome social exclusion); and financial assistance for
diagnostics and hospitalization if needed (to overcome financial barriers). The
health inputs served as a safety net against the income eroding effect of moderate to
severe morbidity. The chronology of the execution of the program is shown in
Fig. 16.1.

16.3 Targeting Process

The most critical step in the implementation of the CFPR program was the targeting
process as selection of appropriate participants is key to the success of the program
(Rahman and Ali 2006; Perkins 2008). A meticulous, evidence and experience
based process with the active participation of villagers and the local BRAC staff was
followed to accomplish this. The successive steps of the process are described below
in Fig. 16.2.
Stage I: Geographical targeting. The districts and sub-districts for the CFPR pro-
gram were identified using spatial poverty and vulnerability maps prepared by the
16 Experiences in Targeting the Poorest: A Case Study from Bangladesh 261

Fig. 16.1 The challenging the frontiers of poverty reduction program activities (Matin et al. 2008)

World Food Programme (Hollema and Begum 2002). Once these were selected,
BRAC staff at the branch level1 used their local knowledge and experience to select
particular village communities with the highest concentration of poor people.
Usually selected sites comprised of a cluster of approximately 100 households
(demarcated by geographical landmarks) that in some cases extend into the neigh-
boring villages when there are insufficient households in a particular village.
Stage II: Participatory Wealth Ranking (PWR). CFPR’s three member selection
team formed at the branch level, comprised of a facilitator, a recorder, and an
organizer, undertake the task of selecting the ultra-poor households at the site level.
The organizer presents the program and invites the villagers to participate in the
PWR meeting during a rapport building tour of the village. The team especially
encourages women to participate in the PWR meeting. In consultation with villagers
a neutral venue and a convenient time for holding the meeting are decided.
On the day of the meeting the villagers sit in a rectangle with the facilitator and
recorder in the middle. The process begins with a social mapping of the site with the
active participation of the villagers, to locate the households with reference to land-
marks such as a ponds, bushes, mosques, schools etc. A map is drawn on the ground
with a stick by the facilitator with the help of the villagers that is later transferred on

1
BRAC has a network of around 3,000 branch offices all over Bangladesh. A three member
(facilitator, recorder and organizer) selection team formed at the branch level undertook the site-
level tasks.
262 S.M. Ahmed

Based on local knowledge, BRAC employees

select a ‘spot’ for CFPR/TUP targeting All the khanas in
the spot

BRAC POs, based on personal observations

and community opinions, determine a target Target is of
area covering approximately 100 khanas approximately
100 khanas
within the spot

The community ranks the 100 khanas by

wealth duringPRA meeting. The bottom one or Khanas ranked
two khanas are then chosen for a in the bottom one
questionnaire survey. or two categories

BRAC POs conduct the survey and

Preliminarily select khanas for the CFPR/TUP
programme. Preliminarily
selected khanas

BRAC ACs or RSS visits preliminarily selected

khanas and finally selects khanas for the
CFPR/TUP programme. Finally selected
khanas

Fig. 16.2 The stages of the challenging the frontiers of poverty reduction program targeting pro-
cess in Bangladesh (Rahman and Ali 2006)

paper by the recorder. Starting from one corner of the map, the villagers and BRAC
staff identify houses serially. With the placement of each house marker, information
on the name of household head, the name of the father of household head, and the
profession of the household head are recorded in a large ledger and on a small card.
Throughout this process the facilitators try their best to encourage participation of
the assembled villagers, especially the women.
The villagers are then asked to rank the households according to their asset
wealth. The facilitator holds up the cards with household information one by one
and asks the assembled villagers to categorize the household according to the level
of poverty among: rich, middle-class, lower-middle class, poor, and ultra-poor.
Usually the participants rank these households into five economic categories, but
sometimes six categories have been used. The bottom one or two are called the
“community defined ultra-poor.” This is a dynamic process and claims and
16 Experiences in Targeting the Poorest: A Case Study from Bangladesh 263

Table 16.3 Group differences: how well did the challenging the frontiers of poverty reduction
program target? (Matin and Halder 2004)
Selected as Not selected as
Variables beneficiary (A) (%) beneficiary (B) (%)
Marital status
Widowed 30 16***
Divorced/abandoned 15 5***
Demographic resources
HHs where husband present but FHHH 17 4***
HHs with physically able husbands 43 71***
HHs with no adult male 36 15***
HHs with working school aged children 18 10***
Assets—land
HHs that do not own arable land 98 88***
HHs that do not own the land their house 62 38***
is located on
Assets—non-land
HHs with no other asset beside the home 56 43***
FHHH female-headed household, HH household
***Indicates significant differences between groups at the 1 % level

counter-claims, arguments, negotiations etc., occur before the villagers arrive at a

consensus. The facilitator ensures that all the voices are heard.
Stage III: Mini survey to cross-check against program specified criteria. On the next
day, the “community defined ultra-poor” households are visited by the BRAC staff
to conduct an on-the-spot survey to verify findings from the PWR meeting and also
to cross-check against some program-defined exclusion (all binding)2 and inclusion
(three out of five to be satisfied)3 criteria for entry into the programs.
A preliminary list of selected households is then made for final verification.
Stage IV: Final verification and selection. The households in the preliminary list are
then scrutinized by senior staff from either regional or the head offices and judged
based on the facts on hand and their expert opinions to finalize the lists of “selected”
households for asset transfer and other inputs. The effectiveness of the “targeting”
process in identifying the extremely poor (ultra-poor) households is confirmed by
empirical findings (Table 16.3) (Matin and Halder 2004).
A number of interesting issues emerged from the PWR exercises. It gave a
chance to the BRAC staff to intimately interact with villagers, to explain the
purpose of the PWR to them, and to understand how “poverty” is envisaged by them.

2
Households were excluded if they were already borrowing from a NGO, a recipient of a main-
stream government antipoverty program, or if there was no adult woman in the household.
3
Five inclusion criteria are: (1) the total land owned by a household must be <10 decimals (0.04 ha)
of land; (2) the household must have no productive assets; (3) there must not be an adult male
income earner in the household; (4) adult women in the household are required to work outside the
home; (5) school age children must work to help support household.
264 S.M. Ahmed

The very detailed knowledge of the poor households backed by community

endorsement make their tasks easier for the subsequent steps. Thus the selection of
households is evidence based and transparent, which facilitates a sense of propri-
etorship of the program by the community. It also kept the staff aware of societal
gender biases, since BRAC staff had to work really hard to get the voices of the
women properly heard and taken into account while the PWRs were in progress.

16.4 Lessons Learned

Much like the “inverse care law” in health care,4 experiences show that the poorest
have the least chance to benefit from any poverty alleviation program unless they
are specifically targeted (Morduch and Haley 2001). Even when targeted, there is a
tendency to select the relatively better-off among the poor (Navajas et al. 2000).
However, targeting is not an end in itself, it is just the beginning of a long process.
There may be different methods of targeting the poorest, but evidence shows that for
success, the intensity of implementation is often more important than methods
(Coady et al. 2002). The CFPR method was particularly successful because the
program put organization-wise emphasis on the rigorous implementation of each
step of the process throughout the scaling-up phases from 2005 onward. The key to
this was facilitating the active participation of villagers in the community mapping
of households and the PWR process. In the mapping phase, disagreements and
confusion can occur about the definition of a household and whom to include as
members, which were ultimately resolved through transparent group discussions
(Rahman and Ali 2006). This reduced the chance of missing any particular house-
hold, especially the (sometimes) invisible households of the ultra-poor.
Mapping is followed by a process of ranking the households according to wealth
and livelihood condition. For enumerating the latter, direct knowledge of the partici-
pants is essential. Sometimes intense debates and quarrels occur about where to
rank a particular household, but ultimately the villagers come to a consensus demo-
cratically (Rahman and Ali 2006). The facilitating BRAC staff questions the rea-
sons behind the ranking before accepting them to ensure that the process is
consistent. The process also helps make the “invisible” households (e.g., destitute
households living at the fringes of the villages about whom most of the villagers are
not well aware) more visible to their communities. The active participation of the
villagers in the mapping and PWR stages makes the selection process transparent
and allows the wisdom of the community to be taken into consideration, which
greatly facilitates implementing the program at subsequent stages.
Another facilitating factor in the targeting process is the development of a
dedicated workforce by BRAC. Acknowledging the fact that the CFPR program

4
The “inverse care law” stipulates that availability of good health care services tends to vary
inversely with the need for it (Hart 2001).
16 Experiences in Targeting the Poorest: A Case Study from Bangladesh 265

would require a different kind of worker than the microcredit/microfinance program,

BRAC recruited a group of fresh graduates, trained them how to interact with the
poorest with compassion and empathy, and managed them with supportive supervi-
sion to achieve desired results. All of these efforts are done to instill a “fresh
perspective and a new work culture” among them, to help the staff view the dis-
course of poverty from an unbiased perspective, and to provide a real-life under-
standing of the stratification of poverty (Matin et al. 2008). A sense of pride and
proprietorship develops, which motivates the staff’s work. Last but not the least, the
objective verification of the information obtained from the PWR through the survey
is absolutely necessary for both allaying any subjective biases or misinformation,
and for matching with program-specific selection criteria. This effectively narrows
down the range of potential beneficiaries and facilitates the selection of the correct
persons/households for the appropriate type of intervention.

16.5 Conclusion

A participatory process that involves local communities and that accommodates

local knowledge and wisdom is the most pragmatic way of identifying the poorest
households in a community. However, for this process to be successful, meticulous
implementation by a motivated workforce is also needed. The targeting process
developed by the CFPR program and implemented by a cadre of motivated workers
helped BRAC reach the poorest of the poor in a consistent and sustained manner.

Acknowledgements This paper is based on work done by researchers of the CFPR group at the
BRAC Research and Evaluation Division as cited in the text. Any errors may be attributed to the
author. Abdur Razzaque of BRAC RED is acknowledged for re-creating the figures.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Alarakhia S, Barua P (2005) Sector scan of TUP enterprises: identifying determinants of sustain-
ability. RED reports. Research and Evaluation Division, Bangladesh Rural Advancement
Committee, Dhaka. http://www.bracresearch.org/reports/sector_scan_report.pdf. Accessed 10
Feb 2013
Barnes C, Gaile G, Kimbombo R (2001) Impact of three microfinance programs in Uganda
USAID-AIMS paper. Management of Systems International, Washington, DC
BBS (2007) Report of the household income and expenditure survey 2005. Bangladesh Bureau of
Statistics, Planning Commission, Ministry of Planning, Government of Bangladesh, Dhaka
BBS (2011) Preliminary report on household income and expenditure survey 2010. Bangladesh
Bureau of Statistics, Planning Commission, Ministry of Planning, Government of Bangladesh,
Dhaka
266 S.M. Ahmed

Chowdhury AMR, Nath SR, Choudhury RK, Ahmed M (2002) Renewed hope daunting challenges:
state of primary education in Bangladesh. University Press Limited, Dhaka
Coady D, Grosh M, Hoddinott J (2002) Targeting outcomes redux. World Bank Res Obs
19(1):61–85
Fernando NA, Meyer RL (2002) ASA: the Ford motor model of microfinance. Q Newsl Focal
Point Microfinanc 3(2):1–4
Grameen Bank (2005) Grameen: banking for the poor. Grameen Bank, Dhaka. http://www.
grameen-info.org. Accessed 10 Feb 2013
Halder SR, Mosley P (2004) Working with the ultra poor: learning from BRAC experiences. J Int
Dev 16:387–406
Haq E (ed) (1979) Meet Bangladesh, 1st edn. Department of Films and Publications, Ministry of
Information, Government of Bangladesh, Dhaka
Hart JT (2001) The inverse care law. Lancet i:405–412
Hashemi S, Shuler S, Riley A (1996) Rural credit programs and women’s empowerment in
Bangladesh. World Dev 24:635–653
Hollema S, Begum NN (2002) Food security assessment in Bangladesh: issues and implications
for vulnerability analysis and mapping. Vulnerability Analysis and Mapping Unit, World Food
Programme–Bangladesh, Dhaka
Hussain AM (ed) (1998) Poverty alleviation and empowerment: the second impact assessment
study of BRAC’s rural development programme. BRAC, Dhaka
Kabeer N (2001) Conflicts over credit: re-evaluating the empowerment potential of loans to women
in rural Bangladesh. World Dev 29:63–84
Khandker S (1998) Fighting poverty with microcredit: experience in Bangladesh. Oxford
University Press, New York
Lovell CH (1992) Breaking the cycle of poverty: the BRAC strategy. University Press Limited, Dhaka
Matin I, Halder SR (2004) Combining methodologies for better targeting of the extreme poor:
lessons from BRAC’s CFPR/TUP program, CFPR working paper 2. Research and Evaluation
Division, Bangladesh Rural Advancement Committee/Aga Khan Foundation Canada, Dhaka
Matin I, Hulme D (2003) Programs for the poorest: learning from the IGVGD program in
Bangladesh. World Dev 31:647–665
Matin I, Sulaiman M, Rabbani M (2008) Crafting a graduation pathway for the ultra poor: lessons
and evidence from a BRAC program, Working paper 109. Chronic Poverty Research Centre,
University of Manchester, Manchester
Micro-credit Summit (1997) The Micro-credit summit: declaration and plan of action. Micro-
credit Summit Campaign, Washington, DC. http://microcreditsummit.org/declaration.htm.
Accessed 10 Feb 2013
MkNelly B, Dunford C (1998) Impact of credit with education on mothers and their young
children’s nutrition: Lower Pra rural bank credit with education programme in Ghana, Freedom
from hunger research paper no 4. Freedom from Hunger, Davis
Mondal SH (2004) Micro credit as an instrument for changing the household economy of the poor
in Bangladesh: a case of PROSHIKA. Impact Monitoring and Evaluation Cell, PROSHIKA,
Dhaka
Morduch J, Haley B (2001) Analysis of the impact of microfinance on poverty reduction, Working
paper series 1014. New York University, New York
Mustafa S, Ara I, Banu D, Hossain A, Kabir A, Moshin M, Yusuf A, Jahan S (1996) Beacon of
hope: an impact assessment study of BRAC’s rural development programme. University Press
Limited, Dhaka
Nanda P (1999) Women’s participation in rural credit programs in Bangladesh and their demand
for formal health care: is there a positive impact? Health Econ 8:415–428
Navajas S, Schreiner M, Meyer RL, Gonzalez-Vega C, Rodriguez-Meza J (2000) Microcredit and
the poorest of the poor: theory and evidence from Bolivia. World Dev 28(2):333–346
Nayar KR (2007) Social exclusion, caste and health: a review based on social determinants
framework. Indian J Med Res 126:355–363
16 Experiences in Targeting the Poorest: A Case Study from Bangladesh 267

Panjaitan-Drioadisuriyo RDM, Cloud K (1999) Gender, self-employment, and micro-credit

programs: an Indonesian case study. Q Rev Econ Financ 39:769–779
Perkins J (2008) Borlaug-Ruan intern report June 5–August 4 2008. http://www.worldfoodprize.
org/documents/filelibrary/images/youth_programs/2008_interns/PerkinsJ_BR08Report_
1CD331520CF13.pdf. Accessed 10 Feb 2013
Pitt MM, Khandker SR, Chowdhury OH, Dl M (2003) Credit program for the poor and the health
status of children in rural Bangladesh. Int Econ Rev 44:87–118
Rahman H (1995) Mora Kartik: seasonal deficits and the vulnerability of the rural poor. In: Rahman
H, Hossain M (eds) Rethinking rural poverty: Bangladesh as a case study. University Press
Limited, Dhaka, pp 234–253
Rahman H, Ali TO (2006) Stories of targeting: process of documentation of selecting the ultra poor
for CFPR/TUP program, CFPR working paper series 1. Research and Evaluation Division,
Bangladesh Rural Advancement Committee and/Aga Khan Foundation Canada, Dhaka
Santana P (2002) Poverty, social exclusion and health in Portugal. Soc Sci Med 55:33–45
Schuler SR, Hashemi SM (1994) Credit programs, women’s empowerment, and contraceptive use
in rural Bangladesh. Stud Fam Plan 25:65–76
Sulaiman M, Gulesci S (2008) Resource booklet: CFPR-II evaluation. Research and Evaluation
Division, BRAC/London School of Economics, Dhaka
Zaman H (2000) Assessing the poverty and vulnerability impact of microcredit in Bangladesh: a
case study of BRAC. World Bank, Washington, DC
Chapter 17
Rural Poverty and Marginalization
in Ethiopia: A Review of Development
Interventions

Assefa Admassie and Degnet Abebaw

Abstract This chapter provides a comprehensive review of Ethiopia’s policy

interventions for poverty reduction and inclusion of marginal population groups,
and discusses a number of indicators that point at significant improvements that
have been achieved in the country. The sector specific policy measures to reduce
poverty are assessed, and the measures targeted at vulnerable groups and in
marginal areas are analysed. The impacts of economic growth on poverty reduction
are traced. Development efforts such as regional development disparity and the
persistence of severe poverty despite concerted anti-poverty efforts and a relatively
high rate of economic growth are highlighted.

Keywords Poverty • Marginality • Economic growth • Vulnerability • Gender

• Disabilities • Economic policy • Agricultural development

17.1 Introduction

Traditionally the Ethiopian economy has been characterized by subsistence farming

that is highly dependent on rainfall and extremely vulnerable to all kinds of shocks.
As a result, an overwhelming majority of the country’s people have long suffered
from extreme poverty and persistent food crises. Political instability, civil conflict,
and misguided economic policies, particularly between 1974 and 1991, have all
contributed to the country’s low level of development.
Despite sustained growth in agricultural production over the last few years, poverty
in rural areas is still severe and more pervasive than in urban areas (MoFED 2008).

A. Admassie (*) • D. Abebaw

Ethiopian Economic Policy Research Institute, Ethiopian Economics
Association, Addis Ababa, Ethiopia
e-mail: [email protected]; [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 269
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_17,
© The Author(s) 2014
270 A. Admassie and D. Abebaw

The combination of recurrent drought, the use of traditional agricultural practices,

fluctuations in agricultural production, underdeveloped communication and road
networks, and a correspondingly inefficient agricultural marketing system, make
illiteracy and poor health characteristic features of rural Ethiopia. Well-being dis-
parity among the rural population is also significant. Many households are trapped
in persistent poverty and caught at the margins of cultural, social, economic, political,
and environmental systems. For instance, in the current 5-year Productive Safety
Net Program (PSNP) the national government intends to provide benefits to about
7.8 million people in order to help them cope with harsh economic conditions
(MoFED 2010). In view of this situation, it is important to assess the policies that
have been pursued by the Ethiopian government to improve the welfare of rural
Ethiopians and how successful these have been at alleviating poverty.
Some of the underlying causes of living at the margins of basic needs are house-
hold specific, while others are external. Household specific problems include the
lack of, or inadequacy of physical inputs (capital, labor, land, etc.), human capital
(skills, innovation, and the ability to cope), and health. External causes include envi-
ronmental and geographical settings, and the interrelationships among different
social groups, which are characterized by natural, inherited, or acquired attributes
such as religion, ethnicity, gender, or occupation.
Understanding the nature of marginality, or being highly disadvantaged in social,
economic, and political terms, enables the design of more effective policy interven-
tions and the identification of development opportunities that fit specific circum-
stances. To gain that understanding it is imperative to have a comprehensive
overview of the policy environment and rural poverty. In this paper we used research
outputs, published data, and government policy, strategy, and program documents to
review the policy measures and program interventions that have been implemented
to alleviate poverty in Ethiopia; the conditions of the people living at the margins of
society; and in particular the general outcome of governmental interventions. The
lack of disaggregated data for geographical, environmental, and demographic vari-
ables precludes a more detailed analysis of their effects.

17.2 Political and Economic Policy Environment in Ethiopia

The government that came to power in 1991 has taken a series of political and eco-
nomic development measures that have had profound impacts on poverty. These
measures focus largely on (a) stimulating economic growth to increase per capita
income in order to reduce absolute poverty, (b) ensuring more equitable distribution
of growth through broad-based growth strategies (addressing relative poverty or
excessive inequality), and (c) ensuring that no section of society is left in persistent
poverty without access to opportunities and resources on the basis of gender, age,
ethnicity, culture, geography, or other factors that lead to marginalization. The fed-
eral government has tried to integrate the Millennium Development Goals (MDGs)
with the country’s overall development strategies.
17 Rural Poverty and Marginalization in Ethiopia… 271

17.2.1 The Political Setting: The 1995 Constitution

The overriding objective of the 1995 constitution is to establish the legal basis for
economic, political, and social freedoms for all Ethiopians.1 The constitution has
fundamentally altered the political landscape of the country, including government
structure, from a unitary state to a federal system encompassing various ethnic-
based entities. The Federal Democratic Republic of Ethiopia now consists of nine
regional states and two city administrations.2 Recently attempts have been made to
decentralize authority. According to Patterson (2007), decentralization brings devel-
opment closer to the community level and could make public service delivery more
efficient and effective.
The Constitution of Ethiopia (FDRE 1995) guarantees human and democratic
rights to all individual citizens and social groups (Article 10). All persons have
equal protection without discrimination on the grounds of color, gender, language,
religion, political or other opinion, property, birth or other status, race, or any other
factor (Article 25). Furthermore nations and nationalities have the right to speak,
write, and develop their own languages; promote their own cultures; establish the
institutions that govern them; and have equitable representation in state and federal
governments (Article 39).
A major obstacle to individual initiatives for growth is the lack of clearly articu-
lated property ownership rights. Article 40 of the constitution stipulates that every
citizen has the right to own private property, however, the ownership of land and all
natural resources is exclusively vested in the state. Ethiopians do have the right to
acquire land without payment and are protected against eviction from these proper-
ties. Pastoralists have the right to use state land for grazing and cultivation without
charge, as well as the right not to be displaced from these lands.
Every citizen has the constitutional right to engage freely in any legal economic
activity and pursue the livelihood of their choice anywhere within national territory, and
the right to equal access to publicly funded social services (Article 41). As social
groups, every nation and nationality has the right to improve living standards and to
sustainable development (Article 43). The constitution also establishes that the duty
of the government to all citizens is to provide equal opportunities to improve their
economic condition, to promote the equitable distribution of wealth, and to provide

1
There is no consensus with regard to some of the content of the constitution. For instance, it is
unknown whether federated states based on ethnicity will ensure peoples’ freedom or lead to sepa-
ration and social conflict, or whether state ownership of land provides greater freedom than private
ownership.
2
Despite claims that regional states are formed based on settlement patterns, language, and cultural
identity there are no strict criteria for statehood. In some cases language groups overlap and yet
form a regional state [e.g., the Southern Nations, Nationalities, and People’s Region (SNNPR)
includes groups that speak more than half of the languages currently spoken in Ethiopia]. Some
regions with ecologically diverse areas and people of different settlement patterns form a state
(e.g., Oromiya, SNNPR).
272 A. Admassie and D. Abebaw

special assistance to the most disadvantaged citizens in terms of economic and

social development (Article 89).
These examples are some of the manifestations of the political good will of the
government with respect to addressing sources of marginalization based on ethnicity,
culture, gender, ecology, geographical location, or any other form. However, some
of the rights enshrined in the constitution may not be fully considered when actually
designing development policies and programs, and during their implementation and
execution. Whether individual and group rights are respected in practice as per the
constitution is an issue that requires empirical study.3

17.2.2 Social and Economic Policy Environment

The federal government formulated a broad-based development strategy to alleviate

poverty, particularly rural poverty, known as the Agricultural Development Led
Industrialization Strategy (ADLI). Based on this policy framework various
economy-wide, sector-specific, and environmental or social group based develop-
ment programs and plans have been drafted and implemented. The government has
clearly indicated that “agriculture should be the starting point for initiating the
structural transformation of the economy” (MoFED 1993).4 The ADLI is the overall
economic development strategy for agricultural development.
Many argue that a global reduction in poverty in the form of a sharp rise in labor
income could result from increased productivity of small-scale agriculture (Lipton
2005). The ADLI is justified on the basis that the vast majority of rural people suffer
from abject poverty. According to the federal government, encouraging small-scale
farmers and pastoralists to efficiently utilize relatively abundant labor and land
resources is the most feasible way to both address food insecurity and ease the prob-
lems of poverty across the country in the short- to mid-term. Growth in the agricul-
tural sector is also envisaged to create linkages to the rest of the economy through
the creation of input and output markets.
Sustainable Development and Poverty Reduction Strategy Paper (SDPRP). The first
comprehensive strategy for poverty reduction in Ethiopia was the SDPRP launched
in 2001. The overarching goals of this strategy were to bring about faster and sus-
tained economic growth, and to extend the benefits of that growth to the poor. The

3
Some measures contradict the constitution and thus require investigation. One policy example is
the Food Security Program (FSP), which facilitates resettlement as one of the ways of addressing
food security through providing land to the landless or those living on degraded land who are willing
to relocate. This program does not encourage movement across regions in contrast to Article 41 of
the constitution (FDRE 1995). In a country-wide context, food insecure households in resource
poor (land scarce) regions are relatively marginalized in terms of access to fertile soils and
adequate sized parcels of land.
4
In the process of development planning, issues such as whether agriculture or industry should lead,
whether a balanced or focused growth strategy should be pursued, and whether markets or the
government should arbitrate economic agents have been recurring controversies that are unresolved.
17 Rural Poverty and Marginalization in Ethiopia… 273

SDPRP considered nurturing small-scale agriculture to be essential for eliminating

food aid dependency, reducing poverty, and boosting economic growth (MoFED
2002). The strategy primarily addressed critical challenges to small-scale agricul-
ture through improved research and extension packages, and by expanding irriga-
tion and training producers in agricultural technology (IMF 2006). The strategy also
pursued decentralization of government and community empowerment activities
(MoFED 2002).5
Plan for Accelerated and Sustained Development to End Poverty (PASDEP). The
PASDEP was the second phase of SDPRP implemented between 2005–2006 and
2009–2010. PASDEP featured people-oriented development efforts to attain the
MDGs by 2015 (MoFED 2006). PASDEP intended to expand education and health
care services, and to continue food security, capacity building, and decentralization
programs. It was also meant to improve subsistence agriculture by consolidating
efforts started under the SDPRP, such as the development of agricultural input, output,
and credit markets; the provision of specialized agricultural extension services; the
establishment of a network of demonstration centers; the expansion of veterinary
services; support for small-scale irrigation; the improvement of land tenure security;
and the generation of off-farm income opportunities (MoFED 2006).
According to MoFED (2011), the PASDEP helped achieve average annual
GDP growth of 11 % from 2005–2006 to 2009–2010, with a corresponding annual
growth rate in per capita income of more than 8 %. Service sector growth contributed
much of this growth, both in terms of the high rate of growth and in the percentage
of the GDP. The agricultural sector grew by about 8 % per annum during the same
period, accounting for about 5 % of the growth in per capita value-added income.
Overall absolute poverty and food insecurity both declined by a margin of around
10 % during the plan period (MoFED 2010). Gross enrollment in the PASDEP
increased by 94 %, health coverage rose by around 89 %, and clean water coverage
by around 66 %. In general, the growth of PASDEP brought relatively fast and
sustained growth with a reasonably tangible reduction in the percentage of people
living below the poverty line. The period was also characterized by high levels of
inflation and irregular rainfall in some parts of the country (MoFED 2011).
Growth and Transformation Plan (GTP). The GTP is the third, 5-year development
plan in continuation of PASDEP that will be implemented from 2010–2011 until
2014–2015. The GTP will differ from previous programs in many contexts (MoFED
2010). Industrialization will be emphasized in addition to agriculture, and the plan
intends to achieve the remaining MDGs. Improvement of the quality of education
and health care services, and the expansion of health care services to include both
prevention and treatment distinguish GTP from previous programs. The plan also

5
Direct involvement of target groups is widely considered to be a precondition for successful
identification of problems and for proper implementation and execution of intervention measures.
Although it is claimed that the SDPRP was designed on the basis of consultation with the target
groups, the contribution of the poor was little more than endorsing those policies (Amdissa 2006).
Development partners, particularly NGOs, did not significantly influence the process of policy
formulation (Bijlmakers 2003).
274 A. Admassie and D. Abebaw

includes efforts to empower youths and women to improve the distribution of

resources and opportunities.
The GTP will continue to support agriculture as a major contributor to economic
growth, albeit conferring emphasis to the industrial sector at end of the plan period.
A special emphasis will be given to areas with high agricultural potential. The intention
of the program is to narrow the gap between low performing farmers and highly
productive farmers through a scaling-up strategy. Development agents and the
extension system will be used as conduits for the transfer of knowledge and good
practices for both crop and livestock production.
Building upon the PASDEP experiences, the GTP will expand joint efforts
between farmers and the government on natural resource conservation, the develop-
ment of underground and surface water sources, and to increase irrigation coverage.
Efforts will be made to enhance the capacity of development agents and farmers,
and to facilitate the transfer of technologies and working practices. The GTP will
improve agricultural marketing through the involvement of farmers’ cooperatives,
modern output market centers, and the private sector. In general the GTP is intended
to improve the lives of the rural people and reduce inequality in all dimensions of
well-being, including: consumption, education, and health. Although the program is
ambitious with respect to its targets, the successful accomplishment of the program
will greatly reduce well-being disparity among Ethiopians due to geographically,
climatically, and culturally induced marginalization.

17.2.3 Sector-Specific Policies Addressing Poverty

and Marginalization

Food Security Program (FSP). The economic situation in rural Ethiopia is diverse,
due in part to a wide range of environmental conditions. In consideration of this
situation the federal government has designed an agro-ecology based and environ-
mentally specific Food Security Strategy based on three major agro-ecological
zones: rainfall adequate areas, moisture stress areas, and pastoral areas. Launched
in 2002, the Food Security Strategy addresses food security in four dimensions: (a)
direct food production interventions, (b) a voluntary resettlement program, (c) an
income diversification program, and (d) a safety net program (Workneh 2008).6
Because the majority of households in rainfall adequate areas produce sufficient
food, these efforts target moisture stress and pastoral areas.

6
Ethiopia has about 140 agro-ecological zones. This diversity of conditions has an immense
bearing on the formulation and implementation of specific local strategies for farming and land
management (MOA/FAO 1984). For instance, because the strategies were designed based on
broad ecological zones, there are pockets of chronically food insecure places within adequate
rainfall zones due to different site conditions that are not considered in the broader strategy
(Workneh 2008).
17 Rural Poverty and Marginalization in Ethiopia… 275

Specific strategies have been used to address circumstances in moisture stressed

and pastoral areas. The main objective is to help communities in these areas withstand
harsh environmental conditions and resulting food production shortages. In the later
stage, the FSP was redesigned structurally into four major components: (a) the PSNP
component, (b) the Household Asset Building Component, (c) the Complementary
Community Investment Component, and (d) the Resettlement Component.
The Productive Safety Net Program (PSNP). The PNSP was launched in 2005 to
address food security, particularly chronically food insecure households, through
facilitating the build-up of household assets, voluntary resettlement, and creating
access to income from non-farming activities (MoFED 2006). These activities focus
on resolving communal problems such as land degradation and the lack of water or
infrastructure. The program provides income through direct public assistance and
food-for-work programs. The PSNP follows two approaches. One approach focuses on
households with physically capable members that receive transfers in return for labor
on public works, including the development of water points, the reclamation and
rehabilitation of grazing areas, and resource conservation or terracing. The second
approach focuses on households with members that are unable to work because of age
or other reasons that receive direct support (MoFED 2006). The program targeted 262
woredas (local administrative districts) to reach around five million beneficiaries.
With respect to the program’s impacts, Gilligan et al. (2008) found tangible evi-
dence that PSNP support payments have improved the well-being of beneficiaries
based on data collected 18 months after the program became operational.
Considering both the PSNP and other food security programs, beneficiaries were
more likely to be food secure and to have improved their production levels and
income generating capacity. These effects were due to the use of improved agricul-
tural technologies and the performance of non-farming business activities. That
study also found that participation in the program did not have crowding out effects
on participation in the labor market or private transfers. Another evaluation of the
PSNP performance found improved food security for 75 % of the beneficiaries in
eight woredas from four regional states (Sharp et al. 2006). Approximately 62 % of
the beneficiaries successfully protected their food assets from depletion and 46 %
received greater access to public services.
Amdissa (2010) found that approximately 75 % of the beneficiary households
reported consuming more or better quality food in 2006 relative to 2005, and that
94 % of these households attributed this change to the PSNP. This study also found
nonparticipants whose well-being decreased during the same period and attributed
this decline to their lack of access to the PSNP. It was found that around 60 % of
beneficiary households avoided having to sell assets in order to buy sufficient food
and that 33 % avoided having to use their savings to buy food. Most households (90 %)
that avoided using assets or savings to purchase food attributed that ability to the
PSNP. Almost half of the beneficiaries stated that they used health care facilities more
in 2005–2006 than in 2004–2005. Half of the beneficiaries reported keeping their
children in school rather than having to withdraw them due to cash or food shortages.
Devereux and Sabates-Wheeler (2006) indicated that almost 25 % of the benefi-
ciaries acquired additional assets or skills in 2005–2006, largely due to the PSNP.
276 A. Admassie and D. Abebaw

The move from purely relief assistance to a mode of assistance that encourages
productive investment helps avoid dependency among target groups and consolidates
social capital (Workneh 2008).
Resettlement Component. This component of the FSP resettles households that suffer
from food insecurity arising from land scarcity and insufficient moisture to areas with
adequate land and water availability (MoARD 2009). The program resettled approxi-
mately 170,000 households by 2006 (World Bank 2006), and between 188,874 (World
Bank 2009) and 205,000 households by 2009 (MoARD 2009). Resettled households
were provided a package of incentives that included being a beneficiary of the PSNP.
Settlers were offered 2 ha of fertile land, simple farm tools, credit for the purchase of
oxen, food rations for 1 year, basic services such as clean water, health care, and educa-
tion, and the option to return to their place of origin if they were unsatisfied with their
new conditions (NCFSE 2003). The program has resettled less than 50 % of the pro-
jected number of households (Tewodaj et al. 2008).
Population and Environmental Policies.7 Traditional farming and land management
methods, coupled with increased population pressure, have had destructive impacts
on the environment such as deforestation and subsequent soil erosion. More than
80,000 ha are deforested every year in Ethiopia because of heavy reliance on fuel
wood and clearing land for farming and grazing. In addition to high levels of degra-
dation and deforestation, the stagnant economic conditions and high population
pressure often reduce the amount of land per capita to far below the minimum
required to support a family.8 According to Desalegn (2008), 87 % of land holders
in the country had only 2 ha or less in 2003 and 37 % of farmers had half of a hectare
or less. The severity of land scarcity varies significantly among regions. Around
62 % of farmers had only 1 ha or less in 2003 and this situation was more severe in
the regional states of SNNPR and Tigray. Ethiopia and Kenya have the lowest aver-
age per capita land holding area among ten East African countries (Omamo et al.
2006). The fragmentation of land holdings continues and as a result it is not only the
supplies of food, fodder, and fuel that are under threat, but also the very land
resources and the livelihoods that depend upon them.
The National Environmental Policy was enacted in 1997 with the goal of preserving
essential ecological processes and life support systems, maintaining biological
diversity, ensuring the sustainable exploitation of natural resources, and improving
public awareness about linkages and trade-offs between the environment and growth.
Implementing the policy is primarily the responsibility of the Environmental

7
The debate on the economic effects of high population pressure has not been resolved. Given the
prevailing structure of the economy and fragmentation of agricultural land, population pressure is
one of the factors contributing to the level of poverty in rural Ethiopia.
8
According to Desalegn (2008), each household requires at least 2.5 ha to sustain the livelihood of
its members with current technologies. It is difficult to determine the exact size of land required,
however, because it depends on regional and site-specific conditions such as soil fertility and
precipitation patterns, as well as household characteristics such as consumption habits and the
crops or livestock produced. The scarcity of land among peasant farmers is reaching critical levels.
17 Rural Poverty and Marginalization in Ethiopia… 277

Protection Authority, but there has not been a specific program designed to address
these daunting environmental problems. Many of the components of rural develop-
ment and food security programs undertaken through the Ministry of Agricultural and
Rural Development (MoARD), such as reforestation, erosion control measures, and
the protection of community natural resource assets such as woodlots and common
grazing areas are expected to have positive environmental impacts (Patterson 2007).
Education Sector Policy. The current Education and Training Policy begun in 1994
and the Education Sector Development Programs (I, II, and III) articulate the pri-
mary principles of educational development in Ethiopia promoting decentralization,
the participation of people and communities, the equitable distribution of educa-
tional services, and development of the diverse cultures and languages of the peo-
ple. The use of the local languages as the medium of instruction at the primary
education level has been the motto of the Universal Primary Education and along
with the expansion of schools into rural areas. The establishment of Farmer Training
Centers (FTC) and Technical Vocational and Educational Training institutions is
intended to provide equitable, relevant, and problem-solving educational services to
the rural poor. These policy directions are also designed to narrow the disparity in
education coverage between localities and genders.
Health Sector Policy (HSP). The first HSP of the current government enacted in
1993 emphasized preventive health care (MoH 2011). The government launched a
20-year, rolling Health Sector Development Program in 1997–1998 to be imple-
mented in phases. This program focuses largely on communicable diseases, com-
mon nutritional disorders, environmental health, hygiene, reproductive health care,
immunizations, the treatment and control of basic infectious diseases like upper
respiratory tract infections, the control of preventable diseases like malaria, and the
control of sexually transmitted diseases, especially HIV/AIDS. The first three
phases of the program have already been completed in 2002, 2005, and 2010 respec-
tively. In all phases, poverty related diseases such as malaria and tuberculosis were
given due emphasis. Other efforts to improve public health and to meet MDG tar-
gets included improving access to primary health care services, increasing aware-
ness about common diseases and their causes, decentralizing and equitably
distributing health care services, and strengthening the collaborative abilities among
the government, communities, and NGOs.
Cognizant of the high interrelation between poverty and health problems, the
HSP has focused on preventative medicine and the decentralization of the health
care services. Accordingly, the policy puts emphasis on the prevention of the most
common poverty-related diseases including malaria, tuberculosis, childhood illnesses,
and HIV/AIDS. The government launched the Health Extension Program that
trained and dispatched health care extension workers to provide basic services to
households without access to basic sanitation, immunization, and other services.
The Health Sector Development Program introduced a 4-tier health care service
system. The first tier consists of primary health care units that include one health
center and five health posts, local hospitals at the second tier, regional hospitals in
the third tier, and a fourth tier consisting of a referral hospital at the national level.
278 A. Admassie and D. Abebaw

17.2.4 Policies for Disadvantaged Areas and Social Groups

In the context of Ethiopia, many citizens live under harsh environmental conditions
and in geographically peripheral areas that are less developed than the rest of the
country. Women, children, orphans, the elderly, and the disabled are expected to
be relatively vulnerable segments of the society in terms of having limited access
to resources and social services. A number of policy measures and programs
were specifically designed to address the problems facing these marginalized
sections of society.

17.2.4.1 Agro-ecologically Disadvantaged Areas

Moisture Stress Areas. Despite some geographical differences, shortages and irreg-
ularities of rainfall have been some of the most critical challenges to the agricultural
sector. This variability causes unpredictable changes in harvest yields, has limited
economic growth, and causes many to depend on food aid for survival. Despite
yield gaps between irrigated and non-irrigated land approaching 40 % between
1997 and 2000, the extension of irrigation use has been limited to approximately
2.5 % of the country’s potential coverage (MoFED 2010). In order to enhance the
available water supply, the government adopted a strategy of facilitating the con-
struction of micro- and medium-scale irrigation schemes, spring diversions, deep
wells, and the construction of multi-purpose large-scale dams in both adequate rain-
fall and moisture stress areas (Ellis and Tassew 2005).
The government envisaged ensuring food security by reducing vulnerability to
rainfall shortages using water harvesting technologies such as the construction
of reservoirs, creating off-farm income generating opportunities, and the voluntary
resettlement of people to more productive and less-populated areas within their
regional states (FDRE 2002).
Pastoral Areas: Most of the residents of the dry lowland areas of the regional states of
Afar, Somali, SNNPR, Oromiya, and some parts of Gambella depend on traditional
livestock husbandry for their livelihoods. These pastoralists regularly move with their
livestock in search of grazing areas and water because they live in environmentally
hostile areas affected by social conflict and recurrent drought. Pastoralists typically
benefit less than other citizens from development efforts because they are geographi-
cally isolated from infrastructure and facilities, and are poorly connected to urban
markets. Poverty in pastoralist areas is generally worse than elsewhere in the country
in terms of conventional human development indicators (MoFED 2006).
The government has demonstrated its political will to address the socio-economic
problems of pastoralist areas. The social sector development programs tailored to
pastoralists such as informal community-based schools, mobile outreach health care
services, and improved veterinary services have been expanded. The government
has established institutions at both the federal and regional levels in predominantly
pastoralist states and other areas that host pastoralist populations.
17 Rural Poverty and Marginalization in Ethiopia… 279

A combination of strategies has been forwarded to address livelihood problems

in pastoralist areas. These strategies involve activities such as providing animal
health facilities to enhance the productivity and quality of livestock, breed improve-
ment, the construction of surface water projects to improve water supplies, and
improving road networks and other transportation infrastructure to create marketing
networks. The encouragement of voluntary resettlement, sedentary farming along
river basins, large-scale ranching, and similar activities have all been included in the
FSP to redress the vulnerability of pastoralists (FDRE 2002).
A 15-year Pastoralist Community Development Project (PCDP) was launched in
2003. Local communities and the federal government contribute 14 % of the project
budget, and the remaining 86 % has been financed by donors. Largely community-
driven, the first phase of the PCDP from 2003 to 2008 was implemented in 30 pas-
toralist woredas in the regional states of Somali, Afar, Oromiya, and SNNPR. These
efforts focused on community-level micro-projects to improve livelihoods. The sec-
ond project phase (PCDP II), which was launched in March 2008 and operates until
2013, intends to enable pastoralists to better withstand external shocks and to
improve their livelihoods. The PCDP II empowers local communities by involving
them in decision making on developmental issues, improves pastoralist early warn-
ing systems, and provides disaster mitigation. Micro-projects focus largely on water
supply improvement, expanding micro-scale irrigation, providing health care and
educational services, rangeland management efforts, and the creation of income
generating schemes through the formation of savings and credit associations.
During the GTP period the main focus in pastoralist areas will be on water resource
development both for livestock and human use through a scaling up strategy that
makes use of experiences gained in other parts of the country. The GTP will also
include similar efforts to those of other programs such as voluntary resettlement, the
expansion of animal and human health care and educational services, improve-
ments to the livestock marketing system including the integration of pastoralists
and domestic investors, and the development of pasture land irrigation schemes.
People at the Periphery: The regional states of Afar, Benishangul-Gumuz, Gambella,
and Somali are designated as disadvantaged regions because of the prevalence of
pastoralist livelihoods and their location at the geographical periphery of the coun-
try. Considering the fact that these areas are relatively marginalized compared to the
rest of the country, the federal government has taken affirmative measures that are
appropriate for the special circumstance of these areas. The federal government
uses a budget subsidy allocation formula that is adjusted on an annual basis. For
instance, in the 2007 fiscal year the size of population, the level of development, and
revenue collection efforts in each region were given weights of 65, 25, and 10 %
respectively in corresponding funding (FDRE 2007). The federal government also
supports these regions through capacity building and advisory services on policy
decisions, and on the planning and implementation of programs and projects. The
various investment proclamations enacted since 1996 provide special privileges to
those who are willing to invest in labor intensive agricultural and related sectors in
these parts of the country. This measure provides opportunities for unskilled rural
280 A. Admassie and D. Abebaw

residents in these areas, improves market access for their produce, and the expansion
of infrastructure and facilities. The ultimate goal of these interventions is to reduce
economic disparity by mitigating the geographic and environmental factors that
contribute to marginality.

17.2.4.2 Policies for Vulnerable Social Groups

Vulnerable or marginalized groups often differ from one country to another due to
unique historical, economic, political, and cultural situations. In most developing
countries women, children, the elderly, and the disabled suffer disproportionately
from hard work, disadvantages with respect to property rights, income inequality,
limited access to education and health care services, limited political participation,
and difficulties in other spheres of life. Because of cultural barriers and a wide-
spread lack of knowledge of legal rights the situation is often more severe in rural
communities.
Ethiopia is a signatory nation of most UN and African conventions on vulnerable
social groups and is expected to incorporate these obligations into its domestic laws
and regulations. The 1995 constitution, the 2002 revised family law, the National
Women Policy, and the National Youth Policy are examples of legislative efforts that
were designed in conformity with the major principles of international conventions
as they apply to the country’s specific circumstances. The various development pro-
grams discussed in this paper (SDPRP, PASDEP, GTP, PSNP, HSDP, etc.) also rec-
ognize the special circumstances of these vulnerable social groups. In addition to
constitutional provisions that affirm the equality of all persons and the need to pro-
tect their rights from any unlawful circumstances, the federal government has
attempted to mainstream the concerns of the different social groups into the differ-
ent policies and programs.
Gender-Related Policies: Women are more likely to suffer from physical abuse,
including rape and early marriage, in addition to being subject to hard work and
problems associated with access to resources and property rights in many parts of
the country. One persistent cultural problem imposed upon Ethiopian women in
rural areas is forced marriage. Mainstreaming gender issues into socio-economic
development plans has been the main strategy to address the concerns of women as
a social group. Institutions have been put in place on the federal, sector, and local
administrative (kebele) levels with different organizational structures to deal with
gender issues. The Revised Family Code which begun to be implemented in 2000,
protects women from forced marriages and imposed a lower age limit for marriage
at 18 years of age in order to reduce health and psychological problems associated
with early marriage. Article 35 of the constitution also included a clause specifically
to remedy the cultural legacy of discrimination and inequality against women
(FDRE 1995). This article stipulates the right of women to own property, including
land, and to participate equally in policy decisions affecting the public and their
personal well-being.
17 Rural Poverty and Marginalization in Ethiopia… 281

The National Policy for Women was launched in 1993 to combat all forms of
discrimination, and to improve access and the involvement of women in all policy
interventions and the activities of development institutions, programs, projects, and
financial assistance of any sort (Demessie and Yitbark 2008). A concrete attempt to
realize the goals of the National Policy for Women was the National Action Plan on
Gender Equality, which was launched in 2001 and operated until 2006, and focused
on poverty reduction, economic empowerment (including improving access to
land), education, reproductive rights, women’s health, the elimination of violence
against females, and institutional mechanisms for the advancement of women in
decision making at all levels. The National Action Plan on Gender Equality was not
integrated with the poverty reduction program SDPRP at the time (Demessie and
Yitbark 2008).
Many notable women-specific programs have been initiatives of the Ethiopian
Women Development Fund and the Women’s Development Initiatives Project.
These programs helped women by creating gender specific income generating
schemes. The Ethiopian Rehabilitation and Development Fund also invested in
small-scale projects that had gender targets of a minimum of 50 % of women as
beneficiaries during the late 1990s (MoFED 2002).
The massive land title registration program has enabled land ownership by
female-headed households. This was a very tangible step to reduce marginality
among women in terms of access to resources in rural areas. Access to land title
certificates is no longer biased against the poor or females because of adequate plan-
ning, public participation, implementation, and dispute settlement mechanisms
(Deininger et al. 2008).
Workneh (2008) argues that cultural and societal norms in rural areas often cre-
ate considerable negative impacts on the nutritional status of women and children,
making them vulnerable social groups. Attempts have been made in the various
phases of the PSNP to consider the participation of women in public works and
asset development efforts (MoFED 2010). Micro-finance institutions (MFIs) also
support women through involvement in income generating activities.
The Health Sector Development Program and subsequent programs targeted
rural households and particularly women through sanitation and disease preven-
tion efforts for waterborne diseases, and services for reproductive health, family
planning, and HIV/AIDS awareness. These efforts aimed to reduce maternal,
infant, and child mortality, and the overall vulnerability of women and children.
The federal government has also established affirmative action measures to
improve women’s access to education (Getachew 2008). All three phases of the
Education Sector Development Program imposed targets to increase the enroll-
ment percentage of female students. In the first phase of the program the target
was an increase in female enrollment from 38 to 45 % between 1997 and 2001
(Amdissa 2008).
The PASDEP also envisaged addressing gender inequality through a holistic
approach. More specifically the plan tried to narrow gender disparity in education
and health by improving women’s health through an extension program operated
282 A. Admassie and D. Abebaw

by female health workers. In addition, it initiated the necessary legislative and insti-
tutional reforms to protect the rights of women. Reducing the domestic work burden
of women by improving access to potable water was also another major emphasis
area (MoFED 2006).
The GTP intends to ensure women’s equal access to resources by building on
previous experiences with land ownership, access to credit, the protection and pro-
motion of reproductive rights, and abolishing harmful traditional cultural practices.
The strategies to be pursued by this program include ensuring full participation in
extension activities, promoting saving and credit services, and creating income gen-
erating activities. The GTP will also engage owners of very small land parcels,
landless youth, and women in nonagricultural income generating activities, provide
business management trainings and credit, and facilitate market access. The GTP
also intends to ensure equitable access to quality primary education in order to nar-
row gender disparity.
Children and Orphans: Article 36 of the constitution stipulates that every child has
the right to life, to the care of parents or legal guardians, and to be free from exploit-
ative or harmful practices (FDRE 1995). The government also accords special pro-
tection to orphans by promoting opportunities for their adoption and education, and
for the advancement of their welfare. The third Education Sector Program explicitly
creates opportunities for disadvantaged children with special needs, or from pasto-
ralist, semi-agricultural, and geographically isolated areas by building formal
schools, alternative education centers, and building schools at the village level
(Amdissa 2008).
The GTP will offer care and support to children, particularly orphans and other
vulnerable children. The education strategy for children with special needs will be
widely implemented along with adult literacy efforts to address socio-economic
challenges within the context of communities. Ethiopia has a long cultural heritage
of supporting vulnerable groups through extended family. Orphaned children are
traditionally adopted by extended family members.
Disabilities: Physically handicapped or disabled people face diverse forms of dis-
crimination. According to Handicap International/CBM (2006) the disabled often
face cultural, environmental, and institutional discrimination. As a result of cultural
attitudes the disabled are often feared or faced with societal biases. Physical inac-
cessibility and harsh working conditions, particularly in rural areas, as well as cus-
tomary laws, prevent the disabled from participation in many economic and social
activities.
In Ethiopia the constitution endows disabled people with equal rights and
opportunities as the rest of society, except in the case of certain physical activities
for which they may not be suited. Disability issues have not been explicitly dealt
with in many development policies and programs (Seleshi 2010). In the PASDEP,
disability was treated only in the context of gender, where it was included in the
discussion on the participation of disabled women in the development process.
In the GTP there is a policy direction that encourages participation of the disabled
in political, economic, and social activities.
17 Rural Poverty and Marginalization in Ethiopia… 283

17.3 Progress in Poverty Reduction

From the material presented above it is clear that the government, the international
community, and NGOs have been trying to help reduce poverty and disparity in the
distribution of the benefits of economic growth across geographic areas and social
groups. During the last 7 years there has been relatively tangible progress in overall
economic performance and poverty alleviation efforts in the country. Unfortunately,
there has not been complete and disaggregated information available to examine the
extent to which these interventions have changed the welfare status of households
in real terms.

17.3.1 The Poverty Situation in Ethiopia

Ethiopia is among the world’s poorest countries. In 2010 it ranked 157th among 169
countries according to the global Human Development Index (HDI) and it continues
to be one of the lowest economic development performers in Sub-Saharan Africa.
Ethiopia has improved its relative position by around 7.4 % between 2000 and 2010
(Table 17.1), due to its relatively higher rate of economic growth over the last
7 years (UNDP 2010). In terms of parity dollars equivalent to purchasing power in
2008, the average per capita income in Sub-Saharan Africa and the rest of the world
in 2010 have been double and 11-fold that of Ethiopia respectively. Because of rela-
tively increased coverage of social services over the last few years, the non-income
HDI values such as life expectancy at birth, maternal mortality, education, and others
show a relatively better picture than the income HDI.
Average measures often conceal distributional disparities among populations.
Poverty is multifaceted and thus quantifying it requires using measures that

Table 17.1 Development indicators from Ethiopia, the region, and the world
Sub-Saharan Sub-Saharan World/
Indicators Year Ethiopia Africa World Africa/Ethiopia Ethiopia
Human development 2000 0.25 0.315 0.570 1.26 2.28
index 2005 0.287 0.366 0.598 1.28 2.08
2010 0.324 0.389 0.624 1.20 1.93
Inequality adjusted HDI 2010 0.216 0.261 0.489 1.21 2.26
GNI per capita 2010 992 2,050 10,631 2.07 10.7
(PPP-2008$)
Non-income HDI value 2010 0.357 0.436 0.663 1.22 1.86
Life expectancy at birth 2010 56.1 52.7 69.3 0.94 1.24
Maternal mortality ratio 2003–2008 720 881 273 1.22 0.38
Mean years of schooling 2010 8.3 4.5 7.4 0.54 0.89
UNDP (2010)
284 A. Admassie and D. Abebaw

Table 17.2 Multidimensional poverty among selected African countries

Population with
Multi- multidimensional % Population with at least one Gini–
dimensional poverty deprivation (2000–2008) co-efficient
poverty Intensity of
indexa Headcount deprivation Living
Country (2000–2008) (%) (%) Education Health standards
Ethiopia 0.582 90 64.7 83.9 48.2 94.2 29.8
Kenya 0.302 60.4 50.0 21.9 41.4 86.2 47.7
Egypt 0.026 6.4 40.4 18.0 16.9 0.9 32.1
Nigeria 0.368 63.5 57.9 42.4 59.5 72.1 42.9
UNDP (2010)
a
The MPI is a product of the multidimensional poverty head count (the share of people who are
multi-dimensionally poor) and the average number of deprivations each multi-dimensionally poor
household experiences (the intensity of their poverty). A household is considered multi-
dimensionally poor if it is deprived in at least two to six of the ten indicators for health, education,
and living standards (UNDP 2010)

reflect this multidimensionality. The Multidimensional Poverty Index (MPI)

complements monetary based methods and also reveals the extent and distribu-
tion of deprivation. The MPI also captures the magnitude and degree of depriva-
tion overlap (UNDP 2010).
About 90 % of the Ethiopian population was deprived access to two or more
indicators related to education, health, and living standard between 2000 and 2008
(UNDP 2010). The scale of multidimensional deprivation was as high as 65 %. Of
the three categories of human development measures, deprivation in living stan-
dards (food, clothing, housing, water, and sanitation) has been greatest. In terms of
practically all MPI criteria, the situation of the Ethiopian people compares poorly to
other populous countries of the region (Nigeria, Egypt, and neighboring Kenya)
(Table 17.2).
The extent of income inequality among Ethiopians (Gini-coefficient) has not
been substantial in comparison to the other three regional countries during the same
period. This situation implies that deprivation and multidimensional poverty in
Ethiopia are phenomena experienced by the majority, and are largely explained by
the lack of economic capacity rather than skewed distribution of resources. This can
largely be attributed to the excessive reliance of the majority of the people on tradi-
tional subsistence agriculture.
Ethiopia has the largest population of livestock and is the third largest contributor
to agricultural production in Sub-Saharan Africa. Agriculture in Ethiopia is character-
ized by high labor and livestock intensity, and the use of traditional practices. Among
the top five agricultural producers on the continent, it utilizes the least number of
tractors and the application of chemical fertilizers is relatively modest. Overall
labor productivity or output per farmer is extremely low. The average farmer in
Ethiopia earns only around 13 % of the annual earnings of a farmer in Nigeria.
Reliance on rainfall and low-technology methods contributes to the extreme poverty
of farming households in rural Ethiopia (Table 17.3).
17 Rural Poverty and Marginalization in Ethiopia… 285

Table 17.3 Agricultural performance of selected Sub-Saharan African countries (Pratt and Yu 2008)
Performance indicators/country Nigeria Sudan Ethiopia Kenya Ivory cost
$US values of output share in % 27 6.6 6.1 4.9 4.7
Farmers per 1,000 ha 472 456 2,204 2,365 458
Tractors per 1,000 ha 0.9 0.7 0.3 2.5 0.6
Livestock per 1,000 ha 321 1,253 1,634 1,240 137
Fertilizer kg/ha 5.8 4.5 13.2 28.3 13
Output per farmer in $US 1,405 676 186 321 1,193

17.3.2 Rural and Poverty Centered Economic Performance

There has been a positive trend in the Ethiopian economy over the last few years.
The GDP declined by an average of 0.3 % per annum in 2001–2002 and 2002–2003,
mainly because of a drop in agricultural output of 6.2 %. In the following year agri-
cultural output recovered and the economy showed significant growth. According to
official figures annual growth in GDP has been in double digits since 2003–2004.
This sustained rate of GDP growth above the rate of population growth is expected
to have improved the well-being of the population. Notable growth has been exhib-
ited by all three major sectors of the economy: agriculture, industry, and service. In
the last few years agriculture has become the driver of consistent economic growth
(Table 17.4).
The growth in value-added agriculture has come as a result of both land expan-
sion and increased productivity. Both the amount of land under cultivation and the
volume of crop production increased consistently between 2003–2004 and 2009–
2010. Except for 2007–2008, the rate of production growth was higher than the rate
of expansion of cultivated land for seven consecutive years, indicating growth in the
yield per hectare of land over the same period (Table 17.5).
Growth in the performance of agriculture in excess of the rate of population
growth may be interpreted as a reduction in rural poverty for three reasons. First,
this growth is primarily a result of improvements in the productivity of small-scale
subsistence farmers, because large-scale commercial farming is not well-established
in Ethiopia. Second, although this growth does not necessarily translate itself into
proportional improvement in the well-being of all households, given that Gini coef-
ficient is relatively lower in rural Ethiopia, (MoFED 2008), there is a high probabil-
ity of corresponding improvement in the well-being of a large segment of the rural
population. Third, according to the Lorenz curve (ECOSOC 2007), improved input
supply among rural households (as a proxy for extension assistance) has been more
or less balanced and thus it is likely that this growth corresponds to contributions
from the majority of households.
The growth of the agricultural sector contributes to the growth of other sectors
through trickledown effects that are also likely to positively influence the lives of
rural people. In an economy-wide model used to analyze growth and poverty reduc-
tion linkages in Ethiopia, Diao et al. (2007) showed that broad based growth in
agriculture is likely to reduce poverty in the country. Growth in agriculture induces
 286

Table 17.4 Recent Ethiopian economic indicators

Average Average
2001/2002– 2003/2004–
Sector/year 2002/2003 2003/2004 2004/2005 2005/2006 2006/2007 2007/2008 2008/2009 2009/2010 2009/2010
Growth rates (percentages)
GDP −0.3 11.73 12.64 11.54 11.79 11.19 10.03 10.44 11.34
Agriculture −6.2 16.97 13.54 10.92 9.45 7.49 6.35 7.63 10.34
Crop −10.1 25.56 19.51 14.97 11.02 7.99 6.47 8.74 13.47
Livestock −1.7 8.02 5.94 4.90 7.88 7.32 7.05 6.19 6.75
Share from GDP in %
Agriculture 47 47.0 47.4 47.1 46.1 44.6 43.1 42.0 44.9
Crop 29.6 27.6 29.3 30.2 30.0 29.2 28.2 27.8 28.9
Livestock 15.1 14.3 13.5 12.7 12.2 11.8 11.5 11.0 12.2
Growth contribution to GDP
Agriculture – 68.0 50.8 44.6 37.0 29.9 27.3 30.7 41.0
Based on World Bank (2010) and MoFED (2010)
A. Admassie and D. Abebaw
17 Rural Poverty and Marginalization in Ethiopia… 287

Table 17.5 Area under cultivation, production, and productivity of food grain in Ethiopia
Cultivated land Production Growth rate (percentages)
(millions of (millions of Yield (quintals/
Year hectares) quintals) hectare) Land Production Yield
2003–2004 8.7 103.6 11.9
2004–2005 9.8 119.1 12.2 12.6 15 2.5
2005–2006 10.2 133.8 13.1 4.1 12.3 7.4
2006–2007 10.5 155 14.8 2.9 15.8 13
2007–2008 11 161.2 14.7 4.8 4 −0.7
2008–2009 11.2 171.1 15.3 1.8 6.1 4.1
2009–2010 11.5 180.8 15.7 2.7 5.7 2.6
Based on CSA (2006, 2007, and 2008)

higher overall growth than other sectors. Poverty growth elasticity is greater when
additional income is driven by agriculture, which is the primary source of rural
livelihoods relative to non-agricultural sectors of the economy.9
The cottage and handicraft industries sub-sector is a major source of income in
rural areas after crop and livestock production, and hunting. Many rural people in
both crop producing areas and pastoralist communities engage in this sub-sector
either full time or as a secondary income generating activity. The value added by the
small-scale cottage and handicraft sub-sector grew by about 7 % between 2003–
2004 and 2009–2010. Given this overall growth, assuming that the increase in cot-
tage and handicraft activities exceeds the rate of population growth, then it is likely
that this sub-sector has also reduced poverty among the rural population that derives
income from it. In addition to increased agricultural and nonagricultural activities,
capital expenditures on poverty related sectors could also reduce poverty through
the creation of job opportunities or other support. Rural antipoverty expenditures on
agriculture, rehabilitation, construction, education, and health care have also
expanded over this period (Fig. 17.1).
Expenditures on the construction of micro-irrigation schemes and micro-dams,
the development of springs, and similar efforts are expected to reduce acute water
shortages during droughts. Expansion of road networks and telecommunication ser-
vices ease market access, reduce transaction costs, and diversify both farm and off-
farm economic activities. Easier access to markets, the expansion of microfinance,
productive safety net programs, and the formation of service cooperatives reduce
the need for small-scale farmers to sell product when prices are low (MoFED 2008).
Improvements in market access not only help reduce price margins and transaction
costs between consumers and producers in a given area, but also reduce price varia-
tions at greater market scales. The implications of the various interventions by dif-
ferent stakeholders, including the government and donors, and the participation of
the people in the improvement of their own lives, can be broadly assessed from the
welfare indicators presented in Table 17.6.

9
According to Diao et al. (2007) a 1 % annual increase in per capita GDP driven by agriculture led
growth leads to a 1.9 % reduction in the poverty head count rate.
288 A. Admassie and D. Abebaw

100.00

80.00

60.00

40.00

20.00

0.00
2004 2005 2006 2007 2008 2009
Agriculture and Rehabilitation Construction Education & Training Public Health Others

Fig. 17.1 Capital expenditures on “pro-poor” economic sectors in Ethiopia (Based on MoFED
2010)

Table 17.6 Living condition indicators in Ethiopia over time

Indicators 2004–2005 2009–2010 Indicators 2004–2005 2009–2010
Macroeconomy Education
Per capita income 472 Gross primary 79.8 94.2
at current prices $US education
(1–8) coverage
(percent)
Consumption/GDP 96.3 90.6 Primary school 0.81 0.93
ratio (percent) girls to boys
ratio
Investment/GDP ratio 21.6 23.7
(percent)
Saving/GDP ratio (percent) 3.7 9.4 Health
Foreign aid/GDP ratio 4.0 3.9 Primary health 70 89
(percent) service
coverage
Poverty related government 14.2 12.5 Under-five 123 101
expenditure to GDP mortality rate
ratio (percent) (per 1,000)
People below absolute 39.0 29.2 Rural potable 35 65.8
poverty line (percent) water coverage
(percent within
1.5 km)
People below food poverty 28.2
line (percent)
Households covered 4,800,000 7,800,000
in the PSNP
Based on MoFED (2006, 2010)

Per capita income in terms of current prices has reached US$472 in 2010 with
incremental growth for the last 7 years. There has been a corresponding decrease in
the percentage of people living below absolute poverty from 39 to 29 %. Savings as
a share of GDP have also risen by more than double, indicating country-wide
improvements in living standards. As long as agriculture continues to grow at a
meaningful rate, rural poverty will be reduced by some margin. Gross primary
17 Rural Poverty and Marginalization in Ethiopia… 289

Table 17.7 Education coverage in Ethiopia, 2004–2005 to 2008–2009

2004–2005 2008–2009
Access indicators Total Male Female Total Male Female
Grade 1: net intake rate 60.9 62.2 59.6 82.0 84.1 79.9
Net enrollment rate–primary 1–4 67.6 69.9 65.1 88.7 90.3 87.0
Net enrollment rate–primary 1–8 68.5 73.2 63.6 83.0 84.6 81.3
Gross enrollment secondary 9–10 27.3 36.6 21.6 38.1 43.7 32.4
Percentage of higher education enrollment 76.0 24.0 71.5 28.5
Primary 1–8 students/section 69 59
Primary 1–4 (girls/boys) 0.87 0.90
Primary 5–8 (girls/boys) 0.69 0.92
Secondary 9–10 (girls/boys) 0.57 0.74
Based on MoE (2010)

education and health coverage have also improved significantly. Child mortality has
been reduced by 22 per 1,000 within 5 years. Rural water coverage has almost
doubled within the last 5 years. The World Bank (2010) indicates that around 62 and
88 % of the Ethiopian people do not have access to improved water and sanitation
services respectively. These changes have already been reflected in the HDI of the
UNDP (Table 17.7).
The rates of enrollment for all education levels indicate expansion in education
coverage. Within a period of 4 years, this achievement has been significant. This
situation reveals the possibility of achieving universal primary education (UPE) by
2015. Despite the changes in overall primary enrollment, the rates for secondary
education are quite low.
Within the health sector, the number of health facilities (health centers, health
posts, and hospitals) has grown more than the rate of population growth as shown in
Table 17.8. As a result, the ratio of people per health facility has declined by 8 % for
hospitals and 16.5 % for health centers. Consequently, health care coverage rose at
an annual rate of around 13 % over the period from 2005–2006 to 2008–2009.
Consistent with the health policy emphasis on preventive health measures, the num-
ber of health extension workers has increased substantially over the same period.
With respect to core health professionals, although the number of health officers has
increased significantly the number of physicians and nurses has declined, leading to
an increase in the number of people per health care professional. Despite the expan-
sion of physical facilities, the shortage of health care professionals likely has nega-
tive impacts on the provision of health care services.
In general the indicators presented in Table 17.8 reflect improvements in public
well-being. However, there is some debate on the effects of antipoverty interven-
tions and expenditures on poverty. Tewodaj et al. (2008) argue that not all types of
antipoverty expenditures improve income levels and income-based measures of
well-being. Public expenditures on agriculture do not affect productivity as substan-
tially as would be expected. Significant effects of expenditures on agriculture have
only been observed among rural households residing near urban centers. Education
spending has widespread and similar effects on welfare across different regions,
whereas health spending does not seem to have similar effects on welfare.
290 A. Admassie and D. Abebaw

Table 17.8 National level health care coverage in Ethiopia, 2005–2009

Annual
Health indicators 2005–2006 2006–2007 2007–2008 2008–2009 growth ratea
Total population 72,045,327 73,918,505 75,840,386 77,812,236 2.7
Primary health service 76.9 86.7 89.6 90 5.7
coverage
Number of hospitals 138 143 149 195 13.8
Number of health centers 635 690 732 1,362 38.2
Number of health stations 1,206 1,376 1,517 No data 12.9
and health centers
Number of health officers 715 1,151 1,242 1,606 41.6
Number of nurses 17,845 18,146 16,765 No data −3.09
Physicians 2,155 1,806 2,085 2,151 −0.1
Health extension workers 9,900 17,653 24,571 30,950 70.9
Population per hospital 522,068 516,913 508,996 399,037 −7.9
Population per health center 1134,57 107,128 103,607 57,131 −16.5
Population per health station 59,739 53,720 49,994 No data −8.2
Population per physician 33,432 40,929 36,374 36,175 2.7
Population per nurse 4,037 4,074 4,524 No data 6
MoH (2010)
a
In percentages

According to the World Bank (2010) around 40 % of Ethiopians still live on less
than US$1.25/day (in PPP terms), implying that the intensity of poverty is still sub-
stantial. Recurrent droughts and acute food shortages continue to be distinguishing
features of the country. Widespread reliance on traditional subsistence farming
practices has led to the degradation of agricultural land. Ethiopia ranked second in
the world after Burkina Faso in the percentage (72 %) of its people that live on
degraded land (World Bank 2010). That study also found that overall life self-
satisfaction among Ethiopians was rated 4.2 out of ten, indicating that approxi-
mately 58 % of the people are not satisfied with their way of life. Only 33 % of
Ethiopians felt that their standard of living is normal and the majority of the popula-
tion had a negative opinion of their life experience. That study also found that 20, 32,
and 39 % of the population were satisfied with the quality of health care, education,
and water respectively (World Bank 2010).

17.4 Growth Divergence Across Geographical Areas

17.4.1 Growth Divergence in Income

Crop Producing Farmers. The main livelihood source in the rural highland and
semi-highland populations of the country is crop production. Over the last 7 years
crop production has exhibited a high rate of growth, exceeding that of the
17 Rural Poverty and Marginalization in Ethiopia… 291

livestock sector. According to official statistics the pace of crop production growth
even exceeded the rate of GDP growth for three consecutive years until 2005–2006.
This production growth is likely to have made some improvement in the welfare of
the majority of residents in rural crop producing areas. Export crops include coffee,
oilseeds, and pulse beans. Most export crops are produced by relatively well-off
cash crop producers, so the growth in export crop production may not have directly
improved rural poverty conditions. It is possible that conditions improved indirectly
due to the boost to the overall economy from greater export crop production.
Food Insecurity in Highland and Semi-Highland Areas. Although the main source
of livelihoods of people in these areas is farming, food shortages occur frequently
because of the irregularity and inadequacy of rainfall. The PSNP focuses largely on
these areas and has somewhat improved livelihoods in food insecure areas. There
have been an estimated 7.8 million chronically food insecure beneficiaries of the
government managed PSNP. This does not mean, however, that the contribution of
the PSNP is sufficient to resolve the problem.
Only 18,538 households have graduated from the program in the three most pop-
ulous regions states: Amhara, Oromiya, and SNNPR (Amdissa 2010). The program
failed to reach approximately 70 % of the food insecure households in those regions
(Sharp et al. 2006), leaving a high number of chronically food insecure households.
Consecutive seasons of failed rainfall, a rapidly growing population, endemic poverty,
and limited government capacity are still major causes of chronic food insecurity
and water shortage in large parts of the country (USAID 2010). The federal govern-
ment estimates that 4.8 million people were in need of emergency food assistance
from January to June 2010, indicating that poverty and food insecurity are still very
serious problems in the country.
Pastoral Areas. The number of pastoralists in Ethiopia was estimated at between 12
and 15 million people in 2004. The pastoralist populations consist of different ethnic
groups and occupy 61 % of the nation’s land area. Approximately 92 % live in
Somali, Afar, and Oromiya (PFE 2004), the remaining 8 % are in parts of Oromiya,
the South Omo area of SNNPR, Gambella, and other regions (Assefa 2008).
Pastoralist livelihoods rely on livestock production and hunting. Economic activ-
ities in pastoral communities contribute an estimated 12 % of the GDP (Assefa
2008). Pastoral areas account for approximately 42 % of the national livestock pop-
ulation (PFE 2004). Pastoral areas of Ethiopia typically have harsh climatic condi-
tions featuring erratic and low average annual rainfall, high temperatures, and
ecologically fragile environments with degraded soils (PFE 2004).
Livestock production is not only important to pastoralists, it is vital to the econ-
omy of many crop farmers as well. As with crop production, the performance of this
sub-sector has been greatly hampered by recurrent drought and other supply-side
problems. This sub-sector has grown faster over the last 7 years than the rate of
population growth, therefore poverty in pastoralist areas is expected to have fallen
as well. The growth rate of value-added complements to the livestock sub-sector has
been relatively lower than for crop production. It has been argued that growth in the
livestock sector could also have a significant effect on economic growth, but with a
292 A. Admassie and D. Abebaw

relatively lesser effect on poverty alleviation because of the fact that this sub-sector
engages fewer people than crop production.
Improving conditions for marginalized segments of the population has para-
mount social and political implications beyond mere economic importance. The
PCDP is one of the mechanisms for applying interventions to address development
inequality. A total of US$60 million was allocated in phase one of the PCDP for
1,804 community initiated micro-projects in 32 (25 % of the total) pastoral woredas
(Assefa 2008). These interventions included contributions to income generating
activities, water projects for humans and livestock, and the construction of schools,
health posts, veterinary clinics, small-scale irrigation projects, and roads. The total
number of beneficiaries of these interventions was estimated be one million pasto-
ralists. These interventions indicate that there is social capital with huge develop-
ment potential in terms of articulating problems, basic needs, and related solutions,
and the commitment for the development of micro-projects (Assefa 2008).
Despite the various initiatives of the government, the international community,
and NGOs, there are still very large imbalances, particularly between the highland
and lowland parts of the country. Growth strategies and plans, including the Food
Security Strategy, did not seriously consider the plight of pastoralists or draw les-
sons from past experiences and realities in pastoral areas (PFE 2004). Some argue
that pastoralists are the most economically and politically marginalized social group
in Ethiopia. In some cases pastoralist land was appropriated unlawfully and without
the consent of pastoralist communities for purposes such as commercial farms and
wildlife conservation areas. The absence of by-laws to apply the provisions of the
constitution that protect communities from dispossession of their land is one of the
manifestations of marginalization.
Despite some improvements over the years, health, education, and other services
are still less available in pastoralist areas than the rest of the country. The effects of
natural resource loss, recurrent drought, and the inadequacy of economic and social
services, have exceeded the capacity of traditional pastoralist coping mechanisms
and aggravate the problems associated with poverty (PFE 2004). Historic move-
ments from the highlands to the lowlands, including both voluntary and involuntary
government resettlement efforts to alleviate land scarcity, have increased social con-
flict due to competition for grazing land and water. As a result, life in the lowlands
became harsher and more prone to conflict than before (Patterson 2007).
Aside from marginalization in terms of inadequate government attention, the
interventions are often perceived as top-down approaches. Local and government
perceptions and evaluations of risk differ significantly in the case of pastoralists in
Afar (Rettberg 2010). The government considers the pastoralist way of life responsible
for the pastoralist livelihood crisis in the state, whereas pastoralists blame the igno-
rance and policies of the government. For the pastoralists in Afar, drought, health risks,
and shortages of food and water are risk factors. For instance, efforts by the government
to settle pastoralists along river banks are perceived by some as a policy intended to
end their traditional pastoralist lifestyle. If there is a lack of understanding of the
root causes of impoverishment in pastoral areas, it may lead to policies that reinforce
marginality by keeping communities in a poverty trap despite costly interventions.
17 Rural Poverty and Marginalization in Ethiopia… 293

Table 17.9 Primary education enrollment in Ethiopia by regional state (MoE 2010)
2008–2009 primary
Share of female enrollments
Primary enrollment to male (1–8) (%)
Regional Growth
state 2004–2005 2008–2009 rate (%) 2004–2005 2008–2009 Male Female Total
Tigray 773,026 1,000,626 6.7 0.49 0.5 95.6 98.1 96.9
Afar 52,671 100,429 17.5 0.36 0.39 25.3 23.2 24.4
Amhara 2,798,860 4,156,399 10.4 0.48 0.49 101.4 103.1 102.2
Oromiya 4,561,378 5,570,628 5.1 0.42 0.46 80.9 74.8 77.9
Somali 192,914 358,224 16.7 0.36 0.38 33.3 29.4 31.6
Benishangul– 131,672 171,170 6.8 0.4 0.43 97 80.1 88.6
Gumuz
SNNPR 2,304,980 3,505,713 11.1 0.42 0.46 94.3 84.5 89.4
Gambella 53,865 77,075 9.4 0.39 0.43 80.2 69.7 75.2
Harari 26,448 39,932 10.8 0.43 0.45 100.2 83.6 91.9
Total 11,448,641 15,553,142 8 0.44 0.47 84.6 81.3 83

17.4.2 Education and Health Coverage

Primary education enrollment increased from 2004–2005 to 2008–2009 in all

regional states (Table 17.9). The rate of increase has not been uniform. Although
this data includes both urban and rural areas, the rural population is relatively much
larger, therefore these trends should reflect the situation in rural areas. In the rela-
tively lagging regions of Afar, Gambella, Benishangul-Gumuz, and Somali, it is
possible that enrollment has grown to compensate for a historic backlog.
The data presented in Table 17.9 clearly show that the enrollment ratio of girls to
boys consistently favors boys with the exception of Tigray and Amhara. In predomi-
nantly pastoralist regional states like Afar and Somali, the enrollment rate is much
lower than the national average and in other regions. The other two regional states
with significant pastoralist populations (SNNPR and Oromiya) also have relatively
lower enrollment rates than predominantly crop producing regional states like
Tigray and Amhara. Despite the fact that Gambella and Benishangul-Gumuz are
relatively undeveloped regions of the country their enrollment performance is rela-
tively good (Table 17.10).
Great regional health care service coverage disparity is evident in Ethiopia.
The immunization coverage was over 70 % in Tigray, Harari, and Addis Ababa,
whereas the coverage in Somali, Gambella, and Dire Dawa was less than 40 %. In
family planning coverage Dire Dawa, Amhara, and Tigray rank the highest, and
Afar, Somali, Gambella, and Benishangul Gumuz were the lowest because popu-
lation pressure has not been a serious problem in these areas. Gambella and
Benishangul-Gumuz have relatively high numbers of health professionals, Afar
and Somali have the least health professionals. There is a relatively high con-
centration of nurses, health officers, and physicians in Tigray, Addis Ababa, Dire
294 A. Admassie and D. Abebaw

Table 17.10 Selected health service indicators in Ethiopia by regional state (MoH 2010)
Immun- Family Health
ization planning Access Number Total health extension
Regional coverage coverage to potable Number of health workers workers
state (%) (%) water (%) of nurses officers (less HEW) (HEW)
Tigray 77.8 67.8 80 2,332 289 5,339 1,259
Afar 44.1 12.8 58 185 44 723 375
Amhara 67.4 69.3 59.3 3,790 738 13,730 6,415
Oromiya 64.9 50.6 57.6 5,040 826 21,125 12,875
Somali 30.3 6.6 33.5 314 83 2,188 1,100
Benishangul- 53.3 25.6 51.5 452 54 1,170 499
Gumuz
SNNPR 69.9 76 74.2 3,980 462 13,859 7,492
Gambella 35.9 20.3 44.6 91 26 618 457
Harari 74.4 35.6 56 276 51 555 32
Addis Ababa 77.6 32.8 3,377 1,104 6,386 0
Dire Dawa 39.8 83.8 75.8 272 72 621 74
National 65.5 56.2 61.5 20,109 3,758 66,314 30,578

Dawa, and Harari. Approximately 80 % of the residents of Tigray have access to

potable water. Afar, Somali, and Gambella have the lowest percentage of residents
with potable water access.

17.4.3 Vulnerable Social Groups

In policy terms, various measures have been taken to address cultural discrimination
against women that limits access to resources and opportunities at both the household-
level and beyond. The most important achievement in terms of reducing the vulner-
ability of women to marginalization in rural Ethiopia is the right to own land. Land
registration procedures in many regions of the country require the name of the buyer’s
spouse to also be registered (Deininger et al. 2008). Widows are also given land titles.
Having a registered land title increases confidence and empowers women to either
invest in activities with long-term dividends (perennial crops or environmental protec-
tion activities) or rent land without fear of it being appropriated by tenant farmers.
Women’s access to education, health care, as well as donor and government co-
supported extension and relief activities, has improved. For instance, between
2004–2005 and 2009–2010, the ratio of girls to boys in primary school across the
country improved from 0.81:1 to 0.93:1. The workloads of women at the household-
level has also been reduced because of improvements in health services (which have
reduced frequency of communicable diseases affecting children and overall house-
holds), and access to potable water (reduced the amount of time spent fetching
water). The expansion of MFIs and gender oriented programs by the government,
and both political party affiliated and independent NGOs, have provided many
opportunities for creating income generating activities.
17 Rural Poverty and Marginalization in Ethiopia… 295

Table 17.11 Alternative rural basic education enrollment in Ethiopia by regional state: 2008–2009
(MoE 2010)
Female/ Regional
Male Female Total total (%) share (%)
Tigray 2,866 2,592 5,458 0.47 0.75
Afar 11,195 6,340 17,535 0.36 2.4
Amhara 207,749 177,882 385,631 0.46 52.8
Oromiya 103,100 81,397 184,497 0.44 25.3
Somali 0 0 0 0
Benishangul–Gumuz 13,241 10,044 23,285 0.43 3.2
SNNPR 57,402 48,993 106,395 0.46 14.6
Gambella 4,147 2,587 6,734 0.38 0.9
Total 400,056 330,223 730,279 0.45 100

There are criticisms of the effectiveness of gender oriented interventions. According

to Demessie and Yitbark (2008), the lives of women have not changed significantly in
the last few decades. Women have not benefitted much from extension services (only
9 %) because of the inability to make down payments of up to 25 % of the total input
purchase. Credit services are mainly for agricultural resources linked with land and
other resource endowments, and thus offer little to the marginalized poor, specifically
women (Hareg Consult PLC 2005). One of the basic constitutional rights intended to
support women is the right to own land. In the context of rural Ethiopia a land parcel
is the most vital property a person or household may own, but women only own
around 11 % of all agricultural land in the country (CSA 2006).
Despite several legal restrictions and measures enacted over the last decade,
women still face widespread violations of their constitutional rights. Female students
frequently confront problems such as early marriage, abduction and rape, unwanted
pregnancy, heavy work-loads at home, and stigmas associated with gender parity
(ADA 2011). In the area of education, enrollment growth has exceeded the rate of
population growth. Both genders have been beneficiaries of this growth despite
enrollment gender disparity in many of the regions. In addition to regular education,
alternative basic education opportunities provide increased access to education for
students above primary age (Table 17.11).
The total number of beneficiaries of the Alternative Basic Education program is
estimated to be around 0.7 million people, and 53 and 25.3 % of the beneficiaries
are from the regional states of Amhara and Oromiya respectively. The shares of
beneficiaries from other regions are extremely low. Somali in particular, has not
been benefiting from this program. In all regional states, the relative proportion of
female beneficiaries has been low.
Many children are deprived of their right to education because either their fami-
lies cannot afford education related costs or the opportunity cost of schooling is
too high (ADA 2011). Koohi-Kamali (2008) found evidence of gender bias against
girls in adult consumption patterns of Ethiopian households after controlling for
the effects of number and age of children. This evidence supports the case for
296 A. Admassie and D. Abebaw

empowering women, including enhancing their income to reduce intra-household

inequality, partly through increased expenditure on goods that benefit the welfare of
children.
Poverty in rural areas has the greatest effects on children. Poor farmers with
small plots or degraded land, particularly from the highland parts of the country,
have few options to putting their young children to work for meager income. Some
are forced to migrate to urban areas to seek work so that they can help maintain their
household. High population pressure, low agricultural productivity, drought, inability
to afford the costs of agricultural inputs, and early marriage are some of the moti-
vators of rural-to-urban migration (Girmachew 2009). Rural-to- urban migration
arises from the intricate socio-economic problems in rural areas and often ultimately
results in the marginalization of the migrants in urban areas.
Although not much is known about the extent of people living with physical dis-
abilities in Ethiopia, its magnitude cannot be underestimated. UNICEF (2006) esti-
mated that the disabled numbered between 5 and 7.7 million. Because of inherent
discrimination and physical constraints, poverty among the disabled is expected be
greater than the rest of the population. Studies of other developing counties indicate
that poverty among the disabled is harsher than for other poor (UNICEF 2006).
In light of the scarcity of information available, it can be assumed that the situation
is not any different in Ethiopia.
The extent of poverty and its underlying causes are expected to be more appar-
ent in the case of the disabled for several reasons. First, a disability often does not
allow one to work productively in typical activities requiring physical labor, such
as the traditional agricultural livelihoods in Ethiopia. Second, access to social ser-
vices such as education is generally limited for the disabled in Ethiopia and more
so in rural communities due to the lack of appropriate facilities, the distance
between schools and villages, and limited transportation options. Some families do
not send their disabled children to school because of the negative attitudes of other
parents and the community (Seleshi 2010). Recently, the attitudes of parents have
begun to change and the number of parents taking their disabled children to school
has increased.

17.5 Conclusions

The general objective of this research effort was to assess the policies and programs
of the federal government of Ethiopia, and also the interventions of other stake-
holders that address rural poverty and particularly ultra-poverty and their overall
effects. Civil conflict and misguided economic policies during the 1970s and
1980s, widespread reliance on rainfall dependent subsistence farming and recurrent
drought have condemned a significant percentage of the Ethiopian people to extreme
poverty. Since 1991 relative peace and stability have prevailed in the country. Within
the general policy framework of the ADLI development strategy, various economic
17 Rural Poverty and Marginalization in Ethiopia… 297

policy measures have been taken to mitigate poverty and the inequitable delivery of
public services.
More specifically the federal government launched three consecutive medium-
term development programs that feature rural development and poverty reduction
as core objectives. The various rural development programs, such as the current
Growth and Transformation Plan, the frequently expanded coverage of the Food
Security Program, the education and health sector programs, and special policies
and programs designed to address the problems of vulnerable areas and social
groups, are some of the measures intended to alleviate rural poverty and inequality.
In addition to the expanded government expenditures on pro-poor sectors of the
economy, donor organizations such as local and international NGOs, civil society
organizations, and MFIs have been offering technical and financial support for
these endeavors.
Between 1991–1992 and 2002–2003 the economy improved considerably, with
sustained and reasonably high growth. In particular, the agriculture sector has shown
a notable rate of growth, unlike during the 1970s and 1980s. As a result, per capita
income has generally increased, particularly in rural areas. As documented in
reports by the federal government and international organizations such as the UNDP,
there have also been relatively meaningful changes in non-income dimensions of
well-being, such as access to education and health care services, and other human
development parameters of the rural population.
Although the effects of increased income and the expansion of social services
are expected to traverse the rural population in different regions, poverty is still
severe and pervasive in rural areas (MoFED 2010). Approximately 30 % of Ethiopians
live below the absolute poverty line. Through the PSNP component of the GTP, the
government intends to benefit about 7.8 million people, implying that around 10 %
of the country’s population requires government support to cope up with their
current economic conditions (MoFED 2010). Despite consistent double-digit eco-
nomic growth for over 5 years, the country still needs committed efforts to fully
feed its people. Although compensating for the shortfalls accumulated over the
years, some regions remain less served than others, regardless of efforts to expand
social services throughout the country. Less developed and pastoralist regional
states such as Afar, Somali, Benishangul-Gumuz, and Gambella still need greater
improvement in the delivery of health and education services.
There have been considerable achievements by some of the efforts to improve
access to economic opportunities and social services regardless of gender.
Among other measures, the land policy, the family code, and the affirmative
action policies in education have improved the socio-economic position of
women. However, growth has not corrected gender inequality and the disparities
are evident across the country. Given this overall background, there continue to
be development issues to address such as regional development disparity and the
persistence of severe poverty despite a relatively high rate of economic growth
and significant efforts to increase the equitable distribution of economic oppor-
tunities and services.
298 A. Admassie and D. Abebaw

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

ADA (2011) Annual performance report. Amhara Development Association, Amharic.

http://www.amhara.org. Accessed 14 Feb 2011
Amdissa T (2006) Agriculture, growth and poverty reduction in Ethiopia: policy processes around
the new PRSP (PASDEP). Paper prepared at the Future Agriculture Consortium workshop,
Institute of Development Studies, Hawassa, 20–22 March 2006
Amdissa T (2008) A review of education policy, strategies and programs. In: Assefa T (ed) Digest
of Ethiopia’s national policies, strategies and programs. Forum for Social Studies, Addis Ababa
Amdissa T (2010) The road for long-term social protection: the case of Amhara region productive
safety net programme. In: Alemu G, Getachew D (eds) Proceedings of the first regional confer-
ence of the Amhara regional state economic development. Ethiopian Economic Association,
Addis Ababa, pp 1–26
Assefa T (2008) Overview of achievements and challenges in implementing CDD projects.
Pastoral communities: the case of pastoral community development project. African
Development Bank, Tunis
Bijlmakers L (2003) Process and impact of the PRSP in Ethiopia. Wemos Foundation, Amsterdam
CSA (2006) Agricultural sample survey (2005–2006), vol 1, Statistical bulletin 361. Central
Statistics Agency, Addis Ababa
CSA (2007) Agricultural sample survey (2007–2008), vol iv, Statistical bulletin 388. Central
Statistics Agency, Addis Ababa
CSA (2008) Agricultural sample survey (2008–2009), vol iv, Statistical bulletin 417. Central
Statistics Agency, Addis Ababa
Deininger K, Ali DA, Tekie A (2008) Impacts of land certification on tenure security, investment,
and land markets: evidence from Ethiopia, Policy research working paper 4764. World Bank,
Washington, DC
Demessie S, Yitbark T (2008) A review of national policy on Ethiopian women. In: Assefa T (ed)
Digest of Ethiopia’s national policies, strategies and programs. Forum for Social Studies,
Addis Ababa, pp 93–126
Desalegn R (2008) Ethiopia: agricultural policy review. In: Assefa T (ed) Digest of Ethiopia’s
national policies, strategies and programs. Forum for Social Studies, Addis Ababa,
pp 129–151
Devereux S, Sabates-Wheeler R (2006) Ethiopia’s productive safety net programme (PSNP):
trends in PSNP transfers within targeted households. Institute of Development Studies,
Addis Ababa
Diao X, Fekadu B, Haggblade S, Seyoum TA, Wamisho K, Yu B (2007) Agricultural growth link-
ages in Ethiopia: estimates using fixed and flexible price models, Discussion paper no 00695.
International Food Policy Research Institute, Washington, DC
ECOSOC (2007) Emerging Ethiopia: strengthening efforts to eradicate poverty and hunger,
including through the global partnership for development. Annual ministerial level substantive
review of the Economic and Social Council, voluntary national report 2007. Economic and
Social Council, Addis Ababa
Ellis F, Tassew W (2005) Ethiopia participatory poverty assessment. Ministry of Finance and
Economic Development, Addis Ababa
FDRE (1995) The constitution of the Federal Democratic Republic of Ethiopia. In: Federal Negarit
Gazetta. 1st year no 1, 21st August 1995, Proclamation no 1, Federal Democratic Republic of
Ethiopia, Addis Abeba
17 Rural Poverty and Marginalization in Ethiopia… 299

FDRE (2002) Food security strategy. Federal Democratic Republic of Ethiopia, Addis Ababa
FDRE (2007) The New federal budget grant distribution formula. Federal Democratic Republic of
Ethiopia, Addis Ababa
Getachew M (2008) A review of national policy of Ethiopia. In: Assefa T (ed) Digest of Ethiopia’s
national policies, strategies and programs. Forum for Social Studies, Addis Ababa, pp 23–45
Gilligan DO, Hoddinott J, Seyoum AT (2008) The impact of Ethiopia’s productive safety net
programme and its linkages, Discussion paper no 00839. International Food Policy Research
Institute, Washington, DC
Girmachew A (2009) Church as a refuge for marginalized children in urban Ethiopia. In: Ege
S, Aspen H, Berhanu T, Shiferaw B (eds) Proceedings of the 16th international conference
of Ethiopian studies. Norwegian University of Science and Technology, Trondheim,
pp 1025–1032
Handicap International and CBM Christoffel-Blindenmission (2006) Making PRSP inclusive.
http://www.making-prsp-inclusive.org/uploads/media/Making_PRSP_Inclusive_311008_E.
pdf. Accessed Nov 2009
Hareg Consult PLC (2005) Gender situation analysis in Ethiopia, Draft paper. Hareg Consult PLC,
Addis Ababa
IMF (2006) The federal democratic republic of Ethiopia: poverty reduction strategy paper –
2003/04 annual progress report, IMF Country report 06/27. International Monetary Fund,
Washington, DC
Koohi-Kamali F (2008) Intra-household inequality and child gender bias in Ethiopia, Policy
research working paper no 4755. World Bank, Washington, DC
Lipton M (2005) Can small farms survive, prosper, or be the key channel to cut mass poverty?
Paper presented at the Food and Agriculture Organization’s symposium on agricultural com-
mercialization and the small farmer, Rome, 4–5 May 2005
MOA/FAO (1984) Ethiopian highland reclamation studies. The degradation of resources and an
evaluation of action to combat it, Working paper 19. Ministry of Agriculture/Food and
Agriculture Organization of the United Nations, Addis Ababa
MoARD (2009) Food security program 2010–2015. Ministry of Agricultural and Rural
Development, Addis Ababa
MoE (2010) National report on the development and state of the art of adult learning and educa-
tion. Ministry of Education, Addis Ababa
MoFED (1993) An economic development strategy for Ethiopia: a comprehensive guidance and
a development strategy for the future. Ministry of Finances and Economic Development,
Addis Ababa
MoFED (2002) Strategy paper for promoting development and poverty reduction. Ministry of
Finance and Economic Development, Addis Ababa
MoFED (2006) Ethiopia: building on progress a plan for accelerated and sustained development to
end poverty (PASDEP). Ministry of Finance and Economic Development, Addis Ababa
MoFED (2008) Dynamics of growth and poverty in Ethiopia (1995/96–2004/05). Ministry of
Finance and Economic Development, Addis Ababa
MoFED (2010) Five years (2010/11–2014/15) growth and transformation plan, Amharic version.
Ministry of Finance and Economic Development, Addis Ababa
MoFED (2011) Growth and transformation plan (GTP): 2010/11–2014/15. Ministry of Finance
and Economic Development, Addis Ababa. http://et.mofcom.gov.cn/accessory/201101/
1296371918402.pdf. Accessed 10 Jan 2011
MoH (2010) Health and health related indicators. Ministry of Health, Addis Ababa
MoH (2011) Health sector development program. Ministry of Health, Addis Ababa
NCFSE (2003) Food security programme, vols 1 and II. New Coalition for Food Security in
Ethiopia, Addis Ababa
Omamo SW, Diao X, Wood S, Chamberlin J, You L, Benin L, Wood-Sichra U, Taywangire A
(2006) Strategic priorities for agricultural development in Eastern and Central Africa,
IFPRI-ASARECA research report 150. International Food Policy Research Institute,
Washington, DC
300 A. Admassie and D. Abebaw

Patterson KP (2007) Integrating population, health, and environment in Ethiopia. Population

Reference Bureau, Washington, DC
PFE (2004) Pastoralist Forum Ethiopia strategic plan: 2004–2008. Pastoralist Forum Ethiopia,
Addis Ababa
Pratt AN, Yu B (2008) An updated look at the recovery of agricultural productivity in Sub-Saharan
Africa, IFPRI discussion paper no 00787. International Food Policy Research Institute,
Washington, DC
Rettberg S (2010) Contested narratives of pastoral vulnerability and risk in Ethiopia’s Afar region.
Practical Action Publishing, Rugby
Seleshi Z (2010) Disability and development: the need to develop disability–inclusive PASDEP in
Ethiopia. Poverty Action Network, Addis Ababa
Sharp K, Brown T, Teshome A (2006) Targeting Ethiopia’s productive safety net program (PSNP).
Overseas Development Institute/The IDL Group, London
Tewodaj M, Gezahegn A, Zelekawork P (2008) The bang for the birr: public expenditures and rural
welfare in Ethiopia, Research report 160. International Food Policy Research Institute,
Washington, DC
UNDP (2010) Human development report 2010: the real wealth of nations: pathways to human
development. United Nations Development Programme, New York
UNICEF United Nations Children’s Fund (2006) UNICEF Ethiopia disability program: key child
disability facts. United Nations Children’s Fund, New York
USAID (2010) Ethiopia–complex emergency, situation report 3, fiscal year (FY) 2010. United
States Agency for International Development, Washington, DC
Workneh N (2008) Food security and productive safety net program in Ethiopia. In: Assefa T (ed)
Digest of Ethiopia’s national policies, strategies and programs. Forum for Social Studies, Addis
Ababa, pp 1–22
World Bank (2006) Productive safety net APL II project in support of the second phase of the
productive safety net program, Report 37224-ET. World Bank, Washington, DC
World Bank (2009) Productive safety net APL III project in support of the third phase of the pro-
ductive safety net program, Report 48633-ET. World Bank, Washington, DC
World Bank (2010) World development report 2010: development and climate change. World
Bank, Washington, DC
 Part V
Responses to Marginality at Different
Levels: State, Business, and Community
Chapter 18
Macro, Fiscal and Decentralization
Options to Address Marginality
and Reach the Extremely Poor

S. Ehtisham Ahmad

Abstract Reaching the extremely poor and marginal groups in countries where
local politicians and officials have little incentive to provide for them is a challenge
to addressing marginality in many countries. Financial and taxation arrangements,
information constraints, and institutional aspects between central and local govern-
ments to address these challenges are identified in this chapter. The instruments
that might be involved include a range of options, from transfers or assistance from
higher levels of government (donors), cash support, or the provision of public services
(particularly health care and education), other forms of income or employment support,
to assistance for small-scale enterprises. In this chapter the author concludes that in
the absence of genuine local interest in providing for the marginalized and extremely
poor, direct provision of funding by central governments and aid agencies appears to
be the main feasible option.

Keywords Fiscal policy • Marginality • Poverty • Transfers • Decentralization

18.1 Introduction

How does one reach the extremely poor and marginal groups in countries where local
politicians and officials may have little incentive to provide for them? This lack of
incentive may be because these groups are minorities in these areas, either due to
ethnicity, language, or caste. In other cases these groups may be predominantly
women and/or children who are not be able to vote and thus exert political pressure
on local politicians (e.g., in some very traditional societies).

S.E. Ahmad (*)

Center for Development Research (ZEF), University of Bonn,
Walter Flex Strasse 3, 53113 Bonn, Germany
Asia Research Center, London School of Economics, London, UK
e-mail: [email protected]; [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 303
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_18,
© The Author(s) 2014
304 S.E. Ahmad

The difficulty for a central government in reaching the extremely poor may be
that it may not have adequate information on the relevant groups, and/or may have
limited ability to provide benefits directly. Providing assistance through tied
transfers to local governments may incur the risk that there could be a diversion of
resources to other interests by local officials or politicians. In other cases, even the
central government may not have the best interests of the extremely poor in mind,
and external aid from foreign governments or altruistic foundations may be subject
to capture and diversion at both the national and local levels of government.
Even in some of the more advanced developing countries, with the possible
exception of Brazil, weak institutional arrangements and inadequate public finan-
cial management systems limit information flows regarding what is spent, by which
level of government, and what the results were. Another aspect contributing to weak
governance arises when local governments do not have access to adequate “own-
source revenues.”1 Under such circumstances incentives to divert resources are
magnified if the implementation of benefit programs for the extremely poor is
handled by local governments, but higher levels of government, or donors, provide
the financing.
A typical problem is when there is an altruistic benefactor, whether a central
government or an external agency, that lacks adequate information on the target
beneficiaries. Examples of such groups could include the Baka and other indige-
nous tribes of the West African rainforest, or girls and women in the northwestern
regions of Pakistan. The relevant information on the target groups may be more
easily available at local levels, where government officials do not necessarily assign
particular importance to the marginalized or extremely poor groups. The problem
can be mitigated, although not eliminated, when the vulnerable groups have a
tradition of political action, or have existing legal or constitutional protection such
as with the scheduled castes and scheduled tribes in India (Gajwani and Zhang
2008; Palaniswamy and Krishnan 2008).
The extremely poor or marginal groups typically do not qualify for categorical
insurance-type benefits, including pensions or unemployment insurance, if they have
not participated in the formal labor force. Their marginality may also constrain their
ability to participate in adequately remunerated informal activities, or to have equal
access to local public services, including basic health care and education. Marginalized
groups often live in remote areas and/or have little education or assets (von Braun
et al. 2009), which is clearly the case with the Baka tribes (World Bank 2011).
Given the importance of institutional structures and associated influence on
incentives to divert resources, I focused here on a typology of options. This high-
lights some of the preconditions that are needed in order to provide effectively for
the most marginal groups. The instruments that might be involved include a range

1
This is defined as some control over revenues at the margin (more likely with respect to the rate
structure than the base). It does not follow that the need for local own-revenues implies that there
should be local administration, indeed even for local property taxes. The cadastre may have to
be kept at higher levels to minimize the incentives to discriminate against the local authorities
opponents.
18 Macro, Fiscal and Decentralization Options to Address Marginality… 305

of options, from transfers or assistance from higher levels of government (donors)

translated into cash support, or the provision of public services (particularly health
care and education), forms of income or employment support, to assistance for small-
scale enterprises.
The alternative delivery mechanisms examined here include a spectrum ranging
from:
• Direct provisions by local governments with their own resources—this generally
works best if the local governments are interested in the welfare of the marginal
groups, which may not be the case
• Provisions by local governments with earmarked financing from higher levels of
government or donors—this may involve more or less conditionality on the
processes and procedures to reach the target groups
• Direct provision by the central government and parallel provision by donors/
NGOs

18.2 Better Governance and Incentive Structures—Hard

Budget Constraints

Well-meaning assistance for the marginalized groups, either from a central govern-
ment or external donors, without proper attention to the overall institutional
framework and fiscal instruments may reduce incentives for efficient provision and
worsen existing governance structures. It may also reduce the political discipline
that comes through operating under hard budget constraints. In this section I focus
on the overall institutional and fiscal framework. I also address the interactions
between instruments, particularly social provision and financing mechanisms. The
overall effectiveness relies on complete and accurate information, effective gover-
nance, and accountability.

18.2.1 Fiscal Instruments and Preferences

Donors, including international agencies and political activists, are increasingly trying
to hold local or subnational governments accountable for the effective delivery of
public services—be those basic health care or primary education of marginalized
groups, or the effective implementation of targeted public works programs [like the
Indian National Rural Employment Guarantee (NREGA)]. The difficulty in many
cases is that it is not clear which level of government holds the functional responsi-
bility and whether this level has control over all the needed economic inputs (e.g.,
employees, wages, operations, maintenance, capital spending) that translate into
programs or sub-programs that can be assessed, such that the cost-effectiveness of
the performance might be measured. Focusing solely on outcomes is not sufficient
in terms of assigning responsibility and achieving cost-effective operations.
306 S.E. Ahmad

Table 18.1 Typical sub-national spending assignments by functions and economic categories as
well as administrative arrangements
Administrative
Economic classification classification
Other economic Central departments/
Functions Wages Other current Capital categories municipal
Education
Primary CCCC CCCCC CCCC M CCCCC CCCCC M
Secondary CCCC CCCCC CCCC CCCCC CCCCC
Health care
Basic care clinics CCCC CCCCC CCCC M CCCC CCCCCC M
Water CCCC CCCCC CCCC M CCCC CCCCC M
Sanitation CCCC CCCCC CCCC M CCCC CCCCC M
Within each broad economic, functional, and administrative grouping, there are several additional
functions that are represented by the full economic classification (see IMF 2001, Government
Financial Statistics Manual), or the UN Classification of the Functions of Government (COFOG).
For instance, we only identify basic preventive health care as a typical local function in advanced
countries, but here it is split as in most developing countries
C represents a central assignment and M represents a municipal assignment

Typically, as in countries as diverse as Bolivia and Cameroon, or even in

Pakistan,2 local governments do not have full responsibility over program functions
or sub-functions. This translates into inadequate control over the economic
categories—wages and salaries, hiring and firing decisions—related to the devolved
functional responsibilities (e.g., basic education). Yet specific health programs or
activities such as building schools are often assigned to local governments that do
not have the full associated economic or functional responsibility. For example in
Cameroon (World Bank 2011) recent attempts to make municipal governments
responsible for the construction of additional schools or clinics, without the corre-
sponding responsibility for providing primary education or basic health care ser-
vices, make it easy for the local government to evade responsibility or accountability
for outcomes. It is quite easy to blame another level of government for the lack of
teachers or health care personnel. A system of overlapping responsibilities cannot
be easily overcome through social action or political sanctions for nonperformance.
The linkages between functions and how these might be implemented are seen in
representations of the typical classification of the budget process (see Table 18.1),
depicting the intersection between functions and sub-functions (rows), and the
inputs classified by economic categories and administrative arrangements (columns).
The responsibilities of lower levels of government appear in designated cells, but
not along the rows (functions) or the columns (economic or administrative categories).
With this pattern of responsibility, it is hard for an electorate to hold government

2
During the 2001–2007 period devolution to districts was attempted by the Musharraf administration
in Pakistan, but central control was strengthened by a weakening of political parties at the provin-
cial level.
18 Macro, Fiscal and Decentralization Options to Address Marginality… 307

responsible for the provision of basic health care, primary education, or even potable
water and sanitation services.3 In the cases presented in Table 18.1, the municipal
government would be responsible for the construction of a school or clinic without
having full responsibility for providing the primary education or basic preventive
care function or sub-function. This makes it easy for the local government to evade
responsibility or accountability for outcomes—as it is quite easy to blame another
level of government for the lack of teachers or health care personnel.

18.2.2 Financing

At the margin, financing mechanisms also matter—both regarding bases over which
the local government has some control (the own-source revenues mentioned below),
the design of transfers from higher levels of government, as well as access to credit.
Poor design in one or more of the dimensions can override any clarification of
spending responsibilities, underlining the importance of interactions across instru-
ments. The lack of accountability at subnational levels often mirrors that found at
the national level (e.g., for countries such as Pakistan). Jurisdictions can become
accustomed to relying on transfers and capital inflows, with little incentive to use
the tax handles that may be available to meet their responsibilities and particularly
to deal with deficits.

18.2.2.1 Own Taxes

From a political economy perspective, the use of own taxes at the margin is critical
for ensuring that the electorates of a locality, region, or country hold governments
responsible for their spending. It is through the need to raise additional revenues to
meet spending needs that there is a clearer political linkage between spending and
financing. This is clearest in the case of debt incurred. Without the ability to raise
additional revenues in order to meet future repayments and liabilities, a local
government is able to pass liability on to the central or other government levels
and related sanctions are just not credible (Ambrosiano and Bordignon 2006).
In such cases there are no hard budget constraints or accountability for spending.
Typically in developing countries there has been relatively little focus on design-
ing adequate sources of subnational own-source revenues (see Ahmad and Brosio
2006, 2009 for a discussion of developing countries). Split bases, such as the Goods
and Services Tax on services in Pakistan are not generally workable, as the crediting
and refunding mechanisms are difficult to handle by subnational governments

3
For a description of the internationally accepted standard for the full structure of the economic
classification system, see IMF (2001) and the UN’s COFOG for the functional classification. The
latter tends to correspond to functions or sub-functions for health care or basic education.
308 S.E. Ahmad

without central coordination and multiple rates make the system even more complex
and open up the possibility of tax wars (Ahmad 2010). Alternatives include dual
Value Added Taxes4 (Bird 2010)—without the establishment of complex adminis-
trations—giving some control to local authorities.
At the local or municipal level, property or land taxes are suitable handles to
generate accountability, but the implementation is patchy (see Ahmad and Brosio
2009 for a review in Latin America). It should be stressed that control over tax rates
or bases does not require that each level of government establish its own tax adminis-
tration. All that is required is that the local jurisdiction be able principally to set its
own tax rates.5
Another alternative is to consider piggybacked income taxes that would also
operate with a single administration. However, this will tend to provide more
revenues to the richest localities, and needs to be supplemented by equalization
mechanisms to permit local governments to be able to provide similar levels of
public services at similar levels of tax effort.

18.2.2.2 Shared Taxes

Tax sharing is quite common and often helps with overall financing needs. Tax
sharing may also be needed for political economy purposes, especially in natural
resource sectors, to prevent centrifugal forces and to help pacify separatist tenden-
cies—as in the case of the sharing of forestry revenues in Cameroon (Morrison et al.
2009; Oyono et al. 2009; Cerutti et al. 2010; Pye-Smith 2010). However, such tax
sharing does not constitute own-source revenues, as additional revenues cannot be
generated independently if needed and it does not promote more efficient governance.
Further, tax sharing does little to support incentives to provide for minorities, such
as the Baka. In this regard tax-sharing revenues are very much like untied transfers,
which are described below.

18.2.2.3 Transfer Design—Gap Filling Fiscal Dentistry?

Deficit-filling transfers negate a recipient jurisdiction’s incentives to use own-source

revenue handles. If deficits are automatically met from higher levels or donors, as
was widespread in the Indian subcontinent (see Rao 1998, who referred to this
phenomenon as fiscal dentistry), there can be little accountability or efficiency
in spending on or targeting vulnerable groups. It can be dangerous when central

4
Typically these involve joint occupancy of the same base, with separate tax rate setting powers.
This does not always involve separate administrations.
5
In unitary states this may involve setting rates within a band legislated at a higher level. In Mexico,
although the constitution is federal with respect to the federal government and the states; the states
and local municipal governments have limited jurisdictional authority and intrastate operations are
effectively unitary.
18 Macro, Fiscal and Decentralization Options to Address Marginality… 309

governments play this game, as has been the case in Pakistan over the past couple of
decades, given its periodic strategic importance. This has led to psychological
dependency, poor governance, and aversion to paying taxes (Ahmad 2010). External
assistance, including for humanitarian purposes, has reinforced these tendencies.

18.2.2.4 Equalization Transfers

Equalization or untied transfers help to meet local government preferences, without

discriminating against regions with low revenue potential or distorting incentives.
Such transfers are desirable in order to supplement own-source revenues, which may
be largely concentrated in the richer areas with more significant revenue bases and are
generally quite disequalizing. But will the local politicians have the incentive to pro-
vide for the marginal groups in the presence of equalization transfers? The short
answer is that local governments will probably not provide for such groups out of
untied resources, and that is one of the contributing factors to marginality problems.

18.2.2.5 Earmarked Transfers

Earmarked transfers are often recommended to meet the objectives of donors or the
central government, specifying conditions that must be fulfilled in order to benefit
from the funds. In some cases, co-financing requirements are also specified.
However, this discriminates against the poorest regions that may not be able to
collect the transfers due to lower than average tax-raising capacity.
With poor information flows on how funds are allocated or accounted for and the
more complex requirements on the outcomes of the spending, the possibility for a
diversion of resources is higher. In the absence of proper benchmarks on spending
and costs across regions and local governments, even social action and the mobili-
zation of public opinion (as is being attempted in relation to the Indian public works
program NREGA in several states) may not be sufficient.
Given that results-based budgeting is an advanced institutional arrangement that
operates mainly in developed countries, withholding funds until the results are dem-
onstrated—the approach now being tried by development partners such as the UK’s
DFID in some countries—may be quite unrealistic. The sanction of withholding
funds in future years may work for investment spending in a repeated games con-
text—if there is both horizontal and inter-temporal competition for funds (Ahmad
2009). However, for basic health care or education services for the most vulnerable
the sanction of withholding funds in future years may not be credible.

18.2.2.6 Tracking the Funds and Establishing Hard Budget Constraints

The effectiveness of provision depends crucially on incentives that the providers may
or may not have to actually spend the funding efficiently and towards the desired
310 S.E. Ahmad

objectives. This depends, as argued above, on some clarity in terms of what these
objectives and responsibilities might be—especially the functional responsibilities
and associated economic components. Tracking these systematically is critical and
a weakness of governance in many countries. The IMF’s Government Financial
Statistics Manual 2001 and the UN’s Classification of the Functions of Government
(COFOG) are essential standards in terms of budgeting, charting of accounts, and
tracking spending.
Tracking funds is also critical, and a typical mechanism used for this purpose is
a Treasury Single Account (TSA)—which consolidates all the government’s cash
and provides a trail of how the funds are disbursed and to whom. If there are lots of
pots of funds in different bank accounts of various agencies, or donor tracking
becomes too difficult, the government loses sight of what is being spent for which
public objective.
Donors frequently do not trust national government budgeting systems and
prefer to establish their own fund allocation mechanisms, as well as associated bank
accounts. This situation has been particularly chaotic in Afghanistan, and has led to
difficulties in other countries as distinct as Bolivia and Timor-Leste. The absence of
complete information on the funds available also makes it very difficult to initiate
social action on the use of funds or to allow the operation of the discipline of yard-
stick competition.
It is possible to use a modified TSA concept for improving the tracking of spending,
especially for lower levels of government and donor agencies. This involves funds
flowing into special “correspondent accounts” within the TSA for local govern-
ments and specific donors. The treasury then stipulates spending limits for zero-
balance accounts in the name of the donor or local government—these accounts
continue to operate such as before—but the funds only flow overnight and there are
no outstanding balances in these accounts. This provides the government with com-
plete information on spending, as well as more efficient financial management. The
donors and local governments operate with their own procedures and safeguards.
Figure 18.1 illustrates a simple correspondent account system within a national
TSA, serving both a hypothetical local government (LG1) and donor agency (TDA).
The main advantage of the modified TSA arrangement for local governments is
the timely and verified generation of complete information on spending for each
local government entity on a consistent basis, facilitating effective cash manage-
ment and creating an electronic trail of payments made. Often as in Cameroon and
indeed Mexico (which also does not use a TSA—see Ahmad in process), and with
disparate and varying standards for budget classification and reporting in different
jurisdictions, the consolidated accounts of local governments are only available
with a considerable lag time of 3 years or more, and are generally not comparable
across local governments.
Without basic information on economic and functional classifications as well as
the TSA, it is very difficult to generate accurate and standardized information on
who spends what amount and for what purpose or when. The question of what
actually happens to the money and how effectively it results in improvements in
living standards or health outcomes is a subsequent and more difficult stage that
presupposes that information on the basic inputs and outputs is available and that
18 Macro, Fiscal and Decentralization Options to Address Marginality… 311

Chart1, Local government and

donor use of correspondent
accounts linked to TSA
MOF

Information
flo s TDA

Payment
request
C/A 2 w
C/A 1

Payee
Payment Order
ZBA

Payment request Payment

Fig. 18.1 Local government and donor use of correspondent accounts linked to a Treasury Single
Account. LG1 is a local government, TDA is a typical donor agency, C/A is a correspondent
account within the TSA, and ZBA is a zero-balance account in a commercial bank, which can be
operated by either the local government or donor. The ministry of finance (MOF) or treasury estab-
lishes spending limits for the ZBA as a function of local government revenues received, or funds
provided by the donor, permitting the local government or TDA to draw from the ZBA up to this
amount. All information for spending is transmitted through the TSA to the MOF, which is able to
produce timely reports on what has actually been spent and for what purpose (Provided that stan-
dard formats and classifications are followed)

standard cost-effectiveness assessments are possible. For social action in the manner
being attempted for NREGA, or yardstick competition, both sets of information are
needed. Reliance by donors solely on evaluation of outcomes is likely to be unrealistic
when basic spending information is not available.
Arrears and the buildup of debt liabilities are also mechanisms that are guaranteed
to lead to inefficient outcomes at the local level. In this manner the local adminis-
trations are able to shift their liabilities on to future generations, the central govern-
ment, donors, or to other jurisdictions. The inter-linkages between policy instruments
are most clearly defined in this case. Without clarity on spending responsibilities, as
well as access to own-source revenues, sanctions on local governments for nonper-
formance or breach of debt limits are not likely to be credible.

18.3 Decentralized Provision and Political Economy

If the transfer or funds made available for a particular purpose are of a “one-time”
nature, the possibility of capture by local vested interests becomes quite acute, espe-
cially if there are broad disbursements and influence, and the intended marginalized
beneficiaries have relatively little voice in local government operations (Bardhan
312 S.E. Ahmad

2006). Repeated targeted transfers also run the risk of being either being appropri-
ated by local officials or used by donor governments for political purposes, resulting
in what is now being increasingly recognized as “clientelism,” especially if the
information on who spends what and for what purpose is hard to attain.
The design of some recent social protection schemes in Latin America originated
in categorical targeting requiring minimal information—attendance in schools and
maternity clinics. This was part of the original design of Mexico’s Progresa/
Oportunidades program that is considered to be the prototype of successful condi-
tional cash transfers (Coady and Harris 2001).6 Applications elsewhere have begun
to focus on asset and income tests that are harder to combine, and that may generate
additional disincentives.
In a multi-year context, a repeated-games framework would permit the operation
of competition for funds across local governments, especially for physical invest-
ments and where the sanction to withhold future funds is credible (Ahmad 2009).
This form of clientelism may occur with the more complex and information
intensive conditional cash transfers, as seen in the Indian subcontinent. In principle,
reaching the poor by offering limited financial incentives and requiring a person’s
time are better categorical targeting mechanisms. However, financing from higher
levels together with poor information flows may contribute to the difficulties of
effective targeting with the Indian NREGA and the social action campaigns being
launched may lack the informational basis to be truly effective.

18.4 Concluding Remarks

In the absence of genuine local interest in providing for the marginalized and
extremely poor, direct provision of funding by central governments and aid agencies
appears to be the only feasible option, which may be reflected by the increasing popu-
larity of centrally sourced conditional cash transfers, such as the Oportunidades pro-
gram in Mexico or the Bolsa project in Brazil. In the Brazilian case, there is adequate
generation of basic information on a standardized basis, but that is not the case in
Mexico (or Pakistan for that matter, see Ahmad in process). In such cases using
simple categorical targeting mechanisms, women through clinics or girls’ through
schools, as vehicles for targeting may be more effective. Similarly public works with
limited financial incentives (based on the 1970s Maharashtra model, see Agarwal
1991) could be an alternative, especially if the community and the targeted margina-
lized groups are able to identify the areas where public works may be most useful.
In the final analysis the difficulty lies with incomplete information on need
and on the actual spending that is available to central governments and donors.

6
However, financing mechanisms for Oportunidades relied on payroll taxes, which in the presence
of labor market informality and evasion, may accentuate disincentives and lower growth potential
(Levy 2008).
18 Macro, Fiscal and Decentralization Options to Address Marginality… 313

Local governments may have better knowledge about need, but may lack the incen-
tives to provide for the marginalized or extremely poor, and indeed may use the lack
of basic information to divert resources. While much depends on individual circum-
stances and institutional arrangements available, in all cases the incentive structures
are important and the interplay between policy instruments needs to be kept in
mind. Above all the basic building blocks for good governance, the ability to track
the funds, and the outcomes of public spending remain as important as individual
interventions.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Agarwal B (1991) Social security and the family: coping with seasonality and calamity in rural
India. In: Ahmad E, Drèze J, Hills J, Sen A (eds) Social security in developing countries.
Clarendon, Oxford, pp 171–246
Ahmad E (2009) The political economy of designing programs to reach the poorest. In: von Braun
J, Hill RV, Pandya-Lorch R (eds) The poorest and hungry: assessments, analyses, and actions.
International Food Policy Research Institute, Washington, DC, pp 237–252
Ahmad E (2010) The political economy of tax reforms in Pakistan, ZEF discussion paper 142.
Center for Development Research, University of Bonn, Bonn
Ahmad E (in process) Governance and institutions. In: Ahmad E, Brosio G (eds) Handbook of fis-
cal federalism II. Edward Elgar, Cheltenham
Ahmad E, Brosio G (eds) (2006) Handbook of fiscal federalism. Edward Elgar, Cheltenham
Ahmad E, Brosio G (2009) The political economy of multi-level tax assignments in Latin America.
National Tax Association, Columbus, pp 191–215
Ambrosiano F, Bordignon M (2006) Positive and normative theories of revenue assignments. In:
Ahmad E, Brosio G (eds) Handbook of fiscal federalism. Edward Elgar, Cheltenham, pp
306–338
Bardhan P (2006) Decentralization and development. In: Ahmad E, Brosio G (eds) Handbook of
fiscal federalism. Edward Elgar, Cheltenham, pp 201–223
Bird R (2010) Central and subnational VATs in federal countries. In: Ahmad E, Al Faris A (eds)
Fiscal reforms in the Middle East. Edward Elgar, Cheltenham, pp 38–75
Cerutti P, Lescuyer G, Assembe-Mvondo S, Tacconi L (2010) The challenges of redistributing
forest-related monetary benefits to local governments: a decade of logging area fees in
Cameroon. Int For Rev 12(2):130–138
Coady DP, Harris RL (2001) A regional general-equilibrium analysis of the welfare impact of cash
transfers: an analysis of Progresa in Mexico. International Food Policy Research Institute,
Washington, DC
Gajwani K, Zhang X (2008) Gender, caste, and public goods provision in Indian village govern-
ments, IFPRI discussion paper no 807. International Food Policy Research Institute,
Washington, DC
IMF (2001) Government fiscal statistics manual 2001. International Monetary Fund, Washington, DC
Levy S (2008) Good intentions, bad outcomes. Brookings Institution, Washington, DC
Morrison K, Cerruti PO, Oyono PR, Stiel M (2009) Broken promises: forestry revenue-sharing in
Cameroon. World Resources Institute, Washington, DC
Oyono PR, Cerutti PO, Morrison K (2009) Forest information and governance in Central Africa.
Center for International Forestry Research and World Resources Institute, Washington, DC
314 S.E. Ahmad

Palaniswamy N, Krishnan N (2008) Local politics, political institutions and public resource allocation,
IFPRI discussion paper no 834. International Food Policy Research Institute, Washington, DC
Pye-Smith C (2010) Cameroon’s hidden harvest. Center for International Forestry Research, Bogor
Rao G (1998) Intergovernmental transfers in India. In: Ahmad E (ed) Financing decentralized
expenditures. Edward Elgar, Cheltenham, pp 234–266
von Braun J, Hill RV, Pandya-Lorch R (2009) The poorest and the hungry: a synthesis of analyses and
actions. In: von Braun J, Hill RV, Pandya-Lorch R (eds) The poorest and hungry: assessments,
analyses, and actions. International Food Policy Research Institute, Washington, DC, pp 1–61
World Bank (2011) Cameroon: the path to decentralization, democracy, development and good
governance, Report no 63369-CM. World Bank, Washington, DC
Chapter 19
Social Protection, Marginality, and Extreme
Poverty: Just Give Money to the Poor?

David Hulme, Joseph Hanlon, and Armando Barrientos

Abstract Social protection programs, especially cash transfer programs, have

spread across low- and middle-income countries since the beginning of the millen-
nium, and are increasingly part of national development strategies to assist the poor
and particularly the poorest. This chapter lays out a wide range of debates about the
specific goals, targets, and conditions of social protection and cash transfers. While
there is no single best program option, the authors identified five overriding princi-
ples for effective efforts. Thus social protection programs and cash transfers work
best when they are: fair, assured, practical, large enough to impact household
income, and popular. These principles need interpretation at the national level,
because no model can be automatically transferred from one country to another.

Keywords Unconditional cash transfers • Trust • Empowerment • Poverty

19.1 Introduction

Social protection programs, especially cash transfer programs, have spread across
low-income and middle-income countries since the beginning of the millennium
and are increasingly part of national development strategies (Hanlon et al. 2010).1

1
This chapter builds on and updates Chapter 10 of Hanlon et al. (2010).
D. Hulme (*) • A. Barrientos
Brooks World Poverty Institute, University of Manchester, Manchester, UK
e-mail: [email protected]; [email protected]
J. Hanlon
Visiting Senior Fellow, Department of International Development,
London School of Economics, UK
Honorary Research Fellow, School of Environment and Development,
University of Manchester, UK
e-mail: [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 315
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_19,
© The Author(s) 2014
316 D. Hulme et al.

When the Millennium Development Goals (MDGs) were being set in 2000–2001
social protection was barely mentioned. Ten years later in 2010 at the United
Nations “MDGs + 10 Summit,” sessions on every one of the eight goals referenced
social protection programs as a means of achieving poverty reduction targets.2
In the short term, well-designed social protection programs reduce poverty levels
and ameliorate related suffering, particularly among the poorest (Barrientos and
Hulme 2008). In the mid-term they allow many poor people to exercise their agency
and pursue micro-level plans to increase their productivity and income. In the lon-
ger term, they create a generation of healthier and better educated people who can
seize economic opportunities and contribute to broad-based economic growth.
Additionally, when sudden crises spread across the world—as with the 2008 triple
whammy of global food, fuel and financial crises—these programs help poor fami-
lies cope with the unpredicted consequences of globalization.
Expanding across the Global South is a heterodox analysis that sees well-
designed social protection schemes as contributing to several goals at the same time.
Brazil, Mexico, South Africa, India, Indonesia, and China are paving the way, but
the idea is spreading via a genuinely southern revolution.3 The focus is on trusting
the poor to use money wisely and to pursue objectives that poor people already want
to do—send their children to school, improve their diet, and make small investments
to increase their income. Cash transfers work especially well when money is tar-
geted at a relatively large group of people and is perceived as social assistance.
Aid donors from developed nations, now calling themselves development part-
ners, have responded after seeing that cash transfers have become an important part
of social protection and increasingly recognize them as a legitimate development
strategy. Aid donors and some elites in developing countries (especially ministers of
finance and business elites), however, have much less trust in the poor. They tend to
use terms like “safety nets,” “handouts,” and “doles” derogatorily, and to view trans-
fers as temporary or short-term solutions, rather than strategic elements that can
help raise agricultural productivity, support micro-enterprise, and foster economic
activity in marginalized regions. From the USA, where the poor are often blamed
for their poverty, the tendency is to emphasize the centrality of conditions—such as
school attendance for recipients’ children or engaging in public service—rather than
to recognize conditionality as one design element of a program. Some programs
work better with conditionality while others do not. Schady, one of the authors of
the World Bank report Conditional Cash Transfers (CCTs) is a strong advocate of
conditions because he is convinced that the problem is the “persistently misguided
beliefs” of the poor (Fizbein and Schady 2009, 9).4 Indeed in the USA, many spe-
cialists are not aware that unconditional cash transfers are a major element of social

2
Hulme attended the MDGs + 10 Summit and witnessed this by agencies keen to incorporate social
protection, social floors, and conditional cash transfers into future strategies.
3
During the preparation of this manuscript Hulme was working with the government of Bangladesh
on a national social protection strategy and the Asian Development Bank reviewing the “silent
revolution” of social protection policy that has swept across Asia.
4
Speaking at a meeting in London on May 26, 2009.
19 Social Protection, Marginality, and Extreme Poverty… 317

protection strategies across Asia and Africa: they talk only of CCTs because in the
USA the general assumption is that all transfers must be conditional.
From Europe and especially from European donors, there is less suspicion of the
poor and greater support for social protection and unconditional cash transfers, but
also a tendency to view cash transfers as welfare or antipoverty measures, rather
than as an essential component of national developmental strategies. The emphasis
of European development efforts is on targeting to minimize expenditures and to
ensure that the ultra-poor are the main (or only) beneficiaries, because poverty
reduction rather than national development is often the goal.
These three tendencies (Global South, USA, and European) reflect different
responses, both to recent economic crises and to the centrality of market-led develop-
ment. Back in 1944 the renowned economic historian Polanyi outlined what he called
the “double movement” in the development of modern capitalism. The “self-regulating
market” was promoted by governments with a host of special legislation and it did
bring growth, but this free market could not work for labor and the environment. Low
wages and unemployment created poverty and misery on a large scale, leading to the
counter movement for “social protection,” which was increasingly supported in
society, forcing governments to intervene in markets (Polanyi 2001; Putzel 2002). The
failure of structural adjustments made in the 1980s and early 1990s to promote
economic growth in low-income countries, and of “social funds” to protect the poor
and newly poor from capability depleting vulnerabilities, showed that palliative
approaches to social protection were ineffective. The lack of development and the
growth of poverty outside of China and India, culminating in the 1997 Asian financial
crisis, demonstrated the failures of neo-liberalism and the “Washington Consensus,”
and brought the second part of the double movement—the push for social protection.
The northern response, epitomized by the MDGs, provides a real increase in social
spending and strict social targets for health and education, but also kept governments
at arm’s length from the economy. The southern response, with Brazil’s Bolsa familia
as a classic example, has been cash transfers that shift the emphasis to providing poor
people with the money necessary for them to take actions to end their own poverty and
make greater use of economic opportunities.5
Mexico, Brazil, and South Africa introduced large-scale social protection
programs in the late 1990s in a Polanyian response to the harm being done by a sole
focus on the market as the answer to all development problems. This was a challenge
to the Washington Consensus, because the programs were (and are) redistributive
and involve governments reallocating resources within countries. The initial
response from the North was criticism and the commissioning of studies (Barrientos
and Hulme 2008). These studies found that social protection programs worked and
that cash transfer programs (with and without conditions) performed particularly
well. They reduce both immediate and intergenerational poverty, stimulate the
economy, and promote development—especially in marginal rural areas as shown
in South Africa (Samson et al. 2004).

5
Brazil, Bolsa Família, see “A Biblioteca Virtual do Bolsa Família,” at http://www.ipc-undp.org/mds.do
318 D. Hulme et al.

Word of the success of the three trailblazers has spread across the Global South.6
At least 45 countries now have cash transfer programs, providing financial support
to more than 110 million families (Barrientos and Hulme 2010). Immediate poverty
is being reduced by these efforts. Hundreds of thousands of children are now in
school because their families can afford to buy them shoes and school clothes, and
can get by without the few pennies the children could earn if kept out of school.
Families are also investing small amounts to raise their own income with, for exam-
ple, cash transfers in rural Africa frequently being used to purchase fertilizer. These
programs are still young and this is a global learning exercise. Experiences are
being shared among continents; research and experimentation are leading to rapid
modifications and improvements in similar programs. Serious analysts also recog-
nize that this is not a quest for the “optimal program,” but a contribution to the
evolution of national social protection systems that need to be able to improve their
performance, maintain political support, and adapt over time in response to changes
in national problems and the international context.
The influence of the USA and European tendencies on the development of south-
ern programs varies from country to country, but key patterns can be recognized. In
middle-income and larger countries domestic policy makers can “listen” to the USA
conditionality and European targeting tendencies, and then decide how best to deal
with these from a position of relative autonomy. In lower-income countries and in
particular those that are smaller and aid dependent, the degree of autonomy is
restricted and there is a real danger that the policies selected derive from external
debates rather than being part of a nationally-owned policy discussion. The danger
of aid donors dominating national policy formulation about social protection
becomes a real prospect in such cases. Where there is a national political consensus
on an action, however, it can be pushed through despite donor concerns.

19.2 Building Local Support

Building the required local political consensus for change is critical, but can be
complex and will surely be different in each country. Political elites need to champion
the role that effective social protection can play in national development, and the
economic and business elites at least need to recognize the legitimacy of social
protection and cash transfers. Debates at the global level are also reflected nationally.
The wealthier in poor countries often hold the attitudes of Victorian Britain, blaming
the undeserving poor for their poverty. These attitudes have been reinforced by three
decades of neoliberalism, which promoted the idea that it was “good” to get rich.
A study of economic elites in five southern countries found a consensus that “trickle

6
The achievements of programs in Mexico, Brazil, and South Africa were rapidly and widely
disseminated. Other successful programs such as Bangladesh’s Secondary School Stipend for
Girls (basically a conditional cash transfer to increase female participation in secondary educa-
tion), were not so well “marketed” and other Asian successes are much less reported.
19 Social Protection, Marginality, and Extreme Poverty… 319

down” growth—growth in the economy as a whole working its way down to the
poorest—would eventually end poverty and that education was the most important
means for improving the welfare of the poor (Reis and Moore 2005).
In a study of economic elites in Malawi, some admitted that they actually benefit
from poverty (Kalebe-Nyamongo 2009). The poor provide cheap labor and votes, as
well as jobs in the aid industry. However, there was not a perception among the elite
in Malawi that the poor are lazy, and thus there was not a distinction between
“deserving” and “undeserving” poor. Nevertheless the elites believed that poverty
will never be reduced and that the poor were responsible for their own poverty—that
the poor do not work hard because they are resigned to poverty as a normal way of
life—and thus the elites worried that cash transfers create dependency. This may be
reflective of Malawi’s experience with a very narrow cash transfer program that
only targeted the labor constrained ultra-poor. In the mid-2000s the government’s
preference for a fertilizer subsidy program for small farmers, rather than a cash
transfer program for the poor, reflected the elitist view that people should be encour-
aged to work (to its credit, the subsidy did raise productivity and improve household
food security very quickly).
By contrast in Brazil only 1 % of a sample of 311 members of the economic elite
blamed poverty on a lack of effort by the poor (Reis 2005, 2009). Nearly half blamed
a lack of state efforts or the lack of political will, and most elites believed that the
state has a responsibility to provide for the poor. Brazilian elites view poverty and
inequality as problems, and are particularly concerned about their relationships with
criminal violence. Perhaps it is not surprising in a country where the poor are not
blamed and the state is given responsibility, that Brazil has taken a global lead in
designing innovative social protection programs (including the systems that allow
them to work well, such as unique identification and national-level data coordina-
tion) and in financing cash transfers.
Politics are important, and effective social protection programs can only be intro-
duced when a critical mass of support can be created. Hickey (2006) makes the
point that civil society does not seem to play an important role in the introduction of
cash transfer programs in Africa because the old, the poor, and the weak do not
constitute politically active pressure groups. Thus it becomes essential to mobilize
support within government, parliament, and political parties, and to use the electoral
process. It is necessary to tap into the social responsibility attitudes of the business
and social elites, and to convince both the elites and the middles classes that social
protection programs are in their best interest—they will help create the conditions
that the elites want, i.e., growing economies and more cohesive societies. The gen-
eral shift from a narrow welfare approach targeting only the poorest to a broader
developmental approach that views social protection, not as poverty reduction but
as national development, seems likely to make it easier to gain support.
Within governments, transfers (whether cash or in-kind) are often the responsi-
bility of administratively and politically weaker ministries that deal with social
welfare or support for women and children. Finance and economic development
ministries, which typically have more political and intellectual clout inside govern-
ments, need to address social protection as an essential component of social
320 D. Hulme et al.

development programs. That in turn means making the case that cash transfers are
not just instruments for mitigating current poverty, but are profitable investments
in long-term development that are as effective as roads and dams (indeed more
effective than many dam projects that are often completed vastly over budget and
under capacity).
Why do low-income governments often prefer fertilizer subsidies to social
welfare programs? Why do donors appear to prefer welfare programs to fertilizer
subsidies? Governments tend to view enhanced access to agricultural inputs as an
investment in production, food security, and economic growth. Conversely they
often regard welfare handouts to widows and orphans as an unaffordable luxury that
generates apathy and dependency among the poor. A great deal of effort has been
expended trying to convince skeptical governments (especially among finance
ministries) that they are wrong about cash transfers, which can achieve poverty
reduction and generate economic growth, but do not necessarily generate depen-
dency. This argument, however, is not yet won according to Devereux (2009).
Making the case for social protection in low-income countries requires making it look polit-
ically attractive as well as fiscally affordable, to stakeholders who have to commit to it and
will eventually have to pay for it. Politicians are more interested in evidence that social
protection will win votes than in evidence that it reduces poverty

This point is underlined in a study of drought relief efforts in Africa by Munemo

(2008), who examined why some governments have preferred universal food aid
while others—or the same governments at different times—have opted for work-
based relief schemes. Munemo finds, not surprisingly, that incumbent governments
that are vulnerable to the loss of political power tend to prefer measures with imme-
diate benefits for a broader base of voters. Political groups that are more secure in
their position can afford to opt for programs that are more targeted on smaller
groups and that are developmental with longer temporal horizons.
Political and even patronage power need to be mobilized in support of social
protection programs. This will vary radically among countries. For example both
China and Brazil have decentralized their cash transfer programs to municipalities,
which allows local elected officials to take the political credit. By contrast, in
Bangladesh political parties and politicians are loath to provide social protection
programs in urban areas where they do not convey influence over voting behaviors
in the ways that social protection efforts in rural areas do.
Public attitudes are important and the media affect the way that people think. The
prevalence of articles about “welfare scroungers” contributes to a climate that
stresses the “undeserving poor,” while articles about rising school attendance and
new businesses begun on a shoestring—although less common—support approaches
designed to give people the economic security they need to take the risks and make
the micro-investments that lift them out of household poverty traps. There is also a
mutually reinforcing relationship between political leadership and media coverage.
In Ghana the Ministry of Manpower did not wait for media coverage, but instead
launched a strong advocacy media campaign to explain to the public that giving
money to poor people is not about “handouts,” but rather about support for children,
19 Social Protection, Marginality, and Extreme Poverty… 321

the elderly, and those who are not able to work (Sultan and Schrofer 2008). In a
study of non-contributory pensions for the Chronic Poverty Research Centre in
Lesotho, Namibia, and South Africa, Pelham (2007) concluded that successful
programs built a bond between citizens and the state based on three factors—social
solidarity linked to the value and contribution of the elderly in the household, the
understanding that pensions are a permanent program that can be depended on, and
acknowledgment of the role of the state in securing the welfare of its citizens. Local
and global events can play important roles in creating change and openings for the
introduction of social protection policies. The Zapatista rebellion in Mexico, the
end of apartheid in South Africa, and the end of the military dictatorship in Brazil
all created space for policy changes. Hickey (2006, 25) points to the way social
protection policies come to the fore “when the social impacts of liberalized capital-
ist economies become too great to be borne in political terms,” which was happen-
ing in Brazil and Mexico—the other half of Polanyi’s double movement.
Hickey’s research underlines the importance of electoral politics. Political par-
ties attempting to stay in power or win power can use cash transfers to win new
constituencies or to strengthen existing ones. The balance can be quite complex.
Middle class support is essential, and Hickey points to the need to include not just
the very poorest. Programs which benefit the middle class are more likely to win
support. Fairness, justice, and social responsibility seem to be important, but are
often relatively intangible elements of such political processes.
Brazil provides a particular example of the political dynamics at work. Brazil
had come out of the 1964–1988 period of military dictatorship and the 1988
constitution stipulated poverty alleviation as a role of the state. The idea of a child
benefit as a first step toward a basic income was proposed by the Workers Party
[Partido dos Trabalhadores (PT)] and entered public debate. Because of decen-
tralization in Brazil, the idea was first picked up by municipalities and by 1998,
60 municipalities and four states had introduced a variety of child benefit pro-
grams. Fernando Henrique Cardoso was running for re-election as national presi-
dent in 1998 against the PT’s Luiz Inácio Lula da Silva (“Lula”), so he adopted
the child benefit (as the Bolsa Escola school grant) as a national program. Cardoso
won a second term and Bolsa Escola was scaled up. Lula subsequently won the
presidential election in 2002 and expanded the program as the Family Grant
(Bolsa Família) (Melo 2007).
Expanding the program increased Lula’s popularity and he was re-elected in
2006. “I like Lula a lot—he gave us Bolsa Família. Many people today have a
better life,” Selma Aguiar who runs a luncheonette in Vale do Mearim in the
state of Maranhão later told BBC Brazil (Corrêa 2009). “He has improved our
life, and that of many families, a lot. I receive R$122 [US$67] per month. I voted
for Lula and I will vote for him again,” added Eliene da Silva Brito, a farmer
with five children.
In Lesotho the government was re-elected in 2008, in part because of the popu-
larity of the pension introduced in 2004. In Mexico the cash transfer not only sur-
vived the historic change in the government of Mexico in the 2000 elections, but the
new administration of President Vicente Fox expanded its coverage from rural to
322 D. Hulme et al.

urban poor areas of the country, renaming it from Progresa to Oportunidades.7 The
program was politically popular because of the overwhelming and unprecedented
evidence that it was alleviating poverty and encouraging the rural poor to send their
children to school (Skoufias 2005). In an opinion survey conducted in South Africa,
exactly two-thirds of the population agreed with the statement “the government
should spend more money on social grants for the poor, even if it means higher
taxes” (Noble et al. 2008). Perhaps more importantly taxpayers also agreed—the
statement had the support of 59 % of the poor and 63 % of paid workers. So in South
Africa social grants also win votes. In Bangladesh the incoming Awami League
government of 2009 cancelled the predecessor government’s “100 days work” cash-
transfer scheme, but immediately launched a new but similar scheme so as not to
lose political popularity.
Globally, social pensions seem to be the most popular programs for national
governments: they are inclusive, satisfy instinctive desire to support the elderly, are
good value, and are perceived as fair. For purely selfish reasons voters are attracted
to the idea of universal pensions because it provides some peace of mind regarding
one’s own fate, or that of a grandparent, aunt, friend, or neighbor, during old age.
Finance and social welfare ministries also recognize that older people spend a sig-
nificant part of their pension on children and others in the household, so pensions
have a broader demographic impact. Child benefit programs, as in the case of South
Africa, have similar broad appeal.
It appears that social protection programs, and especially cash transfers, can start
small. Successful programs, however, are not narrowly targeted at specific groups
with whom most voters cannot identify. They are established in a way that makes
expansion obvious and possible—targeted on the poorest districts or individual
municipalities which makes expansion to other districts a political goal, or targeted
at poor children which makes expansion to a comprehensive child benefit seem
reasonable.
Nicaragua’s donor financed “Social Protection Network” [Red de Protección
Social (RPS)] showed what can happen to a program without broad support. Its first
phase from 2000 to 2002 was spectacularly successful, improving nutrition, health,
and school attendance rates (Moore 2009). For its second phase during 2002–2005,
however, cash payments were cut from US$19 per family a month to US$12.
Although the second phase was also successful, the program was abandoned in
2005 by the government, even though continued funding was available. It was a
textbook case of how not to design a popular program. First, in order to ensure that
the program was not seen as “welfare,” it never mentioned poverty reduction as one
of its goals. Instead the cash was presented purely as a way of buying behavioral
change of the poor in order to build human capital and thus had no buy-in from the
vast majority of people, who felt they already knew how to look after their children

7
See Mexico—Oportunidades external evaluations in English at http://evaluacion.oportunidades.
gob.mx:8010/en/index.php and in Spanish at http://evaluacion.oportunidades.gob.mx:8010/es/
index.php
19 Social Protection, Marginality, and Extreme Poverty… 323

and thus would never benefit. Second, the program was almost entirely driven by the
Inter-American Development Bank (IDB) and by a small group of key civil servants
with experience in World Bank and IDB programs within the Emergency Social
Investment Fund [Fondo de Inversión Social de Emergencia (FISE)].
Program administrators were under a large amount of pressure from the IDB to
implement the RPS quickly, so they did not have time to build political support in
congress with then President Enrique Bolaños and his Constitutional Liberal Party
(PLC) or with the opposition party (Frente Sandinista de Liberacion Nacional) and
their leader Daniel Ortega who was subsequently elected president in 2006 and was
openly opposed to the RPS. Although there was international praise for the pro-
gram, there was little domestic publicity for its success. Finally RPS proponents
even failed to build support within the civil service. In 2002 it was moved from the
FISE to the weaker Family Ministry, where there was discontent about the higher,
donor funded salaries of the RPS staff. It was the family minister who decided in
2005 not to continue the RPS despite continued availability of the IDB funds.
Honduras provides another example of the dangers of a small and politicized
program. Honduras has been experimenting with its family grant program [Programa
de Asignación Familiar (PRAF)] for more than a decade. It was too small, however,
to gain widespread support and created competition between the IDB and the gov-
ernment. Beneficiaries were often chosen on political grounds rather than on need.
Newly elected presidents in both 2002 and 2006 dismissed the entire PRAF staff
and appointed new people. Not surprisingly PRAF became identified as a political
project of the governing party (Moore 2008; Osório 2008). Nicaragua and Honduras
are, however, exceptions to a broader pattern. Where Southern governments have
been able to take the lead and build a political consensus reflecting local conditions
and history, programs are generally proving popular, effective, and durable.

19.3 Cash Transfers Work

Cash transfers can provide the essential boost necessary to lift people out of the
poverty trap—they can be the proverbial boots that allow people to pull themselves
up by their bootstraps. Virtually all evidence supports the conclusions that people
typically spend the assistance money in meaningful ways and that the grants do not
encourage people to be lazy or avoid work (Hanlon et al. 2010). For most poor
people a lack of money is their biggest problem. Small farmers in Malawi do not
need agricultural extension workers to tell them to use fertilizer on their maize, they
need US$3.50 to buy half a bag of subsidized fertilizer. The Oportunidades program
recipients in Mexico have convinced even policy makers that they already knew
how to make profitable investments and that all they needed was the money. Giving
people money is proving to be amongst the best ways to stimulate local economic
development in low-income countries. Social protection and cash transfers are not
social programs that can wait until after economic growth has been achieved, instead
they are an essential precursor to growth and a driver of development.
324 D. Hulme et al.

Vuyiswa Magadla lived in a tiny house at the end of an alley in the Khayelitsha
township of Cape Town, South Africa. She had diabetes and could not walk much
or see well, and received a disability grant. She continued to work selling fruit and
vegetables, and used money from her grant to buy fresh vegetables. She may not
have been healthy, but she was a good trader and traveled quite far to a place in
Nyanga East to buy produce, rather than from Mabhela’s wholesale closer to her
home where vegetables were not as fresh. Fortunately the minibus driver did not
charge her extra for her box of vegetables because she carried it on her lap (du Toit
and Neves 2006).
For the poorest, the elderly, and the disabled, cash transfers are essential social
welfare that can lift people out of destitution and allow them to buy food for a sec-
ond meal a day, but the importance of cash transfers is much broader. A key benefit
is enabling children to be better off as adults than their parents. This starts with more
and better food when they are young, which prevents malnutrition, which is critical
because the stunting caused by malnutrition is a mental as well as physical phenom-
enon, and children who do not develop properly never recapture that lost physical
and mental development. Cash also allows children to attend school—they do not
have to work to help support their family and money is available for clothes and
books. Children who finish secondary school are much less likely to be poor as
adults than those who do not. Cash transfers can be a critical investment in the next
generation and in the long-term elimination of poverty (Hanlon et al. 2010).

19.4 The Southern Alternative

Over the past decade social protection programs and particularly cash transfers have
emerged as the response of the Global South to economic development and poverty
reduction needs. The Washington Consensus free-market model of the 1980s and
1990s did not bring economic growth and development to most developing coun-
tries, but rather often lead to increased poverty and inequality. In the industrialized
North, social protection and cash-benefit schemes expanded rapidly in the second
half of the twentieth century, but these were largely insurance-based schemes that
depended on a household breadwinner with a steady job. In the South, however,
most people are small farmers, landless laborers, or are working in informal market
sectors and cannot qualify for insurance coverage. The USA variant of the northern
model is also built on an inherent distrust of poor people and on the assumption that
the poor are a relatively small group—an assumption that is getting harder to sup-
port considering that 15 % of USA families required federal food assistance in 2011
(Izzo 2011).
The Global South has been rethinking the problem from the bottom up. Poor
people who have struggled to survive on tiny amounts of cash are good economists
who use additional money efficiently. Giving money directly to poor people is prov-
ing to solve three problems at once. First it alleviates immediate problems associ-
ated with poverty; much of the money is spent on more and better food. Second it
19 Social Protection, Marginality, and Extreme Poverty… 325

allows poor people to invest small amounts in their farms and small businesses,
and all evidence supports the conclusion that ordinary people know how to make
profitable investments. Furthermore money is spent locally, which stimulates local
economies, increasing local demand and creating an upward economic spiral. Third,
poor families can send their better fed children to school, creating a healthier and
better educated next generation that will be more able to play an active role in devel-
opment. The key is to trust the judgment of poor people and to prioritize directly
giving them cash—rather than projects, temporary welfare or vouchers (unless there
is evidence from experimental schemes that vouchers work better)—money they
can invest, use, and rely on. Cash transfers can be a key part of the ladder that lets
people climb out of poverty traps. Letting people make their own choices about how
to spend money is also empowering.
The late twentieth century was a very conservative period. The North, as well as
economic elites in many countries of the South, tended to blame the poor for their
poverty, and some still do not believe that poor people are able to act in the best
interests of their children. The first southern cash transfer programs began in middle-
income countries, which could fund them with their own tax revenue and that were
under increasing domestic political pressure to deal with worsening poverty associ-
ated with structural adjustment policies. The high degree of distrust meant that these
programs have been extensively studied—all the more so when “experts” simply
did not believe the initial results showing how well cash transfers were working.
The whole exercise began experimentally, but these programs were big, giving
money to millions of families—not just the poorest of the poor, but also to larger
groups who were below the poverty line. Each country started its programs differ-
ently. Programs were modified in response to initial research, and the ideas and
experiences spread as more countries introduced cash transfer programs. Research
continued and there was increased sharing of information, experience, and ideas.
One lesson was that cash transfers are not a magic bullet; they do not work on
their own. There must be schools and health posts that poor people have access to,
as well as to land, improved agricultural technologies, and jobs. The biggest lesson,
however, has been that people must have the minimum amount of money necessary
to take advantage of schools, health facilities, and land. If they do have sufficient
money, they can take the lead in their own development.

19.5 Northern Responses and Opportunities

Middle-income countries, particularly Mexico, Brazil, and South Africa pioneered

the development of cash transfers, and their early success led many low-income
countries to start programs that attracted donors and international development
banks. This increased tensions, because aid donors and international agencies with
headquarters in the North worried that simply giving money to the poor would waste
their financial aid and for some, reduce their power. Some of the Washington,
DC-based institutions, particularly the International Monetary Fund (IMF), refuse to
326 D. Hulme et al.

trust the judgment of the poor and have tried to retain power by imposing expenditure
ceilings and harsh conditions. Some of the European donors find it hard to accept that
cash transfers can contribute to development and thus might replace some of their
traditional development programs, and therefore tend to support cash transfers only
as social welfare.
Northern institutions are trying to catch up, but the initiative and the action
remains in the South. The southern model of cash transfers is new, and in a learn-
ing culture it is evolving rapidly and it is not just about poverty and welfare. This
movement has been conceptualized in the south and has driven a rethinking of
development models—give money to the poor because they can be trusted to
make better use of it than aid industry project officers and social workers. Lessons
are being learned and experiences exchanged. New York is drawing on the
experiences of both South Africa and Mexico. Indonesia is also drawing on the
experiences of Mexico and Ghana on Brazil’s. Pensions have spread from South
Africa to neighboring states. South-South cooperation is challenging the estab-
lished wisdom of Washington and London as ideas diffuse from Brazil, India,
Mexico, and elsewhere.
Just as cash transfers in the South have been mold-breaking, however, the most
interesting responses in the North have come from outside the normal aid and devel-
opment network. US President Barak Obama specifically invited Indonesian
President Susilo Bambang Yudhoyono to speak at the G20 meeting in Pittsburgh in
September of 2009 to promote his policy of phasing out fuel subsidies and giving
the money to the poor via cash transfers instead. It was an unexpected convergence
of several policy lines. From a climate change perspective Obama was promoting an
end to fossil fuel subsidies. Just 2 months beforehand Yudhoyono had been re-
elected president with more than 60 % of the vote, and his landslide victory was
credited in part to increased support from the poor through cash transfers (Harvey
and Luce 2009; Perry 2009). Thus climate change, democracy, and cash transfers to
help the poor are coming together in a new way.
One question often raised is how social assistance transfers can be delivered.
Cash transfers provide an obvious solution. The most important point is that suc-
cessful cash transfers are both locally designed and transparent—local people
make the choices about targeting and whether or not to impose conditions, but
the distribution of funds is much easier to audit than conventional aid funded
development projects. Thus northern taxpayers and donors could be more
adequately assured that their money was being distributed to children or poor
families or the elderly.
Although the poorest countries, particularly in Africa, can afford limited cash
transfer programs with their own resources, these programs will need outside
financial support for years to come in order to be effective. The shift in thinking,
away from conventional aid projects toward budget support and block grants,
points to cash transfers as the most effective way to distribute social assistance
money—reducing poverty while promoting development, and (if skillfully
designed) slowing climate change and reducing the likelihood of another finan-
cial crisis.
19 Social Protection, Marginality, and Extreme Poverty… 327

19.6 Five Principles

This chapter (and our book Just Give Money to the Poor) has laid out a wide range
of debates about the specific goals, targets, and conditions of social protection and
cash transfers.8 Each country is developing its own model that reflects its own needs,
history, and politics. There is no single “best” program option, but we can outline
five overriding principles. Social protection programs and cash transfers work best
when they are: fair, assured, practical, large enough to impact household income,
and popular.
Fair. Transfers and grants must be perceived as fair, in that most citizens agree on
the choice of who receives money and who does not. Categorical grants—giving to
all or nearly all children or elderly—are usually seen as fair, but may not always
target the neediest. A strategy of excluding the better off, as used in Brazil and
South Africa, is sometimes seen as fairer than trying to distinguish between shades
of poverty. Targeting the poorest or the ultra-poor requires much more care, because
it can be divisive and create conflict between neighbors when some receive support
grants and others do not. Proxy means tests may be relatively accurate, but they are
not easily understood by beneficiaries. There is also the problem of poverty dynam-
ics. A household identified as poor 3 months ago (when the household head was
sick and unable to work) may not be poor today. The household next door, identified
as not being poor 3 months ago may now facing destitution after recent floods
washed away its property.
Assured. Recipients must be certain that every month the money will arrive and that
families can depend on it. Only then will families be able to make long-term plans
like investments in education and income generation. The insurance function of
grants is important because people know that if their crop fails or they fall ill, they
will continue to receive some income—and this security allows people to take risks,
such as experiment with new crops or travel farther to look for work.
Practical. Directly related to the previous two principles, there must be a system to
fairly identify recipients and ensure they regularly receive their grant. This requires
sufficient trained civil servants to oversee and audit the system, and a reliable, secure
banking or cash distribution system to deliver payments. Sophisticated proxy means
tests and complex conditions are of little use if they cannot be applied correctly and
consistently. Some countries have much more experienced civil services than others

8
Other useful websites include: the United Nations Development Programme (UNDP) Inter-
national Policy Centre for Inclusive Growth (IPC-IG), formerly the International Poverty
Centre, in Brazil has an excellent Cash Transfers and Social Protection archive at http://www.
ipc-undp.org/cct.do; the Social Assistance in developing countries database 2007 at http://www.
chronicpoverty.org/pubfiles/socialassistancedatabase_ version4_august2008.pdfversion 4.0; the
Brooks World Poverty Institute, University of Manchester working papers on social protection at
http://www.bwpi.manchester.ac.uk/; and the Chronic Poverty Research Centre, University of
Manchester research on Insecurity, Risk and Vulnerability at http://www.chronicpoverty.org/
research-themes-vulnerability.php
328 D. Hulme et al.

and some countries, particularly in Africa, may be forced to adopt simpler systems
because of a lack of capacity. There are a growing number of effective innovations
for transferring cash where civil services are weak: through post offices, lottery
agents, and even mobile phones.
Not Pennies. Grants must be large enough to cause a real change in behavior, such
as growing new crops or sending children to school. For example, if the amount of
money is only enough to let one extra child in the family go to school it may be
insufficient. In rural African communities where cash incomes are very low and
people produce a significant amount of their own food, even a few dollars a month
makes a huge difference in the choices they can make. In more industrialized coun-
tries where the cash poverty line and cost of living are higher, it requires more
money to make a meaningful difference. Indications are that grants should ideally
be not less than 20 % of household consumption and where this level is not met the
grants are unlikely to have the desired effect.
Popular. Any social protection or grant program must be politically acceptable and
hopefully are popular enough to win votes. Cash transfers are an important step on
the road towards achieving social contracts in developing countries, towards replac-
ing conflict and corruption with solidarity and social bonds. Donor initiated and
driven programs are less likely to win approval than ones that have indigenous roots,
even if they have an “optimal” design they will not have local ownership.
These principles need interpretation at the national level, because no models can
be automatically transferred from one country to another. High quality technical
analysis is needed alongside the recognition that effective programs need to be
based on local political support. Each government will juggle goals and competing
demands for resources, and develop its own approach. A decade of experience,
however, shows that cash transfers work. To reduce poverty and promote develop-
ment, just give money to the poor.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Barrientos A, Hulme D (2008) Social protection for the poor and poorest: concepts, policies and
politics. Routledge, London
Barrientos A, Hulme D (2010) Social protection for the poor and poorest in developing countries:
reflections on a quiet revolution. Oxford Dev Stud 38(1):439–456
Corrêa A (2009) Desabrigados ainda Reverenciam Lula, mas Ignoram Dilma. BBC Brasil, 29 May
Devereux S (2009) Social protection and the global crisis. Regional Hunger and Vulnerability
Programme, Johannesburg
du Toit A, Neves D (2006) Vulnerability and social protection at the margins of the formal econ-
omy—case studies from Khyelitsha and the Eastern Cape. United States Agency for
International Development and Programme for Land and Agrarian Studies, Cape Town
19 Social Protection, Marginality, and Extreme Poverty… 329

Fizbein A, Schady N (2009) Conditional cash transfers. World Bank, Washington, DC

Hanlon J, Barrientos A, Hulme D (2010) Just give money to the poor. Kumarian Press, West
Hartford
Harvey F, Luce E (2009) US backs ending of fossil fuel subsidies. Financial Times, 24 September
Hickey S (2006) The politics of what works in reducing chronic poverty, Chronic poverty research
centre working paper 91. Chronic Poverty Research Centre, Manchester
Izzo P (2011) Some 15 % of US uses food stamps. The Wall Street Journal, 1 November
Kalebe-Nyamongo C (2009) Elite perceptions of poverty in Malawi. Paper presented at the roles
of elites in economic development conference, Helsinki, 12–13 June 2009
Melo M (2007) Political competition can be positive: embedding cash transfer programmes in
Brazil. In: Bebbington A, McCourt W (eds) Development success: statecraft in the south.
Palgrave Macmillan, Basingstoke
Moore C (2008) Assessing Honduras’ CCT Programme PRAF, Programa de Asignación Familiar:
Expected and Unexpected Results, Country study 15. International Poverty Centre, United
Nations Development Programme, Brasilia
Moore C (2009) Nicaragua’s Red de Protección Social: An Exemplary but Short-Lived Conditional
Cash Transfer Programme, Country study 17. International Poverty Centre, United Nations
Development Programme, Brasilia
Munemo N (2008) Political incumbency and drought relief in Africa. In: Barrientos A, Hulme D
(eds) Social protection for the poor and the poorest. Palgrave Macmillan, Basingstoke
Noble M, Ntshongwana P, Surender R (2008) Attitudes to work and social security in South Africa.
Human Science Research Council, Capetown. Available at www.hsrcpress.ac.za
Osório R (2008) The recent impact of government transfers on poverty in Honduras and alternatives
to enhance their effects, Working paper 47. International Poverty Centre, United Nations
Development Programme, Brasilia
Pelham L (2007) The politics behind the non-contributory old age social pensions in Lesotho,
Namibia, and South Africa, Working paper 83. Chronic Poverty Research Centre, University of
Manchester, Manchester
Perry N (2009) Indonesians choose reformist Yudhoyono. UPI-Asia, 13 July
Polanyi K (2001) The great transformation: the political and economic origins of our time, 3rd edn.
Beacon, Boston
Putzel J (2002) Politics, the state and the impulse for social protection: the implications of Karl
Polanyi’s ideas for understanding development and crisis, Crisis states working paper 1.
London School of Economics, London
Reis E (2005) Perceptions of poverty and inequality among Brazilian elites. In: Reis E, Moore M
(eds) Elite perceptions of poverty and inequality. Zed Books, London
Reis E (2009) Poverty and inequality in the eyes of the elites. Paper presented at the roles of elites
in economic development conference, Helsinki, 12–13 June 2009
Reis EP, Moore M (2005) Elite perceptions of poverty and inequality. Zed Books, London
Samson M, Lee U, Ndlebe A, MacWuene K, van Niekerk I, Gandhi I, Harigaya T, Abrahams C
(2004) Final report. The social and economic impact of South Africa’s social security system,
EPRI research paper 37. Economic Policy Research Institute, Cape Town
Skoufias E (2005) PROGRESA and its impacts on the welfare of rural households in Mexico.
International Food Policy Research Institute, Washington, DC
Sultan S, Schrofer T (2008) Building support to have targeted social protection interventions for
the poorest—the case of Ghana. Paper presented at the social protection for the poorest in
Africa: learning from experience conference, Kampala, 8–10 September 2008
Chapter 20
Innovative Business Approaches
for the Reduction of Extreme Poverty
and Marginality?

Heike Baumüller, Christine Husmann, and Joachim von Braun

Abstract The potential of innovative business approaches to target the poor is

increasingly being recognized. This chapter outlines the evolution of thinking in
the business world and explores in detail some of the relatively new business
approaches that have emerged for addressing societal problems. The authors also
examine whether and how these approaches can support not only those living close
to the poverty line, but also help engage the marginalized at the lowest end of the
income scale. While it may be unrealistic to expect businesses to be able to reach all
of the extremely poor and marginalized, the authors suggest that the boundaries of
innovative business operations can be pushed much further to include a far larger
number of the marginalized and extremely poor.

Keywords Corporate social responsibility • Social entrepreneurship • Bottom of

the pyramid • Ultra-poverty

20.1 Introduction

Extreme poverty is an immense political and market failure, wasting the potential of
hundreds of millions of people (von Braun 2010). Investing in the creation of
markets that include the extremely poor and marginalized should thus not only be
considered as a charitable activity, but also has the potential to provide high returns
on investments—in both financial and humanitarian terms. Various new business
approaches have begun to fill these investment gaps, some led by well-known
figures such as Muhammed Yunus, Michael Porter, Mark Kramer, and C. K. Prahalad

H. Baumüller (*) • C. Husmann • J. von Braun

Center for Development Research (ZEF), University of Bonn,
Walter Flex Strasse 3, 53113 Bonn, Germany
e-mail: [email protected]; [email protected]; [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 331
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_20,
© The Author(s) 2014
332 H. Baumüller et al.

on the one hand, and innumerable individual social entrepreneurs on the other, who
together have brought the creation of social value into mainstream business thinking.
While the potential of these approaches for targeting the poor that are close to
poverty lines is increasingly being recognized, the question remains: to what extent
can business also help the extremely poor? In his influential book, The Fortune at
the Bottom of the Pyramid, Prahalad (2010, 8) acknowledges that “[t]here is a
segment of the 4 billion who are so destitute, so deprived, and so consumed by war
and disease that they need other forms of help,” such as government subsidies,
multilateral aid, or philanthropy. How far can business approaches push the margin
to also include those who are extremely poor? How can those who are currently
excluded from development opportunities be brought into and benefit from market-
based systems to improve the quality of their lives? These are the questions that we
examined in this chapter.1
What characterizes the extremely poor from the perspective of business? In some
respects their situation is similar to that of the general poor, albeit more severe.
As consumers their purchasing power is extremely low and they often struggle to
follow regular payment schedules due to income fluctuations. While many may
already operate in commercial markets, for instance by selling their time or engaging
in small-scale trading, returns tend to be low due to limited skills and assets. These
constraints, along with the lack of access to credit and additional barriers that may
result from discrimination or spatial remoteness, also restrict their ability to expand
their business activities.
In other respects extreme poverty has particular characteristics, as highlighted by
the extensive work undertaken by BRAC in Bangladesh to engage the ultra-poor.
BRAC notes that the extremely poor are often subject to overlapping sets of con-
straints and deprivations that make them “structurally different from other catego-
ries of the poor; they are not only poorer than others, but differently so” (Matin et al.
2008, 5). Their main priority is survival, with little time or extra resources to invest
in long-term strategies. As a result they get caught in “[a] below-subsistence trap
from which it is difficult for them to break free using available resources and mech-
anisms” (Matin et al. 2008, 2).
The extremely poor are often marginalized, which also prevents them from
realizing their potential (for an overview on marginality and extreme poverty issues
see Gatzweiler et al. 2011). We define marginality according to Gatzweiler et al.
(2011, 3) as:
an involuntary position and condition of an individual or group at the margins of social,
political, economic, ecological and biophysical systems, preventing them from access to
resources, assets, services, restraining freedom of choice, preventing the development of
capabilities, and eventually causing extreme poverty.

1
This study was part of a research project on marginality and extreme poverty at the Center for
Development Research (ZEF). Support by the Bill & Melinda Gates Foundation for this research
is gratefully acknowledged. For further information see www.zef.de/margip.html
20 Innovative Business Approaches for the Reduction of Extreme Poverty… 333

As we know that the large majorities of the marginalized and extremely poor live
in rural areas of the developing world and typically depend directly or indirectly on
agricultural, forestry, and fishery based livelihoods, any commercial efforts to
address extreme poverty must consider these circumstances. Other cross-cutting
aspects of marginality are the adverse health and nutritional conditions of the
extremely poor, as well as their lack of access to education and information.
The increasing interest in business approaches to combat poverty stems from a
long history of debate on the role of businesses in society, some of which traces as
far back as India’s Kautily in the fourth century BC, and also nearly 200 years later
to Cicero in Rome (Blowfield and Frynas 2005). During the Industrial Revolution
entrepreneurs like John Cadbury, Robert Owen, and Léon Harmel took measures
towards solving the problems of feeding, clothing, and employing a great number
of people, trying to improve the working and living conditions of workers and
their families (Boddice 2009). Over the past four decades the business world has
evolved from “command and control” approaches to addressing environmental
and social issues, towards the pro-active application of business strategies to pur-
sue social goals.
Especially in recent years the proposition has been gaining ground that firms and
investments can be profitable and possibly even improve the competitive position of
companies when the creation of social value is integrated as a core business activity
(Porter and Kramer 2011). This marks an important step on the path of the private
sector from “being less bad” towards “being more good” (McDonough and
Braungart 2002). In the meantime the recognition that governments and traditional
development assistance mechanisms have still not solved the alarming problems of
poverty has also been growing. Indeed, much criticism has been leveled at the
perceived failures and inefficiency of many governmental and non-governmental
development programs and official development aid in recent years.
In this chapter we outline this evolution of thinking in the business world and
explore in detail some of the relatively new business approaches that have emerged
for addressing societal problems. Furthermore we specifically examine whether
and how these approaches can support not only the lower-income and near poverty
line classes, but also help bring the marginalized at the lowest end of the income
scale into market-based systems in meaningful ways.

20.2 Business and Social Values

In the past public discourse and research agendas on the social roles of business
have been largely dominated by North American and European perspectives, in
particular through the concept of corporate social responsibility (CSR). However,
the underlying idea of CSR (i.e., that companies have responsibilities towards
society), is not a “northern” phenomenon, even if the terminology is (Frynas 2005).
334 H. Baumüller et al.

Social Value
Mainstream the promotion of
social values for strategic, profit
Focus
and social reasons.
Shared Value
4 billion low-income consumers
‘Bottom of the Social Entrepreneurship
Pyramid’ market & Social Business
Inclusive Application of
Business entrepreneurial
Include the poor on the demand principles to
side as clients and customers and address a social
Corporate Social on the supply side as employees, problem
Responsibility producers and business owners.
Companies integrate social and environmental
Profit concerns in their business operations and in
Focus their interaction with their stakeholders on a
voluntary basis

Fig. 20.1 Business approaches to poverty reduction

Indeed, examples from Africa show that businesses are often expected to assist the
communities in which they operate and that they are perceived to be part of (Frynas
2005; Idemudia 2010).
Recently leading business thinkers have stressed the benefits to companies of
integrating social considerations into core business strategies rather than treating
them as external add-ons, as is common in CSR. For instance, this can be done by
targeting low-income consumers (“bottom of the pyramid” markets) or strengthening
supply and distribution chains through the involvement of local communities as
part of inclusive business strategies. Others—most notably Muhammed Yunus and
other social entrepreneurs who seek to launch new enterprises that directly address
problems of poverty and sustainability—are taking this argument one step further,
advocating the use of business strategies as a primary means to address social goals,
rather than simply for financial gain. Thus in discussions on the role of business
in society, profit maximization as the primary objective of business operations is
increasingly making way for business initiatives that are guided by social objectives,
whether or not they make a profit (see Fig. 20.1 for a summary of business approaches
presented in this section).
Moreover debates on the role of business in society are moving out of “Western-
dominated” circles to the global level, exemplified by the emergence of new proposi-
tions from leading thinkers in the developing world and growing hubs of social
business activity in low-income countries such as India and Bangladesh. This shift has
helped stimulate a more constructive and practical approach to private sector engage-
ment in development, which remains colored by skepticism about businesses’ motives
and commitments to social goals.
20 Innovative Business Approaches for the Reduction of Extreme Poverty… 335

20.2.1 Corporate Social Responsibility

The term “corporate social responsibility” began to be widely used in the 1960s and
1970s, coinciding with the rise of multinational companies. There is no standard
definition of CSR. One of the most commonly cited definitions comes from the
European Commission (EC), which refers to CSR as “a concept whereby companies
integrate social and environmental concerns into their business operations and in
their interaction with their stakeholders on a voluntary basis” (EC 2011, 6).
More generally Blowfield and Frynas (2005, 503) regard CSR as:
an umbrella term for a variety of theories and practices, all of which recognize the
following:
(a) that companies have a responsibility for their impacts on society and the natural
environment, sometimes beyond legal compliance and the liability of individuals;
(b) that companies have a responsibility for the behavior of others with whom they do
business (e.g., within supply chains); and
(c) that businesses needs to manage their relationships with broader society, whether for
reasons of commercial viability or to add value to society.

The terminology and publicized approaches to CSR are largely shaped by Anglo-
Saxon tradition, with its differentiation between economic and social affairs, focus
on individualistic rather than community values, and the limited role of government
in regulating markets (Blowfield and Frynas 2005; Sadler and Lloyd 2009).
Proponents of CSR tend to emphasize the benefits of volunteerism and self-
regulation as more effective means to promote socially responsible corporate
activities. For instance, the Department for International Development (DFID) in
the UK (2003, 9) warns that “international legally-binding frameworks for multi-
national companies may divert attention and energy away from encouraging
corporate social responsibility and towards legal processes.”
The “northern-centered” view of CSR has driven research agendas over recent
decades, which has largely concentrated on firms in high-income countries, notably in
North America and Europe, and the adoption of universal norms such as workers’ rights.
These approaches have also shaped the way that CSR is being applied in developing
countries. As Blowfield and Frynas (2005, 504) note, “by the time empirical studies
started to be commissioned to investigate whether CSR could benefit the poor and mar-
ginalized, certain conventions and orthodoxies had already been established.” Thus
development related CSR emerged largely as a response to growing criticism of multi-
national companies from activists, consumers, and shareholders about poor labor and
environmental standards, and associated impacts on local communities. An initial focus
on codes of conduct and voluntary company-level standards has expanded to include
ethical sourcing, certifications, community development, and stakeholder engagement.
While NGOs were a key driving force behind promoting CSR in the 1990s, they
have become increasingly critical, accusing some companies of using CSR as
“green wash” for unsustainable business practices and/or to avoid regulations. For
instance the UK-based NGO Christian Aid (2004, 14) acknowledges that civil soci-
ety groups, through their support for CSR, “have unwittingly enhanced company
336 H. Baumüller et al.

images and market profiles.” They suggest that instead, “NGOs may be more effec-
tive by throwing their collective weight behind the drive for international regula-
tion than by tying up their scant resources in bilateral dialogues.”
Critics also point to the gap between CSR rhetoric and practice, exacerbated by
the lack of measurement approaches to assess impacts (Utting 2007). CSR activities
are often seen as cosmetic rather than operational or strategic, summarized in reports
that “aggregate anecdotes about uncoordinated initiatives to demonstrate a compa-
ny’s social sensitivity” (Porter and Kramer 2006, 79). The focus, it is claimed, is on
mitigating tensions between companies and society rather than on recognizing their
interdependence and working for their mutual benefit. As a result the “potential of
companies to take actions that would support both their communities and their
business goals” is lost (Porter and Kramer 2006, 81).
The promotion of CSR as a development tool has also attracted criticism.
Specifically it is argued that CSR neglects issues related to the power dynamics and
political economy of the contexts in which the companies operate. In particular
marginalized groups are often left out of CSR activities, thus they are unable to
engage in stakeholder dialogue or express their concerns when their interests do not
coincide with NGO, donor, or company priorities. This can be particularly problematic
where the choice of CSR activities is primarily driven by business considerations.
As Blowfield and Frynas (2005, 508) point out:
[s]ince inclusion in or exclusion from stakeholder status is not based on either legal rights
or moral obligations, a stakeholder’s recognition is contingent upon the business case for
that recognition. Consequently, the well-being of some groups in developing countries may
be jeopardized by the very pursuit of CSR.

Similarly, lacking contextual analysis can result in projects that may promote
business interests rather than social development priorities. For instance Idemudia
(2010, 841) concludes that basing choices of CSR projects on business logic rather
than on local needs has “contributed to the breakdown of traditional institutions and
the proliferation of failed development projects in the Niger Delta.” At the same
time it is argued that CSR may reinforce corporate power by offering companies an
excuse to avoid regulations and strengthening their influence by providing them the
space to lead and shape the CSR movement (Utting 2007). The economic roles and
political influence of multinational corporations, however, is not fundamentally
questioned in the CSR agenda. Critics also note that CSR only addresses some of
the symptoms of poverty, but ignores underlying development challenges. Utting
(2003, 7) stresses that:
if large corporations are to contribute in a meaningful way to social and sustainable devel-
opment, the CSR agenda needs to address the central question of the structural and policy
determinants of underdevelopment, inequality and poverty, and the relationship of [multi-
national corporations] to these determinants.

Even companies that are implementing CSR activities ostensibly for develop-
ment reasons may at the same time lobby for tax reductions, outsource parts of the
value chain to avoid compliance with labor or environmental standards, or destroy
local livelihoods through their business activities (Utting 2007; Idemudia 2010).
20 Innovative Business Approaches for the Reduction of Extreme Poverty… 337

More recently the rhetoric in the CSR debate—and in broader discussions on the
role of business in society—has shifted from mitigating the negative impacts of
business towards the positive role that business can play in advancing development
and poverty reduction. This shift is also being promoted by leading international
development organizations such as the DFID, the Netherlands Development
Organisation (SNV), the World Bank, and the UNDP (Prieto-Carrón et al. 2006). It
is acknowledged that it may be unrealistic to expect too much of CSR. Indeed CSR
“was never conceived as a tool to tackle poverty” (Newell and Frynas 2007, 678).
Rather, recent debates on how to strengthen the role of business in development
have focused on the need for new business strategies where social objectives are a
core element or even the primary goal of business operations.

20.2.2 “Bottom of the Pyramid” Market

The introduction of the “bottom of the pyramid” (BOP) concept by Prahalad and
Hart in 2002 marked a milestone in the debate surrounding the role of business in
promoting social goals and poverty reduction. By highlighting the “fortune” that
lies at the four billion strong bottom of the income pyramid, they succeeded in
attracting the attention of the business world to recognize and investigate the market
opportunities offered by the “billions of aspiring poor who are joining the market
economy for the first time” (Prahalad and Hart 2002, 1). The BOP argument thereby
places the complementarities of social and corporate objectives at the heart of
business strategy.
While the proponents of the BOP market highlight the potential of this group as
both producers and consumers, the business case initially focused largely on the
poor as consumers (Prahalad and Hart 2002; Hammond et al. 2007; Prahalad 2010).
The BOP market is usually defined by income, with an annual per capita income
limit ranging from US$1,500 to 3,260 (Table 20.1).
Hammond et al. (2007) characterize what they refer to as the BOP as people with
significant unmet needs, including access to finance, housing, water, sanitation,
electricity, and health services. BOP populations are largely dependent on informal

Table 20.1 Delineation of the bottom/base of the pyramid market—in US$, income in purchasing
power parity (ppp)
Annual per capita income Number of people
1 2 3 1 2 3
High >$20,000 >$20,000 >$20,000 75–100 100 million No data
income ($21,731) million
Middle $1,500– $2,000– $3,000– 1,500–1,750 2 billion 1.4 billion
income 20,000 20,000 20,000 million
BOP <$1,500 <$2,000 <$3,000 4 billion 4 billion 4 billion
(<$3,260)
1: Prahalad and Hart (2002); 2: Prahalad and Hammond (2002); 3: Hammond et al. (2007)
338 H. Baumüller et al.

or subsistence livelihoods with limited access to markets, and in rural areas are
highly vulnerable to the destruction of natural resources that their livelihoods
depend upon. They are often subject to a “BOP penalty” in the form of higher prices
and lower quality of goods and services than more affluent consumers due to living
in remote areas and limited purchasing power (see also Gradl and Knobloch 2010).
Elaborating on the nature and potential of the BOP market, Prahalad (2010) argues
that although individual consumers may have low purchasing power, the market is
nevertheless lucrative due to its sheer scale. Contrary to widespread assumptions,
BOP customers are ready to adopt new technologies and are brand conscious—and
thus are also potentially attractive to multinational companies. The use of local
distribution systems and modern communication technologies has greatly facilitated
access to BOP customers. Capitalizing on this market will require companies to
adapt their business strategies to the needs of BOP customers, for instance by creating
the capacity to consume (e.g., through changes in packaging, pricing, or payment
schedules) and by developing new goods and services.
Assessments of the size of the BOP market vary. Prahalad (2010) estimated the
global BOP purchasing power to amount to as much as US$13 trillion annually. A
detailed assessment carried out by the International Finance Corporation (IFC) and
the World Resources Institute (WRI) valued the global BOP market at US$5 trillion
annually (Hammond et al. 2007). The World Economic Forum put the annual income
of the BOP population at US$2.3 trillion, and predicted that this income could
increase to US$4 trillion by 2015 (WEF 2009). Among the four billion, the largest
segment (1.6 billion) was found to be located in the middle tier, earning around
US$1.00–2.50/day, while one billion people earned less than US$1/day, and 1.1 bil-
lion earned US$2–8/day (WEF 2009). In comparison the 1.4 billion strong mid-level
income market is estimated to be worth US$12.5 trillion (Hammond et al. 2007).
Hammond et al. (2007) break down the BOP market geographically and by
sector. They concluded that the largest segment of the market is found in Asia
with 2.86 billion people and an income of US$3.47 trillion, representing 82 % of
the region’s population and 42 % of the purchasing power. The region with the
highest share of the general population in the BOP market is Africa, where 95 %
of the people (486 million) have an estimated combined income of US$429
million. In Latin America the BOP market consists of 360 million people with
an income of US$509 billion, representing less than a third of the regional
household income (28 %).
In terms of sectors, Hammond et al. (2007) identify food as the largest market by
far, valued at US$2.9 trillion. Also important are energy (US$433 billion), housing
(US$332 billion), transportation (US$179 billion), and health care (US$158 billion).
The information and communication technologies sector were estimated at around
US$51 billion, which the authors acknowledge could in reality be as much as twice
that amount given rapid growth. The water market sector is estimated at US$20
billion. BOP markets for water, information and communication technologies, and
housing are predominantly urban in all regions, while food and health care markets
are predominantly rural in most Asian and African countries. In Asia the dominant
rural market sectors were energy and transportation.
20 Innovative Business Approaches for the Reduction of Extreme Poverty… 339

Karnani (2009) on the other hand, contends that claims about the BOP market
are wildly exaggerated. He maintains that most of the successful examples of selling
to the BOP market are in fact targeting the middle class rather than the poor. He
argues that the size of the market is overestimated because of the high per capita
income cut-off of US$3,000/year (US$8/day) used by the IFC/WRI study, which is
considerably higher than the commonly used poverty line. He notes that “[a]ccording
to this report, 98.6 % of the population in India is in the bottom of the pyramid,”
(Karnani 2009, 7). Using US$1,000 annually per capita as the poverty threshold, he
estimates the BOP market to amount to US$1.2 trillion at purchasing power parity
(PPP), and just US$0.36 trillion at market exchange rates.
For Prahalad (2010) the exact delineation of the BOP market and even the termi-
nology used to describe it—be it “base of the pyramid,” the “next billion,” or the
“bottom billion”—are secondary. Reflecting on the debates sparked by the intro-
duction of the concept, he acknowledges that “there is no single universal definition
of the Bottom of the Pyramid that can be useful.” The fact remains that it is widely
recognized that “four billion micro consumers and micro producers constitute a
significant market and represent an engine of innovation, vitality, and growth,” and
companies can choose to serve any segment of this market (Prahalad 2010, 7).
A more fundamental criticism of the BOP approach, however, is whether access
to consumer goods and services will necessarily bring developmental benefits.
Karnani (2009) argues that it is empirically false to view of the poor as “resilient and
creative entrepreneurs and value-conscious consumers” (Prahalad 2010, 25).
Karnani cites research by Banerjee and Duflo (2007) showing that the poor, just like
any consumer, do not necessarily make purchasing decisions that are most beneficial
for the health and food security of the individual or household. He points out the
case of Unilever’s Fair and Lovely skin-whitening cream—an example cited to
highlight the success in servicing BOP markets (Prahalad 2010)—as a product of
dubious value to the well-being of poor consumers and which he claims in fact further
sustains racist and sexist prejudices. Thus Karnani (2009, 4) concludes that “there
is a need to impose some limits on free markets to prevent exploitation of the poor.”
In response Prahalad (2006, 2) maintains that “consumption can and does
increase income.” He also notes that it is ultimately up to the poor to choose what to
spend their money on. As already noted above, he does acknowledge that access to
markets will not provide a solution for all, and that the poorest will continue to rely
on outside help from governments and donors. Even for this segment though, the
goal would be to “build capacity for people to escape poverty and deprivation
through self-sustaining market-based systems” (Prahalad 2010, 8).

20.2.3 Inclusive Business

Another recent incarnation of the BOP approach is “inclusive business,” or “inclusive

markets,” that emphasize the benefits of engaging the poor along the entire supply
chain. Inclusive business models can include the poor on the demand side as clients
340 H. Baumüller et al.

and customers, and also on the supply side as employees, producers, and business
owners.2 Among the main advocates of this approach is the UNDP, which—as a
follow up to the 2004 report of the Commission on the Private Sector and
Development—set up the Growing Inclusive Markets Initiative in 2006 with the
aim of better understanding, “how the private sector can contribute to human
development and to the Millennium Development Goals” (UNDP 2008, v).
The UNDP argues that engaging the poor can be beneficial for both businesses
and poverty alleviation. Business can profit through (potentially) high rates of
return, opening up new markets, innovation driven by the challenges of developing
inclusive markets, expanding labor pools, and strengthening supply chains through
local sourcing. The poor in turn will benefit through greater access to essential
goods and services, income generation, and empowerment. The UNDP acknow-
ledges that inclusive markets also pose serious challenges that need to be addressed,
such as limited market information, ineffective regulatory environments, inadequate
physical infrastructure, a general lack of knowledge and skills, and restricted access
to financial products and services.
Inclusive business has also found strong support in the business community and
among some donors. The World Business Council on Sustainable Development
(WBCSD), in an alliance with the SNV, is promoting inclusive business models that
are both profitable and that have a clear development benefit for the low-income
segment of the population (SNV/WBCSD 2008). The IFC, which explicitly links
the BOP and inclusive business approaches, committed US$780 million to more
than 35 clients with inclusive business models in 2009, while more than 150 active
clients were in its portfolio (Jenkins et al. 2010).
An assessment of 14 inclusive business models published by the IFC and Harvard
University’s Kennedy School of Government highlights a number of common
themes (Jenkins and Ishikawa 2010). This study found that expected growth was the
main driver for businesses to develop such models. The most common outcome for
businesses has been revenue growth, while the main development outcomes for the
poor have been expanded economic opportunities (as suppliers, distributors, or
retailers), and improved access to goods and services. Factors contributing to their
success have included: network and technology platforms that reach low-income
consumers; financing schedules that match the cash flows of poor individuals and
households; capacity building among suppliers, consumers, and distributors; and
partnerships with other companies, governments, or finance institutions.

20.2.4 Creating Shared Value

In their discussion of the concept of “Shared Value,” Porter and Kramer (2006, 2011)
take the argument for the private sector’s role in advancing human development and

2
See UNDP (2008, 2), SNV/WBCSD (2008), Gradl and Knobloch (2010), Jenkins et al. (2010, 2),
and also http://www.inclusivebusiness.org for definitions of inclusive business/markets.
20 Innovative Business Approaches for the Reduction of Extreme Poverty… 341

poverty alleviation one step further by arguing that it is in fact in the interest of all
businesses to be promoting social values—for strategic, economic, and social reasons.
Thus rather than creating special “social” entities that address issues such as poverty
or environmental damage, they call on companies to bring business and society
back together by addressing societal and environmental concerns related to their
products, and designing production processes that benefit workers and their families.
They argue that it is not a matter of altruism to commit to this change, but a strategic
advantage that pays off in the mid- to long-term.
The idea of factoring social and environmental objectives into business strategies
for commercial reasons—to increase labor productivity, to ensure the stability and
quality of production inputs, or to respond to consumer demand—is not new. As far
back as the eighteenth century some of the early industrial entrepreneurs promoted
improved working conditions to enhance labor productivity. More recent examples
include company programs to address HIV/AIDS awareness and health care, improve
smallholder productivity, or sell ethics-based certified products to affluent markets.
What sets Porter and Kramer’s approach aside from these initiatives is the call for
a fundamental rethinking of business strategy, rather than piecemeal activities for a
limited part of business operations or public relations purposes.
“Creating Shared Value” has been most publicly embraced by Nestlé (2012;
Christiansen, Chap. 21 this volume), recognizing that “in order to be successful, a
company must create value for its shareholders and at the same time for the com-
munities where it operates and for society at large.” Nestlé notes that this approach
goes beyond ensuring compliance and sustainability to consider how sound business
principles can create value for both shareholders and for society in the long-term. In
2009 the company set up a high-level Nestlé Creating Shared Value Advisory Board
to increase the company’s positive impact on society. The company recognizes that
implementation of the approach is bound to face challenges including: how to put
the concept into practice, how to measure environmental and social impacts, and
how to positively influence the value chain.

20.2.5 Social Entrepreneurship

While Porter and Kramer stress the complementarity of social and corporate
objectives, profit making remains the primary goal of business, which can support
the pursuit of social objectives. In contrast a social entrepreneur uses entrepreneurial
principles to organize, create, and manage a venture with the primary aim of bringing
about social change. Unlike a business entrepreneur who typically measures perfor-
mance in terms of (long-term) profit and returns, a social entrepreneur measures
success in terms of progress made towards the creation of social value (Dees 1998;
Thompson 2002; Haugh 2006; Nicholls 2006).
While the language of social entrepreneurship may be new, the phenomenon is
not, dating back to the eighteenth and nineteenth centuries (Dees 1998; Boddice
2009). According to Nicholls (2006, 7) the term “social entrepreneur” was coined
342 H. Baumüller et al.

by Banks in 1972 in reference to Robert Owen. While there is no commonly agreed

upon definition of social entrepreneurship, most authors agree that the three defining
features of social entrepreneurship are sociality, innovation, and market orientation
(Nicholls 2006). The explicit and central social mission of social entrepreneurs is
reflected in the special emphasis on projects designed to improve the quality of life
of humans: health, nutrition, education, training, and the creation of stable and
productive jobs (Bornstein 2007).
Opinions differ, however, as to what extent social entrepreneurs are necessarily
engaged in business activities. For instance Ashoka founder Bill Drayton contends
that social entrepreneurs are “change makers,” but not necessarily business people.
According to Drayton people like Florence Nightingale, who revolutionized health
care, are good historical examples of social entrepreneurs. They were committed to
creating change in their societies and employed entrepreneurial principles, but they
did not see themselves as business operators (Lamb 2011).
Observers also disagree over whether social entrepreneurship refers only to non-
profit or also to profit-seeking enterprises. While some authors require the primacy
of social targets over all other objectives, many so-called hybrid organizations also
exist that operate with a double or triple bottom line (i.e., aiming at social, financial,
and environmental returns on investment) (Dees 1998; Guclu et al. 2002; Mair and
Martí 2006; Nicholls 2006; Bornstein 2007; Seelos and Mair 2007; Certo and Miller
2008). The idea of a triple bottom line was introduced by Spreckley (1981), who
proposed to add criteria other than financial performance to measure the success of
enterprises. The term was then promoted by John Elkington (1997) in his book
Cannibals with Forks.
In the absence of a common definition, an exact number of existing social enter-
prises is difficult to find. A recent report counted 55,000 social enterprises in the UK
alone in 2006 (Brown and Campanale 2006). Ashoka, one of the pioneers in sup-
porting social entrepreneurs, has 2,500 “fellows” worldwide.3 The organization,
which was set up in 1980 by Bill Drayton, brings communities of social entrepre-
neurs together to help leverage their impacts, to broaden the scale of their ideas, and
to capture and disseminate examples of best practices. To this end Ashoka selects
social entrepreneurs as “Ashoka Fellows” and provides them with living stipends for
an average of three years, professional support, and access to a global network of
peers in 70 countries. The organization started with an annual budget of US$50,000
and grew to nearly US$35 million by 2008, and is now active in 60 countries.
Several organizations have followed the approach pioneered by Ashoka. Two
well-known examples are the Schwab Foundation for Social Entrepreneurship and
the Skoll Foundation. Launched in 1998 by Klaus Schwab, founder of the World
Economic Forum, and his wife Hilde, the Schwab Foundation promotes “entrepre-
neurial solutions and social commitment with a clear impact at the grassroots level.”4
The foundation’s activities focus on supporting social entrepreneurs strategically by

3
www.ashoka.org
4
www.schwabfound.org
20 Innovative Business Approaches for the Reduction of Extreme Poverty… 343

providing 20–25 “Social Entrepreneur of the Year” awards to recipients who

benefit from networking within and outside the foundation. Through the “Schwab
Foundation community” the organization fosters exchange among social entrepre-
neurs and supports the replication of their methods among one another. The founda-
tion does not provide grants or other financial support to the organizations of its
selected social entrepreneurs.
In addition to providing financing to social funds, the Skoll foundation founded
the Skoll Centre for Social Entrepreneurship at the Saïd Business School of
Oxford University in 2003. The foundation sponsors the Skoll Award for Social
Entrepreneurship, which includes monetary and other elements. The foundation
also hosts the annual Skoll World Forum on Social Entrepreneurship, an interna-
tional high-level platform created to accelerate “entrepreneurial approaches and
innovative solutions to the world’s most pressing social issues” (Skoll World
Forum 2012).

20.2.6 Social Business

The recent interest in social businesses can in part be attributed to the great success
of microfinance operations in the last decades. The concept was mainly pioneered
by Muhammed Yunus, the founder of Grameen Bank in Bangladesh. According to
Yunus (2007) the creation of social value is the main purpose of business activities
for a “social business,” not just a complement to profit creation. The distinction
between social entrepreneurship and social business is not entirely clear. Connota-
tions tend toward calling non-profit enterprises “social enterprises” and for-profit
enterprises “social businesses,” however, there is no agreement on this distinction in
the literature.
According to Yunus’ seven principles of social business (Grameen Creative Lab
2012), the business objective is to “overcome poverty, or one or more problems,
which threaten people and society; not profit maximization.” Social businesses
operate on a “no loss, no dividend” basis, (i.e., they need to be financially sustain-
able and investors only recuperate their initial investment). All financial surpluses
after breaking even are used for the expansion and improvement of the company’s
activities.
These principles are the outcome of a long history of experimentation and
improvement. Starting in the 1970s making small loans to poor borrowers, Yunus
has launched various businesses that provide goods and services that he considered
useful for improving the lives of the poor at prices they could afford. Today the
“Grameen family” consists of 14 social businesses including telecommunications,
education, and garment production through joint ventures with big companies such
as Nestlé, the chemical company BASF, and Telenor.5

5
See www.grameen-info.org
344 H. Baumüller et al.

Operating a business for social objectives rather than monetary returns appears
to fly in the face of conventional capitalistic assumptions that humans by and large
are rational and self-interested individuals (Homo oeconomicus). Yunus (2007, 18)
argues that capitalism’s assumption that people act so as to maximize profit fails “to
capture the essence of what it is to be human.” He contends that it is the multidimen-
sional nature of people that motivates some to pursue social goals rather than only
to maximize personal gain.
Yunus identifies two types of social business (although they may overlap within
a single business operation). One focuses on providing a product and/or service with
a specific social, ethical, or environmental goal. An example of this type is Grameen
Danone, which produces a special yoghurt product with high nutritional value for
children in Bangladesh. The other type is a profit-oriented business that is owned by
the poor or other underprivileged parts of society. The Grameen Bank is owned by
the poor, however, through its activities it also classifies as an example of the former
“specific goal” type of social business (Yunus 2007).
Despite certain challenges, social entrepreneurship and social business show
promising features for reducing extreme poverty and marginality. Since the social
mission is the central task of such ventures, going the extra mile to serve the most
deprived might be more appealing for such entrepreneurs, since the social returns on
their investments are especially high among those people. On the other hand social
enterprises face other constraints when trying to reach out to the extremely poor and
marginalized, such as difficulties with accessing capital.

20.3 Potentials and Constraints of Innovative Business

Approaches in Addressing Extreme Poverty
and Marginality

New and evolving business approaches have shown significant promise in reaching
people with low incomes. How suitable are these different approaches for engaging
the extremely poor and marginalized? The answer depends in part on: (1) the extent
to which the different approaches are able to involve the extremely poor, (2) their
flexibility in directing business objectives towards the reduction of extreme poverty
and marginality, and (3) their ability to successfully operate with non-commercial
public and civil society partners, and in sectors of particular interest to the extremely
poor (see Table 20.2).

20.3.1 Willingness to Include the Extremely Poor as

Producers, Employees, and Consumers

The extremely poor and marginalized are by definition, excluded from many eco-
nomic and social activities. Access to land and natural resources is a case in point.
20 Innovative Business Approaches for the Reduction of Extreme Poverty… 345

Table 20.2 Suitability of different business approaches for engaging the extremely poor
Corporate social Bottom of the pyramid/ Social
responsibility inclusive business entrepreneurs
Inclusion of the extremely poor Low Medium High
Ability to mobilize capital Low High Medium to high
for social goals
Organizational flexibility Low Varied High
Engagement in sectors that Low Medium High
matter most to the poorest

Thus unless businesses make a dedicated effort to engage them, the extremely poor
are unlikely to benefit from corporate activities. A precondition of successful
engagement is a deep knowledge of the contextual situation in which the extremely
poor and marginalized live and operate. In rural areas this must often include
detailed knowledge about local agricultural operations.
In many cases businesses cannot rely on governments to facilitate the participa-
tion of the poorest, as the marginalized may be actively discriminated against by
their national or local governments (e.g., for ethnic or religious reasons), or may
simply not have been recognized as needing special attention. As the United Nations
Children’s Fund (UNICEF 2010, 5) notes:
The poorest and most marginalized communities are not systematically assessed and are
often forgotten when national development plans are laid and resources allocated. They are
also the least likely to have a voice in global and national decision-making forums.
Disaggregating national data to identify these groups and assess the factors that exclude
them is fundamental to designing equitable solutions.

As already noted above, CSR activities often fail to include the extremely poor
and marginalized, unless designed as charitable side activities. Moreover CSR is
likely to favor small-scale projects with high visibility and short-term impacts,
rather than the kinds of long-term projects that are necessary to reduce extreme
poverty and marginality. Scaling-up has rarely been an objective in CSR activities.
In the case of inclusive business approaches (or BOP markets), companies may
prefer to target those living on US$3–4/day as producers and consumers. As Gradl
and Knobloch (2010, 15) point out, “[i]t is often easier to concentrate on groups that
have capital; be it knowledge, land or social inclusion, but that are unable to benefit
from it because of inadequate market access.” For instance, in the case of contract
farming companies, they tend to prefer working with large-scale farmers to reduce
transaction costs and because they expect higher outputs (Fullbrook 2007). The
willingness of inclusive business operations to invest in bringing the extremely poor
into supply chains is likely to be shaped by their business interests, such as the need
to expand the supply base beyond poor farmers or to engage the extremely poor in
distribution networks in areas or for population segments that would otherwise be
difficult to reach.
Among social entrepreneurs for whom social value creation is the main objec-
tive, engaging the very poorest is more attractive, as it is likely to generate the
346 H. Baumüller et al.

greatest social impact returns. A recent study by the UNICEF (2010, 1) concluded
that reaching the most deprived “will be [a] considerably more cost-effective and
sustainable” means of reducing child mortality and improving maternal health.
The study estimated that such an “equity-focused” approach targeting those most
in need—and thereby closing existing health gaps—could avert 60 % more deaths
for every additional US$1 million invested. To maximize social returns, social
businesses may therefore be more likely to actively seek out the extremely
poor and support their participation through communication tools and capacity
building activities.

20.3.2 Flexibility to Engage in Socially-Oriented

Business Activities

The suitability of the different business approaches to pursue activities that aim at
engaging those at the lowest end of the income scale will depend on their ability to
mobilize capital for social activities and the flexibility of organizational structures
to adapt to the particular needs of the extremely poor. In the case of CSR, the pri-
macy of corporate objectives tends to restrict companies’ financial flexibility to
divert internal resources away from competing priorities. Also, the organizational
set-up of established companies is generally rigid, with little openness towards
experimentation and adaptation to meet the needs of the extremely poor.
In contrast, inclusive businesses are in a comparatively good position to mobilize
capital, as their dual objectives open up a wider range of potential funding sources,
including traditional investors as well as impact investors and social funds. The
organizational forms of inclusive businesses can differ widely, with some being
more flexible than others. Inclusive businesses can range from large multinationals
like the Coca-Cola Company or Unilever, that apply such models to some parts of
their operations (e.g., to ensure stable supplies or to access new markets), to
smaller-scale enterprises specifically set up as inclusive businesses. Given the
evolving nature of inclusive business models, there is still considerable room for
experimentation and lesson-learning to identify models that could best reach the
extremely poor and marginalized.
In terms of financial capital, social entrepreneurs mainly rely on funding
sources that are willing to invest in relatively risky enterprises with little or no
assurance of financial profit. While access to capital remains a constraint for
social entrepreneurs, the current expansion in social funds, a growing interest in
market-based approaches among some donors, and ongoing efforts to link social
businesses and investors (e.g., through social stock exchanges) are likely to open
up additional financial resources. Social businesses also tend to be smaller and
have more institutional flexibility than conventional businesses, and are able to set
up organizational structures that are specifically designed for engaging target
populations.
20 Innovative Business Approaches for the Reduction of Extreme Poverty… 347

20.3.3 Focusing on Sectors That Matter Most for the Poorest

Businesses have the greatest potential to benefit the extremely poor if they get
involved in sectors that matter most to the poorest, such as food, health, education,
and low-cost infrastructure, including information and communication services.
These sectors, however, often do not yield high returns, particularly if the aim is to
extend coverage into remote areas and/or to customers with very little purchasing
power. Thus commercial approaches where social returns have an equal or higher
weight than financial profit, as in the case of social entrepreneurs and certain
inclusive businesses models, may be better suited to serve the poorest consumers in
these sectors.
Greater involvement of enterprises in these sectors may require a shift in the
respective roles of private and public investors. Businesses may take over the
provision of certain goods and services that have traditionally been the domain of
governments or development organizations, while the public sector focuses on
facilitating business activities to reach the poorest through financial incentives,
streamlined administration, market guarantees or transportation infrastructure, etc.
Governments would need to ensure that private investments do not entirely crowd out
public investments, and thereby incur the risk that some sectors become neglected
or that the associated goods and services remain out of the reach of the extremely
poor due to high prices or spatial distribution limitations. To minimize these risks,
governments may need to consider regulatory measures or implement complemen-
tary activities to engage those businesses that are currently out of reach.

20.4 Conclusions for the Research and Action Agenda

A major challenge in assessing the suitability of business approaches for the reduc-
tion of extreme poverty and marginality, and for adapting them where needed, is the
lack of standardized and generally accepted methods for measuring social value
creation. Unlike monetary returns, “social value” cannot easily be reduced into a
single or easily quantifiable measure. Various proposals have been made, but a
unified framework is lacking. A key question is how to measure not only outputs
(such as micro-credit volumes or the number of products sold to the poor), but also
whether these outputs have indeed translated into real and sustainable poverty
reduction. Thus further research is needed to identify outcome-focused indicators of
social value impacts and to develop suitable measurement tools to provide compa-
rable assessments of the effectiveness of businesses in reducing extreme poverty
and marginality.
While the need for government support to facilitate business solutions to address
extreme poverty and marginality is clear, the forms that this support should take and
the actual benefits of various support measures remain poorly understood. At times
governments may provide support measures that could in fact reduce social welfare
348 H. Baumüller et al.

overall. The example of fiscal incentives is a case in point; while tax breaks are a
popular tool used by governments to try to attract foreign direct investments into
certain areas or sectors, studies in several countries have shown that these incentives
had little influence on investors’ decisions, while governments lost important
revenue that could have been invested elsewhere (Thomas 2007; Baumüller 2009).
Thus research should explore options for government measures to support business
activities for social development and assess their potential impacts in the different
contexts in which they are applied.
Further research investment is also warranted to address opportunities for scal-
ing-up, which is apparently promising for small-scale projects. Given the fact that
households and communities of marginalized and extremely poor people are often
spatially scattered, this is a particular challenge. A framework for scaling-up initia-
tives that target the extremely poor should not just start with a service or product,
but from case-specific situations themselves.
In this context indirect approaches for inclusive business and shared value gen-
eration that reach the extremely poor may be of particular significance, such as
infrastructure investments in marginal areas, access to improved seeds, health
services, and nutritionally enhanced foods that have elements of comprehensive
coverage and do not exclude the poor. Tracing the results and impacts of such
investments on the extremely poor should, however, be part of such initiatives.
It is important to note that the distinction between the different business
approaches is not as clear-cut as this analysis suggests. While our review has pointed
out the basic pros and cons of the different approaches from a perspective of their
potential contributions to the extremely poor and marginalized, new business
approaches to include the poor may also be viewed along a continuum. The scope
of the continuum and overlap between these approaches can extend their reach and
effectiveness through collaborative efforts among different approaches and the
formation of hybrid models that combine different elements. For instance, business
activities targeting the extremely poor may start as CSR projects or pilot business
models by existing companies, and be subsequently scaled-up or outsourced through
dedicated inclusive or social businesses.
We recognize that it is unrealistic to expect businesses to be able to reach all of
the extremely poor and marginalized. There are limits on what any individual or
type of business can achieve to overcome extreme poverty and marginality, however,
these limits are often context specific and should not be viewed as categorically
prohibitive. The entrepreneurial capacities of the marginalized and extremely poor
tend to be underestimated. Most of the approaches reviewed here are relatively new
and still need to be evaluated in terms of how they could best help reduce extreme
poverty and whether they are replicable or can be scaled-up. Moreover none of the
presented business approaches will succeed by themselves, but will require equally
innovative cooperation with public authorities, development organizations, and
above all, the extremely poor themselves.
In summary it seems that the boundaries of innovative business operations can be
pushed much further to include a far larger number of the marginalized and
extremely poor. The corporate sector should be encouraged to look into these
20 Innovative Business Approaches for the Reduction of Extreme Poverty… 349

opportunities as potentially low risk/high return ventures and continue experiment-

ing. Development partner communities may best serve these initiatives by providing
their own insights gained and co-funding. The research community may best serve
these efforts by considering innovative ideas that foster institutional arrangements
that bring together unusual alliances, by accompanying solid impact studies, and
with insights from comparative assessments of cases of scaling-up.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Banerjee B, Duflo E (2007) The economic lives of the poor. J Econ Perspect 21(1):141–167
Baumüller H (2009) Competing for business: sustainable development impacts of investment
incentives in Southeast Asia. International Institute for Sustainable Development, Winnipeg
Blowfield M, Frynas JG (2005) Setting new agendas: critical perspectives on corporate social
responsibility in the developing world. Int Aff 81(3):499–513
Boddice R (2009) Forgotten antecedents: entrepreneurship, ideology and history. In: Ziegler R (ed)
An introduction to social entrepreneurship: voices, preconditions, contexts. Edward Elgar
Publishing, Cheltenham/Northampton, pp 133–153
Bornstein D (2007) How to change the world: social entrepreneurs and the power of new ideas.
Oxford University Press, New York
Brown J, Campanale M (2006) Developing a social equity capital market. New Economics
Foundation and Committed to Effective Giving, London
Certo ST, Miller T (2008) Social entrepreneurship: key issues and concepts. Bus Horiz
51(4):267–271
Christian Aid (2004) Behind the mask: the real face of corporate social responsibility. Christian
Aid, London
Dees JG (1998) The meaning of ‘social entrepreneurship’—comments and suggestions contri-
buted from the Social Entrepreneurship Funders Working Group. The Fuqua School of
Business, Duke University, Durham
DFID (2003) DFID and corporate social responsibility. Department for International Development,
London
EC (2011) Green paper on promoting a European framework for corporate social responsibility.
European Commission, Brussels
Elkington J (1997) Cannibals with forks: triple bottom line of 21st century business. Capstone
Publishing, Mankato
Frynas JG (2005) The false developmental promise of corporate social responsibility: evidence
from multinational oil companies. Int Aff 81(3):581–598
Fullbrook D (2007) Contract farming in Lao PDR: cases and questions. Laos Extension for
Agriculture Project, Vientiane
Gatzweiler F, Baumüller H, von Braun J, Ladenburger C (2011) Marginality: addressing the root
causes of extreme poverty, ZEF working paper no 77. Center for Development Research,
University of Bonn, Bonn
Gradl C, Knobloch C (2010) Inclusive business guide: how to develop business and fight poverty.
Endeva, Berlin
Grameen Creative Lab (2012) The 7 principles of social business. http://www.grameencreativelab.
com/a-concept-to-eradicate-poverty/7-principles.html. Accessed 12 Aug 2012
350 H. Baumüller et al.

Guclu A, Dees JG, Anderson BB (2002) The process of social entrepreneurship: creating opportu-
nities worthy of serious pursuit. Center for Advancement of Social Entrepreneurship, The
Fuqua School of Business, Duke University, Durham
Hammond AL, Kramer WJ, Tran J, Katz R, Walker C (2007) The next 4 billion: market size and
business size at the base of the pyramid. World Resources Institute and International Finance
Corporation, Washington, DC
Haugh H (2006) Social enterprise: beyond economic outcomes and individual returns. In: Mair J,
Robinson J, Hockerts K (eds) Social entrepreneurship. Palgrave Macmillan, New York, pp
180–205
Idemudia U (2010) Rethinking the role of corporate social responsibility in the Nigerian oil conflict:
the limits of CSR. J Int Dev 22(7):833–845
Jenkins B, Ishikawa E (2010) Scaling up inclusive business: advancing the knowledge and action
agenda. International Finance Corporation/Harvard John F. Kennedy School of Government,
Cambridge/Washington, DC
Jenkins B, Ishikawa E, Geaneotes A, Paul J (2010) Inclusive business solutions: expanding
opportunity and access at the base of the pyramid. International Finance Corporation,
Washington, DC
Karnani A (2009) The bottom of the pyramid strategy for reducing poverty: a failed promise.
United Nations Department of Economic and Social Affairs, New York
Lamb GM (2011) Bill Drayton sees a world where ‘everyone is a changemaker.’ Christian Science
Monitor, 16 May 2011
Mair J, Martí I (2006) Social entrepreneurship research: a source of explanation, prediction, and
delight. J World Bus 41(1):36–44
Matin I, Sulaiman M, Rabbani M (2008) Crafting a graduation pathway for the ultra poor: lessons
and evidence from a BRAC programme. Bangladesh Rural Advancement Committee, Dhaka
McDonough W, Braungart M (2002) Cradle to cradle: remaking the way we make things, 1st edn.
North Point Press, New York
Nestlé (2012) A message from our Chairman and CEO. http://www.nestle.com/csv/Nestle/
messagechairman/Pages/messageChairman.aspx. Accessed 12 Aug 2012
Newell P, Frynas JG (2007) Beyond CSR? Business, poverty and social justice: an introduction.
Third World Q 28(4):669
Nicholls A (2006) Social entrepreneurship: new models of sustainable social change. Oxford
University Press, Oxford
Porter ME, Kramer MR (2006) Strategy and society: the link between competitive advantage and
corporate social responsibility. Harv Bus Rev 84(2):78–92
Porter ME, Kramer MR (2011) The big idea: creating shared value. Harvard Business Review,
January/February, pp 62–77
Prahalad CK (2006) C. K. Prahalad responds to ‘mirage at the bottom of the pyramid.’ http://www.
nextbillion.net/archive/multimedia/2006/08/31/c-k-prahalad-responds-to-mirage-at-the-
bottom-of-the-pyramid. Accessed 19 May 2011
Prahalad CK (2010) The fortune at the bottom of the pyramid: eradicating poverty through profits
(revised and updated 5th anniversary edition). Wharton School Publishing, Upper Saddle River
Prahalad CK, Hammond A (2002) Serving the world’s poor, profitably. Harv Bus Rev 80(9):48–59
Prahalad CK, Hart SL (2002) The fortune at the bottom of the pyramid. Strategy + Bus 26:54–67
Prieto-Carrón M, Lund-Thomsen P, Chan A, Muro A, Bhushan C (2006) Critical perspectives on
CSR and development: what we know, what we don’t know, and what we need to know. Int Aff
82(5):977–987
Sadler D, Lloyd S (2009) Neo-liberalising corporate social responsibility: a political economy of
corporate citizenship. Geoforum 40(4):613–622
Seelos C, Mair J (2007) Profitable business models and market creation in the context of deep
poverty: a strategy view. Acad Manag Perspect 21(4):49–63
Skoll World Forum (2012) About the Forum. http://skollworldforum.org/about. Accessed 11 Aug
2012
20 Innovative Business Approaches for the Reduction of Extreme Poverty… 351

SNV/WBCSD (2008) Inclusive business: profitable business for successful development. The
Netherlands Development Organisation and World Business Council on Sustainable
Development, The Hague/Geneva
Spreckley F (1981) Social audit. A management tool for co-operative working. Beechwood
College, Leeds
Thomas KP (2007) Investment incentives: growing use, uncertain benefits, uneven controls. Global
Subsidies Initiative, International Institute for Sustainable Development, Winnipeg
Thompson J (2002) The world of the social entrepreneur. Int J Public Sect Manag 15(5):412–431
UNDP United Nations Development Programme (2008) Creating value for all—growing inclusive
markets. United Nations Development Programme, New York
UNICEF (2010) Narrowing the gaps to meet the goals: an equity-focused approach to child sur-
vival and development. United Nations Children’s Fund, New York
Utting P (2003) CSR and development: is a new agenda needed? In: Summaries of presentations
made at the UNRISD conference on ‘corporate social responsibility and development: towards
a new agenda?’ Geneva, Switzerland, 17–18 November 2003. United Nations Research
Institute for Social Development, Geneva, pp 6–9
Utting P (2007) CSR and equality. Third World Q 28(4):697–712
von Braun J (2010) Extreme Armut—eine Investitionschance, KfW Meinungsforum
Entwicklungspolitik. kfw Entwicklungsbank, Frankfurt
WEF (2009) The next billions: unleashing business potential in untapped markets. World Economic
Forum, Geneva
Yunus M (2007) Creating a world without poverty: social business and the future of capitalism.
Public Affairs, New York
Chapter 21
Business Initiatives That Overcome Rural
Poverty and Marginality Through Creating
Shared Value

Niels Christiansen

Abstract This chapter examines the Creating Shared Value (CSV) approach to
reach the poor through integrated social and business goals. CSV simply means that
when making business decisions on future plans and investments, companies simul-
taneously consider what long-term value can be created both for society and for
shareholders. The chapter describes how the approach has been applied in Nestlé’s
international dairy programs and identifies three major results: including more
small farmers in supply chains, reducing their poverty, and increasing the local
availability of dietary calories, protein, calcium, and various micronutrients.

Keywords Private sector • Shared value • Poverty reduction • Dairy sector

21.1 Introduction

Addressing world poverty and hunger must begin with world’s 1.2 billion subsis-
tence farmers, many of whom are economically, socially, and/or geographically
marginalized. Unfortunately global economic winds are blowing against this imper-
ative for poverty reduction, resulting from the world-wide financial turmoil that
began in 2008, the continued global economic problems, and increasing food prices.
It is estimated that in 2011, 64 million additional persons were added to the ranks of
extreme poverty, defined as living daily on US$1.25 or less per capita. While much
progress has been made in recent years in reducing both urban and rural poverty,
particularly in India, China, and other key countries, at this point in time the world
faces a reversal of the progress that has been made over the past two decades in
reducing poverty and malnutrition.

N. Christiansen (*)
Creating Shared Value Advisors, Lutry, Switzerland
e-mail: [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 353
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_21,
© The Author(s) 2014
354 N. Christiansen

In this economic context and after disappointing results from many well-intentioned
but unsuccessful development projects based on donations, leaders are searching for
new models of development that are economically sustainable, have sufficient scale
to make a difference for large numbers of people, and that are based on economic
foundations that are both self-sustaining and allow growth. One approach that is
gaining interest involving business is called “Creating Shared Value” (CSV). CSV
simply means that in making business decisions on future plans and investments,
companies simultaneously consider what long-term value can be created both for
society and for shareholders. This requires a long-term business perspective and is
based on the assumption that chances for sustainable business success are increased
when a company, as appropriate within the business strategy, invests in social
aspects that improve business conditions. CSV also implies a process wherein a
company assesses its value chain and identifies those primary points of intersection
between the company and society where social investments can improve the chances
of business success.
I coined and defined the term CSV in 2005 while serving as vice-president of
public affairs at Nestlé and it was shortly thereafter approved by the Nestlé Executive
Committee to describe the Nestlé approach (Nestlé 2009; Bockstette and Stamp
2011). It has since been adopted by other authors and by other companies and orga-
nizations (Porter and Kramer 2011). While the term is relatively new, the general
idea is not, and many companies began with a single person identifying a societal
need and fulfilling it with a very long-term vision. However, as these companies
have grown, the long-term vision of the corporation’s social function has often been
lost due to the day-to-day competitive pressures for survival in the marketplace, and
investor markets that disproportionally reward high short-term rates of return. Other
related terms which carry a meaning similar to CSV include: “Creative Capitalism”
(Bill Gates), “Business with a Purpose” (Pepsico), and “Ecoimagination” (General
Electric). What these terms share is the idea of building societal goals into the core
business strategies and operations of a company. The shared aim of these approaches
is to leverage the power of a company’s core business for health, social and economic
progress, and to do so with a very long-term perspective.

21.2 Contrasts with Other Business Models

The CSV approach contrasts with alternative business strategies focused on mini-
mizing current costs and maximizing short-term profit. Buying raw materials on
an opportunistic basis by jumping in and out of the market, development of low-cost
production facilities or outsourcing manufacturing to low-cost suppliers, and
offering the lowest cost products with minimal value-added features characterize
these latter business approaches. CSV companies, which also have short-term
profit goals, require a much longer term perspective and definition of success,
with the aim of establishing a strong market position in the long term. Among
consumer goods companies the aims are to secure a trusted brand reputation
21 Business Initiatives That Overcome Rural Poverty and Marginality… 355

among consumers for high-quality products, to build a reputation of trust with

suppliers by helping them become successful, and to secure a highly competent
and loyal work force by investing in long-term relationships with employees and
suppliers through training and advancement incentives. As a consequence those
companies that follow a CSV strategy generally have greater potential for positive
impact on society than those companies tied to a short-term view, and have higher
potential for consistent and reliable creation of long-term shareholder value and
rates of return to their investors.
Taking this approach requires four things. First, it requires a willingness to
assume long-term risk and strong financial management to permit making major
investments that may not pay off for 10 years. This is particularly true when enter-
ing areas that are populated by the marginalized poor and that lack the basic infra-
structure, educated work force, communications, and reliable business suppliers
necessary to serve the company’s manufacturing or raw material needs. Second, it
requires an understanding of what will benefit society in the long term, and where
those potential benefits intersect with a company’s business interests. Third, it
requires a planning process and mindset that translates the interests of shareholders
and the interests of society into concrete business plans. Fourth, this approach
requires patience and persistence aligned to a common value set and continuity in
personnel, as well as not being dependent on that significant segment of the investor
community that only pursues short-term results (for example Nestlé S. A. refuses to
be listed on any stock exchange that requires reporting of quarterly results).

21.3 What CSV Is Not

CSV means creating profit for shareholders and improvement in the standard of
living or the quality of life of suppliers, employees, and/or customers. It is profit
based and thus does not include social enterprises that do not have profit built into
their model and that are dependent on financial support from free capital or other
sources. These activities may be laudable and can be appropriate for some enter-
prises, but generally lack the potential for large-scale expansion of profit-based
models. Nor does CSV signify investing in activities that are not related to the
company’s value chain or that do not have a business rationale. Such activities have
a lower chance of long-term sustainability, as the fundamental reason for a com-
pany to engage in them can be lost. CSV also does not mean philanthropy, although
companies may decide to make philanthropic contributions to outside organiza-
tions as a part of being an active participant in the community and/or for maintaining
a license to operate. The impacts of a company’s donations to social enterprises or
philanthropic activities, even if substantial, pale in comparison with the positive
impacts of a company’s basic business investments if those are made properly
through the CSV approach.
CSV is also not simply business as usual or identifying customer needs and trying
to fulfill them. It implies a much longer business perspective, analyzing value chains
356 N. Christiansen

and developing a focus on where societal needs and a company’s business needs and
abilities to make a difference intersect. CSV means going beyond business as usual.
In summary, by building societal goals into the basic business strategy and opera-
tions, the potential for long-term sustainability is greatly enhanced. If an action is
part of a business plan and is tied to the long-term enhancement of corporate profit,
it will endure over time because its financial support is not tied to the largesse of the
company, but to a self-sustaining financial model. The potential for impact on a
massive scale is therefore also greatly enhanced because it is a part of core business
activities rather than something that is an additional activity to the main business.

21.4 Marginalized Rural Poor—Five Ways

to Escape Poverty

Reaching the marginalized rural poor is particularly difficult where the lack of
agricultural knowledge, infrastructure to reach markets, reliable access to water,
and access to credit and alternative work opportunities all present major obstacles
and result in low agricultural productivity and poverty. Robert Thompson (2011),
has laid out the following five primary ways for subsistence farmers to escape
poverty:
1. To increase productivity of current crops
2. To change to (or add) higher value crops
3. To gain ownership or access to more land or livestock
4. To obtain farm income through home manufacture of products or part-time work
in nonagricultural employment
5. To change to nonagricultural employment locally if opportunities exist, or by
moving to other (primarily urban) areas
In spite of the obstacles that face rural development among marginalized rural
people, some far-sighted organizations are having significant success in the reduc-
tion of poverty and malnutrition through investments that help increase food
production and allow farmers to climb out of poverty through one or more of these
five ways, without the necessity of migrating to urban areas. These organizations
include private businesses, cooperatives, foundations, and intergovernmental agen-
cies. They are working separately and/or jointly to make long-term investments in
remote rural areas including: manufacturing facilities, infrastructure development,
education of farmers for improved productivity, training of local suppliers, creation
of a skilled industrial labor force, and investments in the health of the population in
a way that has a dramatic impact on the poor, including the most marginalized. For
food companies, progress is being made through developing local production and
manufacturing of food products in remote rural areas to meet the rising demand in
emerging markets. Pioneering examples of this include the efforts of Nestlé in the
21 Business Initiatives That Overcome Rural Poverty and Marginality… 357

area of dairy product manufacturing (Goldberg and Herman 2006), as well as in

coffee, cocoa, grain, rice, legume, vegetable, nut, and spice farming.

21.5 The CSV Approach: The Nestlé Dairy Case

Nestlé currently operates about 450 food processing factories globally, of which
about 200 are in the developing world. About 150 of these are located in rural areas.
Nestlé first began its manufacturing operations in a developing country in 1920 in
Brazil, where it started its long history of going to remote rural areas in Latin
America, Asia, and some parts of Africa, and investing in poor marginalized farm-
ers in order to create a local dairy industry. These investments in farmers were made
in order build a reliable source of high quality milk to supply new, large scale milk-
processing and manufacturing facilities that were built in these rural areas. Initially
Latin America (including Brazil, Mexico, Colombia, Chile, Venezuela, Argentina,
and parts of Central America) was the focus of these activities in the early decades,
but in the latter part of the twentieth century investment in rural development in
Asia became widespread, including parts of India, Pakistan, Sri Lanka, China, and
Indonesia. Investments in the dairy industry in rural areas extend today to over 30
countries (Nestlé 2010).
This approach has yielded three major results. First, Nestlé has become the
world’s largest manufacturer of dairy products (in terms of sales). Second, millions
of farmers have climbed out of poverty in the process. Third, the availability of calo-
ries, protein, calcium, and various micronutrients has increased significantly in
these countries as the result of increased food production and availability.

21.5.1 The Dairy Farmer Development Process

Taking a system of organization and management of milk districts that was first
developed in Switzerland in the late 1800s and adapting it to each specific country
in emerging markets, potential dairy farmers are contacted among existing subsis-
tence crop farmers or farmers who already have one or two dairy cows for house-
hold needs. They are offered free education on raising dairy cattle, including how
to feed them for higher productivity and nutritional value, and how to maintain
hygiene standards necessary to maintain bacterial levels within safe levels. The
productivity of local dairy cattle is improved through free artificial insemination
from modern dairy breeds and free veterinary services are provided to help keep
animals healthy. Financial aid is provided at no cost to farmers that want to increase
their herds and/or improve their production facilities. Lastly Nestlé agrees to buy
all of the milk they can produce even though the farmers are free to sell their milk
to whomever they want.
358 N. Christiansen

21.5.2 Impact of Rural Factories in Marginalized Areas

The factories themselves have a major impact on poverty reduction, above and
beyond providing a guaranteed market for local dairy farmers. First, unskilled rural
workers are trained and transformed into skilled industrial workers with higher pay-
ing jobs. Second, local industrial suppliers are educated in how to provide needed
goods and services to the factory that meet high standards of quality and hygiene
brought from Switzerland, as well as given training on small business management.
Third, a skilled local factory management is developed and most expatriates ini-
tially brought in for training and start up purposes are replaced with local hires as
quickly as is feasible. Fourth, the region acquires a new tax base from the factory,
leading to local improvements in infrastructure, schools, and medical facilities.
Fifth, new standards of doing business are introduced, including environmental
standards regarding maintaining the water, land, and air, along with honest business
practices that help build trust in the brand.
As a result of the establishment of these new industrial complexes in rural areas,
towns grow into cities. This creates new employment opportunities for those who
have no future on the land and reduces the need to migrate to distant urban areas to
find acceptable work. In this way entire regions begin to rise out of poverty, includ-
ing the marginalized poor residents.

21.5.3 Nestlé Milk Districts in the Punjab: India and Pakistan

One example of the CSV approach is the Nestlé milk district network in the Punjab
region of India and Pakistan. While Nestlé started its milk district approach in Brazil
in the 1920s, work in India started much later with the building of the Nestlé Moga
milk factory in the Punjab region in 1961. While the Punjab traditionally has been a
major agricultural area, the production of milk products on an industrial scale was
not present in this region before the Moga factory was built. Today Moga is a huge
factory complex, employing 1,400 full-time staff and supporting jobs for over 5,000
people who work for suppliers.

21.5.3.1 Sourcing Milk in the Moga Region

In 1961 there were virtually no commercial dairy farmers in the Moga region.
Nestlé started the Moga factory with 180 farmers who produced 511 kg of milk per
day. Today about 85,000 farmers sell an average total of 837 tons of milk to the
Nestlé Moga factory every day, primarily in quantities of 5–10 l each. One of the
first obstacles to creating a dairy industry in the Moga was that the idea of produc-
ing milk for sale was not in the local farming culture. Locals typically kept one or
two cows in the yard of their village home for household dairy needs, but the idea
of selling milk from a cow or buffalo was initially met with suspicion. However,
21 Business Initiatives That Overcome Rural Poverty and Marginality… 359

the pragmatic farmers of Moga, nearly all of whom are Sikhs, soon realized the
advantages of diversifying their farm income and having a cash income every day
of the year rather than only when cash field crops could be harvested and sold.
The next obstacle was the bacteriological quality of the milk. For most farmers
two or three additional cows were added to their herds and kept in paddocks next to
the family home in the village. New practices of keeping the environment around
the cows clean, milking the cows in a hygienic way, and keeping the cows healthy
were all necessary to produce milk that is bacteriologically acceptable. Regular
visits by the Nestlé veterinary staff to offer free animal husbandry advice helped
solve this problem.
The milk needed to be received from the farmers in an accessible place where it
could be tested and stored until it could be taken to the Moga factory for processing
into milk products. For this purpose village cooling stations were set up in store
fronts with refrigerated tanks, and provided with a scale to weigh the milk and test-
ing equipment for purity. The staff at cooling stations also provides simple advice to
the farmers, who typically bring their milk can to the cooling station by donkey cart
or bicycle.
Increasing productivity per cow was also a necessity. Nestlé provided free artifi-
cial insemination of cows using superior dairy breeds to improve the milk produc-
ing capability of local cattle breeds and also provided access to financing a no cost
for the acquisition of additional cows. Cost-effective feeding to produce higher
quality milk was also necessary, so instruction was provided on what feed crops to
grow and low-cost chopping machines to convert crops into animal feed were sold
to the farmers. Those farmers who preferred to buy feed were offered commercial
feed at reduced costs.
As the men spend most of their time during the day tending field crops, it is the
women who take primary responsibility for care of the cows, and thus the women
from participating families suddenly became cash income earners. This new found
earning capacity helped to improve the social status of women across the Moga
region. Farmers were paid in cash every day for the milk they brought to the cooling
stations. Today with the advent of cell phones, cash payments are deposited directly
into family bank accounts electronically.
Appropriate disposal of manure is also an environmental necessity for maintain-
ing milk quality. Participating farmers were taught how to build brick-lined cisterns
that function as biogas digesters, where manure is digested and the methane gas
collected through the top of the structure. A pipe runs from the methane gas collector
to the house to provide fuel for both cooking and heating the household.
Given the importance of the Moga village farmers as stable suppliers of high-
quality milk for the Nestlé Moga milk factory, discussions were held between the
Nestlé agricultural extension workers and village elders about what was most
needed in the villages. Clean water was identified as a top priority. Drinking water
wells and pumps were installed next to the school in each village, and the villagers
and students were taught how to maintain the wells and pumps.
It was also discovered that attendance at schools by girls was low and that the
underlying problem was a lack of private sanitary facilities for girls. In response,
360 N. Christiansen

toilets that assured privacy were installed next to the village schools. Furthermore,
while the Punjab region is highly suited for agriculture, local nutritional knowledge
was limited, particularly what should be fed to children, therefore nutritional education
programs were introduced in the milk district villages. It was decided that educa-
tional efforts should focus on adolescent girls, as they are approaching the age at
which they will marry and begin raising families. Thus they would learn about nutri-
tional and dietary needs before and during pregnancy, and became better prepared
to properly feed their children from their first days of life.
The benefits to the farm families from this system are clear. Farm incomes
climbed significantly, women achieved a higher social status, households were more
able to afford education costs, and conditions in the villages improved, especially
for the children. The significance of these improvements to the villagers becomes
evident when new drinking water sources are inaugurated and entire villages come
together for a large celebration involving prayers, dancing, speeches, and a feast.
The benefit of these efforts for Nestlé is a large number of loyal and reliable
providers of milk, who can be trusted to produce an expanding quantity of high-
quality milk over the long term. The farmers are free to sell their milk to whomever
they want, but they continue to sell to Nestlé as their preferred client because Nestlé
is a reliable customer who will buy every day at stable prices. Even in times of milk
scarcity when farmers could get more money selling to other buyers they continue
to sell to Nestlé because of the bonds of trust that have been established.

21.5.3.2 Impacts of the Moga Factory on Job Training, Nonagricultural

Employment, and Urban Development

When the Moga factory was started in 1961 there was no skilled factory work force
available in Moga, so all of the current 1,400 factory employees had to be trained
to perform the operations of each factory department. To accomplish this, an
apprentice system developed in Switzerland and used in other countries where
Nestlé operates was translated and adapted to the local setting. Thus a skilled
workforce was developed in food processing technology that requires extremely
high standards of hygiene and food safety. Trucks are not allowed onto the factory
grounds, but must discharge milk and goods at loading docks at the exterior of the
factory. Employees must put on laboratory coats, bonnets that cover their hair, and
covers over their shoes. Hands must be washed at designated stations when enter-
ing each separate section of the factory. The grounds of the factory are kept in strict
order to reduce the possibility of contamination from foreign objects. This not only
has had an impact on the employees and their work standards, but also on the com-
munity as a whole. In one case a development minister of the Punjab told the
author that she took hospital managers on tours of the Nestlé Moga factory so that
they could better understand what measures a hospital should take to operate with
proper regard to hygiene.
Local suppliers needed to be developed for the various needs of the factory,
including such mundane operations as laundering the hygienic factory uniforms.
21 Business Initiatives That Overcome Rural Poverty and Marginality… 361

Classes in entrepreneurship were held to help develop successful local businessmen

and women, as well as literacy classes as needed. It is estimated that four jobs are
generated in the community outside the factory for every job in Nestlé’s factory.
This has led to the urban growth of Moga and to the availability of trained local
business people capable of operating their own businesses. While the marginalized
poor were not the only ones helped by this process of economic development, they
have benefitted greatly through increased work opportunities both outside and
inside the factory.

21.5.3.3 Environmental Impacts

In addition to teaching farmers how to collect biogas (methane) through home-

collection systems, which reduces wood fuel consumption and greenhouse gas
emissions, water management is a major focus of Nestlé’s education efforts.
Besides teaching farmers how to keep their sources of water clean, Nestlé has
sponsored a major study to determine means of stopping the depletion of water
resources in the Punjab. The repeated planting of water-intensive crops and virtu-
ally free electricity to run irrigation pumps on a continuous basis have seriously
lowered water tables in the Punjab. Together with the International Water
Management Institute, Nestlé undertook a study of the water intensities of millet,
wheat, and rice production in the region. The result was a plan for a comprehensive
response using best farming practices to make water sources more reliable and to
ensure the future of farming in the region. Rice, which requires 1,034 m3 of water
per ton to produce, is the most widely produced crop, whereas milk requires only
565 m3 of water per ton to produce. It was thus recommended that fodder area for
dairy cattle be increased at the expense of rice cultivation areas, as well as the
number of cows and the productivity of dairy cows. Research findings also sug-
gested that improvements in rice cultivation using technologies developed in other
regions could lead to lower demand on the water supply.

21.5.3.4 Development of the Dairy Industry in Pakistan

While Nestlé has been a supplier of milk products in Pakistan for many years, in
2007 Nestlé inaugurated its largest dairy product factory in the world in Kabirwala,
Punjab, Pakistan. This large investment was made based on the belief that the pro-
duction of milk by Pakistani farmers could be greatly increased. Approximately
seven million of Pakistan’s ten million farm families live in the Punjab, many of
which are landless. A number of collaborative efforts were begun to convert margin-
alized subsistence farmers into successful dairy farmers.
Nestlé had previously adapted its milk district system to the Pakistani Punjab on
a smaller scale and in 2006 joined forces with the United Nations Development
Programme (UNDP) and Engro Foods to create the “Community Empowerment
Through Livestock Development Program.” Nestlé had developed a program to
362 N. Christiansen

train female livestock workers, and together with the UNDP, this program was
adapted and expanded on a larger scale. The aim of the 3-year program was to
develop women entrepreneurs who possessed skills in animal husbandry and dairy
farming, and who could return to their villages and share their expertise with other
local women. They are provided basic training in livestock management and also
serve as a conduit for microfinance efforts.
So far over 4,000 women have successfully completed the program and returned
to their villages. Once trained, they are provided with instructional materials and a
veterinary kit to use in their villages that includes medical instruments, medicines,
and vaccines. These livestock workers provide a vital service. While it is almost
always the women that take care of the dairy cattle, they do not have access to
knowledge about animal husbandry due to cultural barriers, as nearly all of the vet-
erinarians are men. A woman veterinarian heads this training program and her
knowledge about animal care is transferred first to the trainees in the program, and
then to the village women who receive (and pay a nominal fee for) the advice given
by the trainees.
A second partnership in Pakistan involving the Swiss Agency for Development
and Cooperation is called “The Small Entrepreneur Development Project.” Livestock
and dairy farmers who are already productive are given training to both enhance
their skills as small business managers and to develop links to markets to improve
milk and meat sales. To have appropriate facilities to instruct the farmers Nestlé is
building two demonstration farms, the first to be completed being the 103 acre
(41.7 ha) Sarsabz Demonstration and Training Farm. The Sarsabz farm houses
about 250 cattle and includes animal sheds, milking machines, other necessary farm
equipment, classrooms, and lodging facilities for visiting farmers. The instructors at
the facility are drawn from the 75 professionals who make up the Nestlé Pakistan
Agricultural Services division, and both basic and advanced farm management
courses are offered. Selected farmers are given training kits and accessories for
promoting good dairy farming practices and serving as trainers in their villages.
Approximately 30,000 farmers per year receive training through this program in
Pakistan. Of course the benefit of these activities for Nestlé is a reliable source of
quality milk for the Kabirwala factory, while the farmers simultaneously emerge
from poverty using the skills they have learned and the assistance they have received.

21.5.3.5 Impacts on Food Availability and Consumers

Milk is a particularly good source of protein and calcium, and can be an excellent
carrier of micronutrients (vitamins, zinc, iron) when those are added through forti-
fication. Milk and dairy products are important traditional sources of food in India
and Pakistan, and there is a large-scale shortage of milk, particularly of milk of
known nutritional value and safety. Most of the milk consumed locally is raw,
unpasteurized milk that can pose significant health risks when transported or stored
improperly, and liquid milk purchased through traditional channels is often
adulterated in ways that decrease quality.
21 Business Initiatives That Overcome Rural Poverty and Marginality… 363

Nestlé has developed milk products that are particularly affordable to lower
income groups using food technology to replace butter fat with lower cost fats and
that are fortified with iron, zinc, and essential vitamins. The broad impacts of this
fortification are only possible because milk processing has been introduced on a
large industrial scale and because dairy farming has greatly increased production
through the milk district system. Millions of tons of milk and milk products are
added annually to Indian and Pakistani diets as a result of these investments and
efforts. Virtually all of the milk products have been nutritionally enhanced through
the addition of iron, vitamins, minerals, as well as prebiotics and probiotics aimed
at improving resistance to infection.

21.6 Targeted Versus Community-Based Solutions to Lift

the Rural, Marginalized Poor Out of Poverty

This CSV approach to poverty reduction is obviously a community-based approach

and doesn’t begin by exclusively targeting the marginalized poor. To make this
development approach work, various levels of the community have to be involved
and those who are helped economically aren’t just the marginalized poor, but a
broad spectrum of the population that makes up the community. The overall impacts
on rural development include the construction of factories, the training of farmers,
and the creation of new clusters of collaborating sectors that benefit the community
as a whole, as does the local tax revenue and urban growth associated with this
approach to business and development. Helping the marginalized poor is possible
because of massive investments to produce products that are distributed to large
urban markets through the distribution system of the company involved.
There are certainly other instances where the rural poor are too segregated for
this approach to work, such as in refugee camps or areas where the geographic and
climatic prerequisites for increased food production do not exist. The CSV approach
does require preconditions that provide the potential for business creation. In the
case of dairy products this potential requires adequate conditions for cattle raising
such as water and grazing lands.

21.7 Conclusion

It is clear that substantial underinvestment in agriculture has been a major factor in

maintaining the current states of poverty among the rural, marginalized poor. The
CSV approach as illustrated by the Nestlé business model illustrates one seemingly
important part of the solution to this problem, but each situation has its own charac-
teristics and solutions. Significantly increased investment in agriculture from the
public as well as the private sector, the application of modern science and technologies,
the transfer of knowledge and skills, access to credit and markets, as well as a better
364 N. Christiansen

understanding of how the private and public sectors can work together, will be
necessary if we are to significantly reduce the levels poverty and malnutrition cur-
rently experienced by over one billion people on our planet.

Acknowledgements The author thanks Hans Joehr, Head of the Nestlé Agriculture Department,
and his staff, for reviewing and providing data and comments to drafts of this chapter.

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Bockstette V, Stamp M (2011) Creating shared value: how-to guide for the new corporate
(R)evolution. Foundation Strategy Group, Geneva
Goldberg R, Herman K (2006) Nestlé’s milk district model: economic development for a value-
added food chain and improved nutrition, Harvard business school case study. Harvard Business
Publishing, Boston
Nestlé (2009) Creating shared value report, pp 2–3
Nestlé (2010) Rural development report. Nestlé SA, Geneva
Porter ME, Kramer MR (2011) Creating shared value, pp 62–77. Harvard Business Review,
January 2011
Thompson R (2011) Food security and the future of farmers: 2020 and toward 2050. J Swed
R Acad For Agric 1:43
Chapter 22
The Marginalized and Poorest in Different
Communities and Settings of Ethiopia

Tadesse Woldemariam Gole, Fite Getaneh Ilfata, Motuma Tafa,

and Aleka Aregachew

Abstract This chapter describes how the extent of poverty and the causes and
nature of marginality vary from place to place, depending on the natural resource,
livelihood strategy, climatic, agro-ecological, and socio-cultural conditions. Local
attributes of marginality were assessed in four different districts of rural Ethiopia.
Extreme land degradation was a key force in one district, lack of credit in another,
competitive commercial pressure in another district, and resource management
practices according to cultural values in the remaining district. There was always
a complex interplay of various factors that exclude some people from the benefits
of economic growth that others enjoy. Independently from the different features of
marginality, this interplay was found to be a root cause of poverty in all of the
districts studied.

Keywords Ethiopia • Marginality • Poverty • Agro-ecology • Local narratives

22.1 Introduction

Poverty is widespread in developing countries like Ethiopia. There are both global
initiatives and national poverty reduction programs, like the United Nations’
Millennium Development Goals (MDG) and domestic efforts such as the Poverty
Reduction Strategy Papers and the Productive Safety Net Programs (Admassie and
Abebaw Chap. 17 this volume). The national government, NGOs, and bilateral and
multilateral organizations are all making concerted efforts to meet the MDGs and to
implement national poverty reduction efforts. However, poverty seems to persist

T.W. Gole (*) • F.G. Ilfata • M. Tafa • A. Aregachew

Environment and Coffee Forest Forum (ECFF), Addis Ababa 28513, Ethiopia
e-mail: [email protected]

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 365
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4_22,
© The Author(s) 2014
366 T.W. Gole et al.

everywhere in Ethiopia. The severity of poverty and the proportions of the poor
vary from place to place, depending on various factors. Marginality is often found
to be a root cause of poverty (von Braun and Gatzweiler Chap. 1 this volume).
Socio-economic development efforts may not equally reach all of the poor because
the barriers that marginalize some segments of society prevent policies and devel-
opment efforts from comprehensively incorporating the poor.
Marginality has different meanings depending on context. In this chapter, we
consider marginality as Gatzweiler et al. (2011; see Chap. 2 this volume) define it.
This study was conducted as part of a pilot research project on marginality and
extreme poverty in Ethiopia led by the Environment and Coffee Forest Forum in
Ethiopia. We conducted a field investigation in four communities located in separate
districts of the regional state Oromiya with distinct environmental, social, and
cultural circumstances. The districts (locally called woredas) are: Jida in North
Shoa, Yayu in Ilubabor, Adami-Tullu Jiddo Kombolcha in East Shoa, and Arero in
Borana. The case study sites were selected in order to capture some of the diverse
aspects of marginality by considering the differences in agro-ecological contexts,
livelihood strategies, cultural practices, and natural resource endowments. The
objectives of the study were:
• to assess the diverse factors that contribute to marginality and poverty across
gradients of agro-ecological and climatic conditions and livelihood strategies
in selected parts of Ethiopia
• to better understand the perceptions of poverty among the rural marginalized
poor communities

22.1.1 Agro-ecological Characteristics of the Study Sites

Jida district—Central Highlands:

• cool humid or semi-humid climate
• highly degraded land, steep slopes, and seasonally flooded plains at the base of
slopes
• local livelihoods are primarily based on cereal production and mixed-agricultural
systems
• annual precipitation is highly variable (635–1,350 mm)
• mean annual temperature is 15 °C
• the primary sources of livelihood are food crops and livestock production
Yayu district—Southwestern Highlands:
• area is well endowed with natural resources, over 50 % forest cover remaining,
fertile soils, and streams and rivers flow throughout the year
• this area is relatively far from major markets
• local livelihoods are primarily based on food crops and coffee production. Coffee
alone was the source of around 75 % of the cash income for the study community
or kebele (Ethiopia’s smallest geopolitical unit) of Wabo near the town of Yayu
22 The Marginalized and Poorest in Different Communities and Settings of Ethiopia 367

• mean annual precipitation is about 1,850 mm, with annual variation ranging
1,600–2,600 mm
• mean annual temperature is 20 °C
Adami-Tullu Jiddo Kombolcha district—Central Rift Valley:
• arid to semiarid climate with topography ranging from lowlands to mid-altitude
highlands
• annual precipitation is between 600 and 800 mm
• mean annual temperature is 20 °C
• area has fertile, young, volcanic soils and extensive surface and groundwater
resources
• water for irrigation is available (Lake Ziway and its tributaries)
• livelihoods are primarily based on the production of food crops such as fruit
and vegetables supplemented by livestock production
• relatively close proximity to major domestic markets
• this is a major vegetable producing area of the country, both by commercial
and small-scale producers (commercial farms also produce flowers for export,
and grapes and other fruits for local markets and industries)
Arero district—Southern Lowlands:
• arid climate
• predominantly covered by Acacia spp. woodlands and dry forests that are
managed as rangeland for livestock production
• mean annual precipitation is about 600 mm with a bimodal distribution pattern
• mean annual temperature is 20 °C (Halake 2010)
• local livelihoods primarily depend on nomadic livestock production, although
a minority has adopted cultivating maize on small parcels close to permanent
settlements
• rangelands are highly degraded with exposed soils that are nearly devoid of
vegetation in most areas

22.1.2 Methods

The four districts were systematically selected based on their distinct climatic,
physical, and agro-ecological conditions. In each district, consultations were held
with administrators and senior agricultural development officers in order to select
each case-study community. At the selected study kebeles, discussions were held
with local experts and key local informants such as development agents and
extension workers to select the specific case study community within each kebele.
In the selected communities, group discussions were held with representatives of
different demographic groups (youths, women, elders) and community representa-
tives. Interviews with key local informants were held with two or three individual
household heads from each of three locally determined relative wealth categories:
poor, middle-income, and wealthy households.
368 T.W. Gole et al.

Semi-structured interviews were used to consult administrators, agricultural

development officers, extension workers, and development agents at the district and
kebele levels. The local farming community consultations were held in two stages,
first with a group of key informants and community representatives, and later with
individual household heads from each wealth category. Participatory Rural Appraisal
(PRA) tools such as group discussions, transect walks, participatory mapping, and
wealth-ranking exercises were conducted with the community representatives. At
this stage the key local informants defined and characterized the three household
wealth categories and identified two or three household heads from each wealth
category for further individual discussions. Afterwards, semi-structured interviews
were conducted using checklists of discussion points to obtain life history and other
household data. Guided walks with the key informants were undertaken to observe
local biophysical conditions and the status of natural resources.

22.2 Findings

The marginalized poor identified in the study sites included individuals of all ages,
although the perceived root causes of poverty differed among social and demographic
groups. Perceptions of poverty and wealth also varied by location. The major
emphasis of our investigations was to characterize natural resource bases, economic
and livelihood conditions, infrastructure, and other legal and human rights related
issues that we expected to have an impact on marginality and poverty. The importance
of these factors differed among locations and hence emphasis was placed on factors
that appeared to be more relevant to specific case study sites.
Jida. This district is located 70 km northeast of Addis Ababa. Despite its vicinity to
the capital city it has suffered from a lack of basic infrastructure like electrical
power, telecommunications, and environmental related investments. Climatically
the area is classified as temperate and there are three distinct seasons locally known
as the Bega (a dry season from October to January), the Belg (a brief rainy season
from February to May), and the Kiremt (a longer rainy season from June to
September). The greatest amount of annual precipitation occurs during the Kiremt.
The major land cover types are cropland and rangeland, with remaining areas
covered by forest and scrub. The total human population of the case study kebele
in 2007 was 8,329, with nearly equal numbers of male (4,129) and female (4,200)
residents (CSA 2007). The local livestock population was approximately 18,310
head, dominated by sheep, followed by cattle, donkeys, horses, goats, and mules.
Jida is a typical example of the highly degraded conditions found in much of the
Central Highlands of Ethiopia. Erosion has led to low infiltration capacity and high
runoff on upper slopes, resulting in seasonal flooding and gully formation in the
valley bottoms. The short rainy season (Belg) is the best season for crop production
because the higher rainfall during the Kiremt causes flooding and water-logged soils
that damage crops. Some informants mentioned that the Belg rains only come at the
22 The Marginalized and Poorest in Different Communities and Settings of Ethiopia 369

Table 22.1 Locally determined attributes of wealth in the community of Gango, Jida district, some
indicators apply to more than one wealth category, distinctions were made using a combination
of indicators
Rich Middle-income Poor
Owns a house with a corrugated Owns one or two pairs Does not own oxen or sheep
steel roof with separate rooms of oxen Does not own land or provides
for family, guests, kitchen, Owns two head of cattle land to share croppers
livestock, and storage Owns around ten sheep Owns a simple house
Owns more than two pairs of oxen Generally food self with a grass roof
Owns more than three head of cattle sufficient Unable to send children
Owns more than ten sheep Capable of sending to school
Food self-sufficient even children to school Unable to provide appropriate
in unproductive years Capable of going clothing for family
Capable of lending money to others to health center Unable to go to health center
Owns a horse, mule, or donkey without support in case of sickness and
Capable of sending children to Capable of supporting at risk of death as a result
school up to the secondary-level family for 1 year in of serious illness
(which requires renting a room case of crop failure
in a nearby town) Owns a horse,
mule or donkey

appropriate time and amounts for crop production once every 3 years, leaving the
population in a perpetual food deficit for two out of every 3 years. As a coping
mechanism most young male household heads migrate to Addis Ababa to work in
the construction sector as unskilled day laborers. Elderly people, women, and other
residents who are incapable of migratory labor have only one choice to deal with
adversity, to take out loans from illegal money lenders.
Livestock production seemed to be a stable livelihood option in the area due
to its proximity to Addis Ababa, where there is a very high demand for sheep. The
main constraints for livestock production were water shortages during the dry
season and disease. Based on their own economic status and perspectives, local
participants in the focus group discussion in the community of Gango helped
to characterize the local attributes of the three wealth categories (Table 22.1).
Local representatives of each of the wealth categories were interviewed to better
understand their perceptions of poverty and associated causes.
Mr. Hailu Gurmu was a 72 year-old household head identified as poor.
His household had seven members. Mr. Gurmu was frustrated during the study
year because his barley crop failed due to meager Belg rainfall. To survive the sub-
sequent food shortage, he needed to either sell one of his sheep or else borrow
money at a high interest rate from a rich man in the village. His son had to migrate
to Addis Ababa to work as a day laborer. This case revealed that even for house-
holds with some assets (land), productivity depends exclusively on environmental/
climatic factors like rainfall. Although Mr. Gurmu was losing hope on the prospects
of crop production, he viewed raising sheep as a good livelihood opportunity;
however, he did not have the financial resources to buy more sheep.
370 T.W. Gole et al.

Box 22.1 Narrative of Mr. Tadesse Teshome

Tadesse Teshome was a 69 year-old male household head responsible for a
family of eight. Most of his children were married and he lived with his wife
and some of his grandchildren. He owned two houses with corrugated steel
roofs, one for his family and the other for livestock. There was a third grass
thatch roofed house used to shelter horses and as a cooking area. He owned
2.5 ha of land and also cultivated another 2.5 ha as a share cropper. He owned
2 pairs of oxen, 4 head of cattle, 15 sheep, 2 chickens (others had died from
disease), 3 donkeys, 2 horses, and a mule. At the time of the interview he was
cultivating eucalyptus plantations and was planning to plant more. He sold
sheep and sometimes oxen to generate income. When rainfall was good, he
sold surplus crops (barley, wheat, beans). He mentioned the erratic nature of
rainfall as the major challenge for both crop and livestock production in the
kebele. He credited hard work as a reason for his success. As a well-off person
in the village, he claimed to support other people by providing interest-free
loans (a matter later disputed by poor local residents who claimed to have
taken out loans from him with interest) although he admitted that loan
recipients sometimes provided labor. He saved money through Iqub. When
asked what he did with his money, he mentioned that he had bought a grinding
mill. He also mentioned that he used to be a tailor and that his business
had been inherited by his son. Mr. Teshome was planning to purchase another
grinding mill in the near future.

Mrs. Ilfe Badho was a 55 year-old female household head identified as belonging
to the middle-income category. In addition to agriculture she diversified her sources
of income by preparing drinks and selling them locally, and was able to save money
through local institutions such as Iqub. She also took advantage of access to
micro-finance loans. Mrs. Badho claimed that her self-motivation, hardworking
nature, and support from her children also contributed to her economic status.
Mr. Tadesse Teshome belongs to the rich local wealth category. Income diversi-
fication was one of the primary reasons for his economic success. He receives
income from share cropping, livestock husbandry, a grain grinding mill, money-
lending to poor households, tree plantations, and a home garden. Details from the
interview with Mr. Teshome are summarized in Box 22.1.
Yayu. This district is located in the Illubabor Zone of the regional state of Oromiya,
approximately 550 km west of Addis Ababa on the major highway to Metu. The
district is composed of 18 kebeles, of which 17 are rural and one is urban. Five land-
cover types were observed in the district, which was dominated by forest and agri-
cultural crops, and included smaller areas of rangelands, wetlands, and settlements.
22 The Marginalized and Poorest in Different Communities and Settings of Ethiopia 371

Yayu district has the highest percentage of forest cover in Ethiopia. Most forested
areas and the surrounding landscape are managed as the UNESCO (2010) desig-
nated “Yayu Coffee Forest Biosphere Reserve.” The local communities depend
heavily on the forest, mainly for coffee, spices, and honey. The main local food
crops were maize, sorghum, and teff. Coffee was the major cash crop in the area
(although it is also consumed locally), followed by chat, which is a leaf chewed as
a stimulant by many people in urban areas (Gole 2003; Gole et al. 2009; Fite 2008).
According to the most recent census data available the human population of the
district was 52,829, of which 26,720 were male and 26,109 were female (CSA
2007). There were 9,917 households, out of which 1,348 were female-headed.
Around 4,103 (41 %) of the household heads were unemployed at the time of
our interviews.
The district is characterized by the high amount of forest cover, high rainfall,
high suitability for coffee and honey production, and considerable potential for the
production of food crops like maize, sorghum, teff, and others. Local communities
considered the undulating terrain, soil erosion, and low productivity as major
challenges to crop agriculture. Many areas have steep slopes and require soil
conservation measures to avoid degradation and severe soil erosion. Most high
slope areas, however, are forests managed for coffee production with minimal
risk of erosion. For croplands on slopes, NGOs like Menschen für Menschen,
Ethio-Wetlands, and the Natural Resources Association have introduced “bund”
terraces that are planted with vetiver grass to prevent erosion.
The area is accessible via a highway that connects Addis Ababa to Metu,
Gambella, and South Sudan. However, the local road network connecting the kebele
to the highway and local towns was not well developed. Given the steep slopes and
difficult terrain, transporting products to markets was a major challenge in the area.
Telecommunications infrastructure and access to electricity, clean water, and health
services were at rudimentary stages of development. Most farmers find it difficult to
transport their harvests (especially coffee) to market during the rainy season. As a
result, a few wealthy merchants from local towns built warehouses in rural areas
where they purchase coffee at low prices directly from producers and store the
product until they can sell it for a higher price. Poor farmers who cannot process
their own product or keep their product until the dry season (when prices are better)
have no choice but to sell their coffee at very low prices. There are few secondary
schools and health centers accessible to most rural residents, in some kebeles people
must walk up to 3 or 4 h in order to reach the closest health centers.
Coffee production was the major source of income (over 70 %) for most house-
holds in Yayu. Honey and spices were other important sources of income. Recently
some households also started growing fruits, vegetables, and chat. Most of these
livelihood options were linked to the ownership or access to land. In the absence of
land markets, the typical size of inherited land parcels has gotten smaller over time
due to population growth. As a result, harvest yields (especially of food crops) were
getting smaller and in many cases could no longer support household consumption,
making most small-producers reliant on food purchased from markets.
372 T.W. Gole et al.

Coffee prices are influenced by the global market and vary considerably. The
productivity of coffee plantations depends on the intensity of management, which
requires labor that the poor may not be able to afford or provide, thereby reducing
available income. This situation prevented most households from being able to
support their family year-round based on coffee income, making it necessary to find
other sources of income. Some farmers also worked as day laborers on larger coffee
plantations or in nearby towns. Elderly and female landowners often had to offer
their land to others for share cropping. In the case of shortages of food and other
consumables for the family, households occasionally have to borrow money from
local lenders at high interest rates.
Share cropping is a common practice in many parts of rural Ethiopia. It is an
option for households that own land but cannot afford the production inputs such
as oxen for plowing or the required labor. In such cases landowners look for a
share cropper who can provide the inputs and labor required to cultivate the land
in return for half of the harvest. This is an attractive opportunity, especially for
households headed by women or elderly people who cannot manage farmland
themselves. In the case of crop production, the share cropper provides seed, fer-
tilizer, labor, and oxen for plowing. In the case of coffee production the share
cropper employs labor for weeding (which is carried out at least three times per
year) and harvesting.
A problem with share cropping is the lack of regulatory oversight or legal
mechanisms for ensuring compliance with agreements made between land own-
ers and share croppers. As a result, the share croppers sometimes refuse to pro-
vide the appropriate harvest share to the land owner or else find ways to claim
legal rights to the use of the land under false pretenses without the assent of the
owner. During the community consultations respondents mentioned that there
were many households that had lost their land in this way and afterwards
remained landless.
In the Yayu district the poor often receive money from the rich under an agree-
ment to repay the debt with coffee. Agreements are often made when coffee prices
are very low (mainly in the rainy season), with loan payments fixed at the current
price of coffee. Payments are then made later in the year, when the coffee prices
increase (reaching double or over three times the price when loan agreements were
made). Such contracts reduce the earning capacity of small-producers, helping to
keep them in a state of poverty (Table 22.2).
One of the poor household heads identified by the community and interviewed
was Mrs. Alemi Kumsa, a single woman over 60 years of age who lived alone. She
owned a 0.5 ha coffee plantation that she contracted to share croppers and 0.5 ha of
cropland, which was rarely cultivated due to lack of labor. She claimed that share
croppers often cheated her and rarely provided the proper harvest shares. The tragic
loss of her husband and sons significantly impacted her life. Share cropping agree-
ments were made through the mediation of brokers with no legal backgrounds.
Due to illiteracy, people like Ms. Kumsa are not able to verify the terms of written
agreements and are required to believe what they are told by mediators, which in
some cases is not consistent with the signed agreement.
22 The Marginalized and Poorest in Different Communities and Settings of Ethiopia 373

Table 22.2 Locally determined attributes of wealth in the kebele of Wabo, Yayu district
Rich Middle-income Poor
Owns a house with a Owns a house with either Cannot provide adequate food
corrugated steel roof and a corrugated steel or for family
full furnishings, including grass thatch roof Works as day laborer or sells
a sofa, TV, and shelf Owns 0.5–1.0 ha of fuel wood for income
Owns more than coffee plantation with May own a coffee plantation
one pair oxen moderate management but must allow another person
Owns more than 1 ha Owns 0.5 ha of cropland as share cropper or in exchange
of well-managed Owns a single ox for loan and may work
coffee plantation Co-owner of a cow on own land as laborer
Owns more than 0.5 ha (shared among If they possess a coffee plantation,
of cropland households) it is less productive because
Food self sufficient Owns two or three sheep of poor management
Capable of sending Food self sufficient Unable to send children to school,
children to school Unable to withstand poor who must either live with
Capable of providing decent growing conditions others or work
clothing for family Capable of sending Does not own livestock of any type
Capable of storing maize children to school Depends completely on labor
Capable of supporting family May own a television for income
for more than a year in set and sofa Needs support for minor health
case of production failure services
Owns one donkey
Owns more than five sheep
Owns more than five goats

In contrast to Alemi Kumsa’s situation, middle-income household head Sukare

Firisa’s (Box 22.2) educational status enabled her to avoid similar problems. Her
children were highly educated, which seemed to have contributed to her ability to
defend her rights with respect to her share of harvests from sharecroppers and the
ability to manage her own coffee plantation. Apart from her education, she was
relatively young and healthy and therefore had the physical strength to work. She
also earned income by brewing drinks that she sold locally. She did not have savings
of any type, which might have enabled her to improve her economic status.
Similar to the other sites, diversification of livelihood means was common among
households identified as belonging to the rich wealth category in the community of
Wabo. Mr. Fikadu Hailu, a Wabo household head belonging to the rich category,
was engaged in coffee, food crops, chat, and sugarcane production. He also owned
eucalyptus plantations. Some households identified as rich in this area had also
expanded into providing specialized services and opening consumable goods shops.
Adami-Tullu Jiddo Kombolcha. This district is in the East Shoa Zone of Oromiya
and had a total population of 142,861 at the most recent census (CSA 2007). The
case study was conducted in the kebele of Bochessa, which is located 6 km east of
the town of Ziway. Bochessa had a total of 480 households and 2,855 people. The
area is well endowed with natural resources. It has fertile, relatively young, volcanic
ash soils and abundant surface and groundwater resources. The area is suitable for
374 T.W. Gole et al.

Box 22.2 Narrative of Ms. Sukare Firisa

Sukare Firisa is a female household head identified as belonging to the
middle-income wealth category. She was 45 years old and had five children,
only three of which lived in her home, which had a corrugated steel roof.
The house had three rooms for the kitchen, a bedroom, and a salon furnished
with a sofa, television set, and a shelf. Two of her children lived with her
ex-husband. Her son was a university student and her daughter had graduated
with a nursing degree and was looking for a job. Ms. Firisa was a ninth grade
dropout and reliant on agriculture for her livelihood. She owned a 0.75 ha
coffee plantation and 0.4 ha of cropland. She also owned a shared ox, a cow,
three calves, and shared a fourth calf. She provided her cropland to share
croppers due to her inability to cover labor and input costs, for which she
obtained 600 kg of maize each harvest. With part of the maize harvest she
prepared a local alcoholic drink called arak that she sold locally. Income from
arak sales enabled her to buy teff to add to the remaining maize for the house-
hold food supply. She would have inherited a larger coffee plantation, but her
late step-mother made a share-cropping agreement on part of the land to
someone who later claimed it after the death of the step-mother. Firisa was
capable of defending her rights with respect to the share croppers. She
managed the coffee plantation by herself. When coffee income declines,
she increases arak production and sales. She did not participate in the Iqub
saving system or save by other means. She claimed that her reluctance to save
stemmed from her mistrust in others, based on prior experiences with people
who did not return money she had given them.

all kinds of agriculture using irrigation, including high-value fruits and vegetables,
oil seeds, flowers, other food crops, and livestock production. This has attracted
considerable investment in floriculture, large and small-scale irrigated agriculture,
agro-industries, and recreation. The most fertile and accessible areas were allocated
to such investments, reducing the availability of irrigable land for the local com-
munity and causing high rates of landlessness.
Rain-fed agriculture is not promising in the area due to the relatively low amount
of rainfall. Therefore access to irrigable land was the most critical factor determining
the wealth status of households in the area. The soils in the area have salinity
problems, which further reduce productivity. The interplay of these three factors
(low land ownership, erratic rainfall, and saline soils) greatly diminishes production,
leading to a widespread lack of household self-sufficiency in food production.
The district is relatively well developed compared to other parts of Ethiopia.
The major highway connecting Addis Ababa to the southern part of the country
and Kenya crosses the district and good feeder roads connect the kebeles to major
towns and the highway. Schools, mobile phone services, clinics, and health posts
are available in most areas. All towns and major rural settlements have electricity.
22 The Marginalized and Poorest in Different Communities and Settings of Ethiopia 375

Table 22.3 Locally determined attributes of wealth in the kebele of Bochessa, Adami-Tullu Jiddo
Kombolcha district
Rich Middle-income Poor
Owns more than 5 ha of fertile Owns 4–5 ha of fertile land of Lack grain storage
land out of which at least 2 ha which a minimum of 1 ha is capacity near home
are irrigable irrigable Owns almost no land
Owns more than four pair of oxen Owns 2–3 pairs of oxen except home/garden
Owns a good house, with a Owns 10–15 head of cattle and plot
corrugated steel roof 8–10 goats Does not own oxen and
Produces sufficient yield to secure Four or more family members other livestock
household nutrition needs capable of working on farm Owns few home utensils
Capable of providing loans and Sufficient financial resources or Is in poor health
accessing credit access to credit to buy Physically incapable of
Capable of sending children to agricultural inputs to boost work on farm or as
school yields day laborer
Physically capable of working Recipients of government
Large family labor force to work agricultural extension
on farm services, input supplies, and
Household heads’ ideas/opinions asset protection during poor
are accepted by many and harvests
well respected by most Capable of storing sufficient seed
community members for planting next crop and
enough pasture for livestock

Transportation is not a problem in the area due to the good road networks and
common use of carts in both rural and urban areas. The town of Ziway is the
commercial center of the area, connecting the larger markets in Addis Ababa,
Adama, Shashamanne, and Hawassa.
Mixed agriculture (crop and livestock production) is the major livelihood activity
in the area. The major staple food crops produced in the kebele are maize, teff,
wheat, and barley. Local farmers also produced Haricot bean, tomato, and onion as
cash crops. The typical size of land holdings per household range from 0.75 to
3.0 ha. About 46 % of the rural population is landless, 27 % owned 1–2 ha, and
22 % owned less than 1 ha. Share cropping and land leasing for periods of up to a
few years are common practices in the area, by both local subsistence farmers and
commercial producers from urban areas.
With increased urbanization and commercial agriculture around Lake Ziway,
employment as day laborers and petty tradesmen is increasingly becoming an
alternative or complementary source of income for local residents. Other livelihood
sources included fishing, livestock production, livestock finishing (pre-slaughter),
and poultry production. The results of the local wealth ranking exercise in Bochessa
are presented in Table 22.3.
As the representative head of a poor household, Mr. Wario Dinko, a 60 year-old,
physically disabled man with a family of three owned 0.75 ha of non-irrigated land.
He did not own oxen, but did own three chickens. Because of his inferior economic
status, Mr. Wario felt marginalized. He believed that the voices of the poor like
himself are not heard or are not taken seriously by the community, and that their
376 T.W. Gole et al.

Box 22.3 Narrative of Mr. Kumbi Ababa Tufa

Kumbi Ababa Tufa was an illiterate 65 year-old man and the head of a family
of 11. He was able to send all of his children to school at the appropriate age.
He owned 1.5 ha cropland (of which 1 ha was irrigated), 2 oxen, 2 dairy cows
and 4 other cattle, 2 sheep, and 2 donkeys. In addition, Mr. Tufa had one
house in the town of Batu where he rented three rooms and earned 500 birr
(US$30) per month. In his home he had a television set and a tape recorder.
He also had a bicycle for transportation and a motorized pump for irrigation.
In 2010–2011 he harvested 4,000 kg of maize and 100 kg of teff. He attributed
his success to hard work. He earned enough money to pay for most of his
assets from fishing on Lake Ziway for many years. He was convinced that
hard work and saving money can change the life of any person.

problems are not known to development agencies or policy makers. From his
perspective, many policies and development activities fail to include the priorities
of the marginalized poor. Similar poor households encountered in the area had
very little land, none of which was irrigable. Groundwater is available but requires
investment to access, which poor households could not afford. Most poor house-
holds are unable to support their families and depend on direct government aid
through social safety-net programs.
On the other side of the spectrum, Mr. Kumbi Ababa Tufa was identified as a
rich household head in the community (Box 22.3). The amount of land owned by
Mr. Tufa was not much more than typical members of the middle-income category.
However, he managed the land on his own, was able to irrigate, and produced
high-value vegetables. In addition, his life story demonstrates the potential of
fishing as a livelihood option that landless farmers could take advantage of.
Arero. This district is located in the Borana Zone of Oromiya, the district capital
Mata Gafarsa is located 675 km south of Addis Ababa. The district has a total area
of 957,639 ha, of which only 5 % is cultivated land. The greatest land cover in the
area was rangeland, followed by a lesser amount of forest and settlements. There
are 21 kebeles in the district, of which 20 are rural and one is urban. The case
study kebele, Alona, is a rural kebele with the same land cover pattern that is
typical for the whole district. Rainfall during both rainy seasons is erratic. In
2010–2011 there was no rain at all during the shorter rainy season. The district is
an important watershed of the Dawa River, one of the major tributaries of the Juba
River in Somalia. The total population of Arero district was estimated to be 73,500
in 2007 (CSA 2007). The number of households in the district was 12,487, of
which 6,722 (53.8 %) were female-headed. In Alona female-headed households
constituted only 36 % of the total. The principal reasons for the exceptionally
high percentage of female-headed households in the district were polygamy and
a high divorce rate, and violent conflict among ethnic groups. About 95 % of the
22 The Marginalized and Poorest in Different Communities and Settings of Ethiopia 377

rural population belonged to the Oromo ethnic group, others included Somali,
Konso, Gari, Burji, and other ethnicities.
Natural resources in Arero include high altitude dry forests, lowland woodlands,
diverse and unique species of wild animals and mineral resources. Dry Afro-montane
forests cover highland areas above 1,500 m. The lowland woodlands were rich in
various useful species of acacia, gums, and incense-bearing species. The district is
rich in mineral resources like gold, platinum, tantalum, and precious stones. Most of
the natural resources are not being exploited by local communities.
The communities in the area are predominantly pastoralists and the major
economic activity is livestock production. Over time the pastoralists of this area
have developed unique breeds of cattle, camel, goat, and sheep. The local Borana
livestock breeds are known in Ethiopia for the quality of their dairy and meat,
which are also exported. Livestock resources are threatened by frequent drought,
shortages of animal feed, and the encroachment of invasive woody plants on
pasture lands.
The road connecting the district to major administrative towns and markets is in
very poor condition, making traveling and transporting goods in the area very
difficult. Most rural villages are not connected to the road network and residents had
difficulty accessing health care services and markets. The major local commercial
center is the town Mata Gafarsa. The nearest livestock markets are in the Yabello
and Mega districts that require 2 or 3 days to reach on foot. Telecommunications,
drinking water distribution, and electricity infrastructure were not available in the
area. The district residents fetch water from rivers or streams during the rainy
season and from springs, deep wells, ponds, and even from areas as far as 100 km
away like the town of Yabello during the dry season. Generally, water was a scarce
resource in this district.
On the other hand, human and veterinary health posts and schools are available
in many kebeles, including the case study site. There were also three health centers
in the district. Yet, the health posts, clinics, and health centers lacked qualified staff
and medical supplies to provide their intended services. For proper medical services,
people had to travel over 100 km to reach Yabello. There was one lower primary
school (1–4th grade) in each kebele, and upper primary schools (1–8th grade) in
some kebeles. Alona had schools of both kinds.
Livestock production contributed over 90 % of the total household income. In the
entire district there were 157,000 cattle, 58,462 sheep, 80,252 goats, 35,165 camels,
18 horses, 1,403 mules, 5,742 donkeys and 11,980 chickens (CSA 2007). In Alona
there were about 11,754 head of cattle, 4,900 sheep, 5,220 goats, 4,251 camels, 98
mules, 314 donkeys, and 135 chickens. The major constraints on livestock
production are drought, feed shortages, diseases, and market access. In spite of
all the problems livestock population has grown beyond the carrying capacity of
the area, leading to land degradation and deterioration of critical resources like
pasture and water.
Some households supplement income through crop production (maize), trade,
mining, and employment as day laborers, especially among the poor. According to
Boku (2008), 75 % of the households that engaged in crop cultivation belonged to
378 T.W. Gole et al.

Table 22.4 Locally determined attributes of wealth in the kebele of Alona, Arero district, counting
the total number of cattle is culturally taboo in the Borana communities, where only the number of
dairy cows is counted
Rich Middle-income Poor
Owns a house with a corrugated May or may not own a Owns one small house with a
steel roof in town house with a corrugated grass roof in a rural village
Food self-sufficient steel roof in town Not food self-sufficient
Owns more than 50 dairy cows Food self-sufficient Owns 5–30 dairy cows
Capable of sending children to Owns 30–50 dairy cows (someone who does
school and covering all costs Capable of sending not own cows is referred
Capable of saving money at credit children to school to as “Qollee,” meaning
and saving institutions covering some costs absolutely or extremely
Able to cover health care costs Capable of saving poor)
Owns a mix of livestock to some extent, Unable to send children
(cattle, oxen, goats, sheep, and also of helping to school
camels, donkeys, horses, other households that Unable to save money, either
and mules) suffer shocks or deaths assisted by clan or aid
Capable of helping poor Owns a mix of livestock May own a mix of livestock
neighbors and other (cattle, oxen, goats, Dependent on support
households in all aspects sheep, camels, and of others, aid, or work
(providing money, milk, donkeys) as a day laborer
materials, and grain) in case Self-sufficient in every
of funeral or food shortages aspect

the customary economic category of “self-reliant.” Trade mainly involved livestock,

livestock products, consumable goods, and grain. Small-scale mining of gold and
other minerals by otherwise unemployed youths also occurs locally, a livelihood
that is being learned from immigrant miners. There is significant variability in the
wealth status among community members. A traditional local indicator of wealth is
the number of livestock. A few individuals had large herds of cattle, but the majority
of local residents had few or none.
The stories of some of the poor households illustrate how drought has the ability
to impoverish people in Borana. In 2010–2011, many households lost all of their
livestock due to 10 months of drought. It was evident from the stories of the “rich”
and “middle-income” category household heads that income diversification helped
them cope with vulnerability in a marginal environment. In addition to traditional
livelihoods based on livestock production, the creation of alternative businesses or
employment opportunities could contribute to poverty reduction.
Aside from the naturally marginal environmental conditions, culture also played
an important role in asset accumulation of individuals in Borana. The strong
cultural value attached to livestock ownership and the reluctance to sell livestock
when the animals are in good condition during the season when prices are highest
was the main barrier to reducing herds to more sustainable levels. Food shortages
for people coincided with the dry season, when there were also feed and water
shortages for livestock. This means that the physical condition of livestock suffers
during the dry season, reducing their market value (Table 22.4).
22 The Marginalized and Poorest in Different Communities and Settings of Ethiopia 379

Box 22.4 Narrative of Mr. Doyo Guyo

Doyo Guyo was an 85 year-old man with 17 children from 3 wives, of
which 11 were married, and 6 were living with the family. At the time of
the interview he was the head of a household of ten living in a multiple-
room house. He was also the head of his community. Mr. Guyo owned
more than 50 dairy cows, 21 goats, 10 sheep, a donkey, 10 chickens, and a
mule. In addition Mr. Guyo owned a home with a corrugated steel roof in
Alona and 0.5 ha of cropland for maize production. He harvested around
1,000 kg of grain for food and the cornstalks served as animal feed. The
household did not produce enough food for the family and Mr. Guyo aug-
mented his income by selling livestock. He was not receiving any aid from
the government or NGOs. He had lost his parents and all of his livestock
to a drought long time ago and afterwards had to take care of his brothers
and sisters. He claimed to have earned his assets through hard work. Most
of his cattle were taken by members of the Guji and Garba clans during a
conflict in 1992. At that point he received support from his clan through
the traditional Buusaa-Gonofaa mutual support system of the Borana
Oromo. He also earned additional income working as a day laborer, sell-
ing fuel wood, and locally trading tobacco, cattle, and consumable goods.
He was not saving money since there was no culture of saving like the
Iqub in the area. He started cultivating land to reduce his dependency on
livestock. As a future strategy, he planned to cultivate more land for sub-
sistence using improved agricultural techniques to increase production, to
reduce the number of livestock he owned, to open a savings account at a
bank, to purchase a house with a corrugated steel roof in Yabello, to har-
vest and store hay for the dry season, and to seek other means to diversify
his income.

The entrenched cultural practice in Borana of providing help to the poor by

rich individuals, as well as a collective self-help system called “Buusa-Gonofaa”
are excellent concepts for reducing abject poverty. In most of the areas studied
in Ethiopia, the “rich” community members sometimes took advantage of
(and were partly responsible for) the poverty of others. In Borana, however,
the wealthier families help the poor without charging interest or seeking other
economic gain. Similarly, social-oriented businesses could help the poor to build
assets based on locally available resources and opportunities. The story of Mr. Guyo
(Box 22.4) provides evidence of how the community-based coping systems like
Buusaa-Gonofaa and individual efforts help residents recover from natural and
human-induced shocks.
380 T.W. Gole et al.

22.3 Conclusion

We found that poverty continues to be widespread in Ethiopia, even in areas

endowed with a broad range of natural resources and where the cultivation of cash
crops is possible. The extent of poverty and the causes and nature of marginality of
the poor vary from place to place, depending on the resource base, livelihood
strategies, and the climatic, agro-ecological, and socio-cultural conditions. In the
Jida district in the central highlands, extreme land degradation on slopes, erratic
rainfall, and unsustainable forestry practices on the valley plains have contributed to
the marginality of local households. Environmental degradation was often partially
caused by over-exploitation of resources. As a consequence, people suffer from
frequent crop failures, poverty, and hunger. Exploitation of the poor by wealthier
individuals through money lending at high interest rates worsens the situation of
many of the poor. As a result, they need to migrate seasonally to urban areas in
search of employment opportunities. Marginalization here seems to be predominantly
caused by anthropogenic environmental factors.
Contrary to Jida, the Yayu district is endowed with plentiful natural resources.
The region offers immense agricultural potential, which has attracted immigrants
from other parts of the country. The lack of formal credit facilities, the practice
of providing high-interest personal loans, shortages of family labor, and the lim-
ited size of coffee plantations all contributed to the poverty of some community
members. In addition, the lack of regulatory institutions for enforcing compli-
ance with share cropping agreements exposed the poor and elderly to exploita-
tion. Thus, marginalization here seemed to be mainly rooted in institutional
factors.
In the Adami Tullu Jiddo Kombolcha district the potential for income generation
depends on the possession of land with potential for irrigated agriculture and
livestock production. The proximity of the district to major domestic markets has
attracted investment in commercial agriculture, displacing local residents and
leaving them with only marginal production areas. The proportion of landless
households in the district is very high compared to other rural areas in Ethiopia.
Here, economic factors in the form of agricultural investments and related factors
such as access to irrigation water are the main causes for marginality.
On the other hand, degraded rangeland and water resources, cultural values
attached to livestock, poor range management and livestock production practices,
and marginal environmental conditions were all factors contributing to poverty in
the pastoralist community of Borana. Although drought is a natural calamity, the
high social value given to cattle has inhibited the practice of accumulating financial
savings, which would help households cope with adverse conditions during seasonal
droughts and reduce pressure on grazing areas and surface water resources. The
traditional self-help community institution of Buusaa-Gonofa among the Borana
pastoralists was an important coping mechanism for incidences of drought and
human-induced disasters. As a result, ecological and spatial factors are the main
causes for marginality experienced by people in this area.
22 The Marginalized and Poorest in Different Communities and Settings of Ethiopia 381

Open Access This chapter is distributed under the terms of the Creative Commons Attribution
Noncommercial License, which permits any noncommercial use, distribution, and reproduction in
any medium, provided the original author(s) and source are credited.

References

Boku TD (2008) Pastoralism under stress: resources, institutions and poverty among the Borana
Oromo in Southern Ethiopia. Dissertation, Norwegian University of Life Sciences, Aas
CSA (2007) Summary and statistical report of the 2007 population and housing census results.
Central Statistical Agency of Ethiopia, Addis Ababa. http://www.csa.gov.et/pdf/Cen2007_
prelimineray.pdf. Accessed 20 Sept 2012
Fite GI (2008) Remote sensing and GIS assisted participatory biosphere reserve zoning for wild
coffee conservation; case of Yayu forest. Thesis, Addis Ababa University, Addis Ababa
Gatzweiler F, Baumüller H, von Braun J, Ladenburger C (2011) Marginality: addressing the root
causes of extreme poverty. ZEF working paper no. 77, Center for Development Research,
University of Bonn, Bonn
Gole TW (2003) Vegetation of the Yayu forest in SW Ethiopia: impacts of human use and implications
for in situ conservation of wild Coffea arabica L. populations. Cuvillier Verlag, Göttingen
Gole TW, Feyera S, Kassahun T, Fite G (2009) Yayu coffee forest biosphere reserve nomination
form. Ethiopian MAB National Committee, Addis Ababa
Halake D (2010) The impacts of development interventions on customary institutions of forest
resource management among the Borana Oromo Of southern Ethiopia. Thesis, Addis Ababa
University, Addis Ababa
UNESCO (2010) World Network of Biosphere Reserves 2010: sites for sustainable development.
United Nations Education, Science, and Cultural Organization, Paris
Index

A Bangladesh, 7, 9, 13, 19, 73, 77, 80, 94, 95,

Abebaw, D., 9 101–114, 170, 171, 175, 176, 257–265,
Acemoglu, D., 29 316, 318, 320, 322, 332, 334, 343, 344
Adaptability, 17, 61–63 Bangladesh Rural Advancement Committee
Adebaw, D., 13 (BRAC), 13, 14, 41, 259–265, 332
Admassie, A., 9, 13 Barbier, E., 190
Agricultural productivity, 10, 34, 81, 151–155, Baumüller, H., 5, 15
157–166 Beneficiary targeting, 102, 111–114
Agriculture, 8, 32, 61, 72, 117, 151, 170, Besley, T., 98
183–184, 206, 225, 241, 258, 269, Bhalla, A., 29
320, 333, 356, 366 Blakely, E.J., 29
Agro-ecological zone, 152, 153, 155, 159, Blowfield, M., 335, 336
160, 274 Boku, T.D., 377–378
Ahmad, S.E., 19 Bottom of the pyramid, 15, 332, 334, 337–339
Ahmed, A.U., 1, 7, 13, 14, 86, 222 Bourguignon, F., 97
Ahmed, S.M., 19 BRAC. See Bangladesh Rural Advancement
Aid, 2, 14, 71, 96, 163, 227, 251, 259, 273, Committee (BRAC)
278, 304, 312, 318–320, 325, 326, Bradshaw, T.K., 29
332, 333, 335, 357, 376, 379 Brodwin, P., 33
Amdissa, T., 275 Brousseau, E., 46
Area prioritization, 102, 104, 107–110 Bruner, J., 136
Asset ownership, 95, 118, 120, 121, 125–129 Budget constraints, 9, 191, 305–311
Assets, 3–5, 7, 8, 17, 30, 35, 52, 60, 70, 86, Bulte, E.H., 191, 198
95, 112, 118, 119, 121, 125, 128–129, Burgess, R., 98
174, 176, 190, 205, 217, 218, 250, Business approaches, 2, 15–16, 331–349, 354
258–260, 262, 263, 275, 277, 281,
304, 312, 332, 369, 376, 378, 379
Attachment, 10, 133–145 C
Attributes of wealth, 369, 373, 375, 378 Callo-Concha, D., 6
Azomahou, T., 191 Capability, 4, 15, 141, 143, 144, 174, 206,
247, 275, 317, 359, 361, 374, 375
Capability approach, 4, 33
B Carlucci, M., 192
Bai, Z.G., 195 Carrying capacity, 58, 377
Balmford, A., 182 Causal complexities, 27–37
Banerjee, B., 339 Causal relationships, 37, 70, 185, 198

J. von Braun and F.W. Gatzweiler (eds.), Marginality: Addressing the Nexus 383
of Poverty, Exclusion and Ecology, DOI 10.1007/978-94-007-7061-4,
© The Author(s) 2014
384 Index

Caviglia-Harris, J.L., 191 program, 8, 13, 42, 102, 104, 228, 234,
Chen, S., 208 235, 259, 278, 297, 319–320, 326, 333
China, 2, 12, 13, 19, 89, 90, 92, 93, 239–253, strategies, 14, 117–118, 152, 156–158,
353, 357 164, 270, 272, 315, 316
Choudhury, N.Y., 9 theory, 31
Christiansen, N., 15, 19 Devereux, S., 276, 320
Civil society, 2, 14–15, 20, 173, 297, Diamond, J., 29
319, 335, 344 Diao, X., 285, 287
Classes of poverty theory, 5, 29 Di Bao program, 12, 242–248, 251–253
Coffee, 16–17, 291, 357, 366, 371–374, 380 Dietary calorie intake, 118, 120, 122–124
Collier, P., 75 Disabilities, 119, 242, 245, 282, 296, 324
Community-based approach, 234, 363 Disaster risk reduction, 9, 104–111, 113
Community nutrition, 104–112 Discount rate, 185, 186, 190
Competition, 18, 19, 41, 54, 58, 59, 61, 119, Discrimination, 2, 4, 11, 17–18, 119, 142,
164, 198, 292, 309–312, 323, 333, 354 143, 207, 209, 217, 271, 280–282,
Complex systems, 17, 62 294, 296, 332
Conceptual framework, 5, 28, 31, 34–37, 60, Disenfranchisement, 10, 142
169–170, 183–186 Distance, 6, 10, 18, 31, 32, 51–53, 80–81, 119,
Consumption behavior, 12, 221–235 129, 173, 296
Consumption expenditure, 102, 118, 120, Duflo, E., 339
122–128, 212, 231 Dunne, R.J., 30
Contract, 6, 42, 45–49, 54–56, 134, 194, 195, Duraiappah, A.K., 29
328, 345, 372 Dynamic systems, 57
Cooley, C.H., 143
Corporate social responsibility (CSR),
2, 333–337, 345, 346, 348 E
Correlate, 75, 117–129, 190 Earmarked transfers, 309
Creating Shared Value (CSV), 15, 340–341, Ecology, 4, 10, 31, 57–59, 62, 272, 274
353–364 Economic growth, 1, 6, 9, 12, 17, 18, 43,
Credit, 16, 94, 119, 273, 276, 279, 282, 295, 70, 75, 89, 117, 118, 156, 162–163,
307, 319, 320, 326, 332, 356, 363–364, 187–188, 191, 192, 207–209, 212,
375, 378, 380 215, 241, 257–258, 270, 272, 278,
CSR. See Corporate social 283, 285, 291–292, 297, 316, 317,
responsibility (CSR) 320, 323, 324
CSV. See Creating Shared Value (CSV) Economic policy, 270–282
Cut-off point, 71–73, 75, 80, 88 Ecosystem services, 10, 11, 18, 63, 152, 157,
165, 166, 169–178, 181–186, 189
Education, 7, 9, 12–15, 31, 32, 34, 47–50, 54,
D 60, 86, 94–95, 97, 104, 106–108,
Dairy, 12, 15, 217, 223, 224, 226, 228–231, 113, 119–122, 125, 127–129, 142, 169,
233, 234, 258–263, 376–379 205–207, 209, 217, 218, 222, 226–227,
Dasgupta, P., 4, 29 231, 233–235, 241, 242, 244, 245, 247,
Decentralization, 14, 19, 271, 273, 277, 303–313 251, 252, 273, 274, 277, 279–284,
Definition of marginality, 3, 30, 31 287–290, 292–297, 304–307, 309, 317,
Dellink, R.B., 29 319, 327, 333, 342, 343, 347, 356, 357,
Demessie, S., 295 360, 361, 373
Desai, R.M., 29 Education sector policy, 13, 277
Desalegn, R., 276 EKC. See Environmental Kuznets curve (EKC)
Determinants, 2, 10, 11, 31–32, 34, 118, 208, Elbers, C., 103
227, 229, 336 Elites, 29, 234, 316, 318, 319, 325
Development, 2, 27, 41, 70, 102, 117, 133, Elkington, J., 342
151, 170, 183, 206, 223, 240, 258, Emotions, 10, 135–137, 140–144
269, 309, 315, 332, 354, 366 Empowerment, 4, 144, 197, 235, 258, 273,
practice, 6, 41–56 281, 340
Index 385

Environment, 2, 28, 44, 58, 80, 118, 133, Goldberg, M., 33

151, 170, 182, 225, 259, 270, 317, Goldsmith, W.W., 29
333, 358, 366 Gole, T.W., 16
Environmental degradation, 133, 152, 164, Good governance, 14, 313
165, 172–175, 178, 190, 191, 198, 380 Governments, 2, 32, 43, 61, 69, 94,
Environmental Kuznets curve (EKC), 102, 117, 160, 171, 195, 206,
191, 192, 198 222, 240, 258, 270, 304, 316,
Equalization transfers, 309 332, 365
Ethiopia, 7, 36, 77, 92, 117, 269, 365 Gradl, C., 340, 345
Exclusion, 2, 31, 41, 60, 97, 102, 118, 205, 234, Grameen, 258, 343, 344
242, 259 Grant, U., 8
Expenditure, 12, 80, 94, 101, 103, 118, 120, Graw, V., 7
122–129, 211, 212, 222–224, 227, Gunderson, L.H., 62
229–233, 287–289, 296, 297, 317, Gurung, G.S., 118, 120
325–326 Gwartney, J., 29
Extreme poor, 70, 86–88, 93, 110

H
F Hagedorn, K., 35
Ferenczi, S., 140 Hamilton, D., 29
Fiscal policy, 14 Hammond, A.L., 337, 338
Food Haque, E.E., 171
expenditure, 223 Hart, S.L., 337
security, 12, 13, 17, 63, 105–106, 112, 117, Hawking, S., 43
163, 171, 177, 185, 188, 225, 231–233, Health, 6, 34, 43, 59, 73, 94, 118, 133, 169,
241, 252, 273, 275, 278, 320, 339 182, 221, 222, 242, 258, 270, 304,
subsidies, 222, 231 317, 333, 354, 371
Food security programme (FSP), Health sector policy (HSP), 277
117, 231–233, 274–276, 279 Helliwell, J.F., 29
Freedom, 3, 4, 6, 10, 118, 144, 170, 271 Herrnstein, R.J., 29
Freud, S., 141, 142 Hickey, S., 319, 321
Frynas, J.G., 335, 336 Holling, C.S., 62
FSP. See Food security programme (FSP) Household variables, 121
Functioning, 4, 33–34, 141, 191 HSP. See Health sector policy (HSP)
Hulme, D., 14, 15
Hunger, 7, 9, 73, 86, 95, 103, 156, 159,
G 163, 172, 206, 221–222, 252,
Gallup, J.L., 32 353, 380
Gatzweiler, F., 5, 30, 31, 34, 63, 70, 118, Husmann, C., 7
332, 366 Hutton, D., 171
Gender, 12, 20, 31, 33, 34, 44, 119, 121, 127,
128, 133, 135, 142, 227, 235, 264,
270–272, 277, 280–282, 294, 295, 297 I
Geographical information systems (GIS), Idemudia, U., 336
7, 70, 74, 75, 107, 187 Identity, 10, 33, 62, 136, 138, 143, 144,
Geographic targeting, 103, 104, 106–111 206, 260
Geography, 4, 17–18, 32, 270 Incentives structures, 305–311
Gerber, N., 11 Inclusive business, 2, 15, 16, 334, 339–340,
Gilligan, D.O., 275 345–348
GIS. See Geographical information Income generation, 17, 63, 327
systems (GIS) India, 8, 11, 15, 80, 92–95, 134, 170, 171, 175,
Glachant, J.M., 46 176, 192, 193, 196–198, 206, 209–215,
Global, 6, 33, 60, 70, 87, 125, 154, 173, 182, 217, 304, 317, 318, 326, 333, 334, 339,
222, 240, 272, 316, 334, 353, 365 353, 357–363
386 Index

Indicator, 5, 7, 10, 13, 18, 32, 36–37, 70–75, Land management, 11, 183, 184, 186, 189, 190,
81, 86, 94, 95, 103, 104, 107, 108, 118, 192–198, 276
120–122, 126, 129, 178, 186–188, Lapeyre, F., 29
190–192, 198, 223, 232, 247–248, 253, LD. See Land degradation (LD)
283, 284, 286–290, 294, 347, 369, 378 Lear, J., 135
Indonesia, 2, 8, 12, 92, 221–235, 316, 326, 357 Levy, S., 15
Inhibition, 141–142 Lewis, H.B., 144
Institutional innovation, 243–247 Lewis, O., 29
Institutions, 3–5, 11, 17–21, 34–36, 74, 75, Linkages, 3, 4, 10, 32, 42, 69, 81, 108, 170,
118, 145, 162–166, 183, 185, 191, 175, 178, 189, 190, 195, 198, 206, 207,
194, 195, 197, 198, 242–247, 251, 215, 253, 272, 276, 285, 306, 307, 311
253, 259, 271, 277, 278, 280–282, Livelihood, 3, 35, 60, 95, 114, 119, 170, 182,
304, 305, 309, 313, 325, 326, 336, 206, 225, 241, 259, 271, 333, 366
340, 346, 349, 370, 380 categories, 210–211, 215, 216
Interdisciplinary, 4, 10, 20, 62 strategy, 366, 380
Investments, 2, 11, 12, 15–18, 36, 97, Livestock, 16, 17, 121, 129, 171, 176, 194,
151–152, 154, 156–160, 162–166, 173, 195, 217, 259, 274, 278, 279, 284, 287,
184–186, 189, 190, 195, 197, 198, 212, 290–292, 356, 361–362, 366–370,
241, 242, 252, 253, 276, 279, 309, 316, 373–375, 377–380
320, 323–325, 327, 331, 333, 342, 344, Local
347, 348, 354–357, 361, 363, 368, 374, government, 14, 112, 114, 164, 228, 240,
376, 380 241, 244–246, 248, 253, 304–313, 345
Irrigation, 17, 165, 185, 192, 196, 198, 273, knowledge, 261, 265
274, 278, 279, 287, 292, 361, 367, support, 318–323
373–374, 376, 380

M
J Mäler, K.-G., 29
Jencks, C., 29 Malnutrition, 72, 73, 97, 108, 118, 221, 223,
Jenkins, B., 340 225, 324, 353, 356, 364
Jensen, L., 29 Mapping, 7, 9, 44, 69–81, 103–104, 261,
264, 368
Marginality, 1, 28, 42, 69, 102, 118, 152,
K 181, 205, 239, 270, 280, 303, 315,
Kant, E., 28 331, 353, 366
Kapp, K.W., 29 dimension, 7, 70, 75, 78, 79
Karnani, A., 339 hotspot, 7, 69–81
Kaufmann, D., 74 Market, 2, 42, 73, 86, 118, 151, 171, 183, 206,
Kirkby, A., 119 222, 240, 270, 312, 317, 331, 354, 366
Kleinman, A., 144 Marx, K., 29
Knobloch, C., 340, 345 Mauss, M., 46
Kollmair, M., 118 Mbakogu, I.A., 29
Kontoleon, A., 177 McCaleb, T.S., 29
Koohi-Kamali, F., 295 McGregor, J.A., 183
Kramer, M.R., 331, 340, 341 Mehretu, A., 118, 119
Krugman, P., 32 Milanovic, B., 98
Kumar, M., 10 Mlodinow, L., 43
Kuznets, S., 207 Moga factory, 358–361
Moscovici, S., 136
Moynahan, D., 29
L Muller, B., 29
Land-cover change, 187, 192 Munemo, N., 320
Land degradation (LD), 11, 18, 153, Murray, C., 29
181–199, 275, 377 Mwangi, E., 19
Index 387

N Policy, 1–21, 96, 117, 119, 152, 158, 162–166,

Nachtergaele, F., 182 171, 172, 178, 195, 212, 215–218,
Natural resources, 4, 10, 16, 18, 20, 61, 62, 74, 221–235, 240–242, 244–246, 251–253,
152, 157–158, 165, 166, 170–177, 189, 270, 272, 273, 276–282, 289, 292, 294,
191, 195, 197, 198, 271, 274, 276, 277, 296–297, 311, 313, 316, 318, 321, 323,
292, 308, 337–338, 344, 366, 368, 373, 326, 327, 336, 375–376
377, 380 environment, 152, 154–156, 159, 164,
Nelson, A., 72, 73 270–282
Nestlé, 15, 341, 343, 354–363 intervention, 125, 152, 156, 163, 165, 167,
NGOs. See Non-governmental 222, 253, 270, 281
organization (NGOs) Political economy, 3, 29–31, 44, 45, 55, 70,
Nicholls, A., 341 118, 119, 170, 269–282, 292, 307,
Niger, 77, 92, 192–198, 284, 336 308, 311–312, 332, 336
Nkonya, E., 189, 190, 193, 198 Poorest, 2, 7–9, 12–15, 70, 85–98, 101–114,
Nkonya, I., 18, 187, 191, 193, 194 118, 154, 171–172, 174–175, 192, 208,
Non-governmental organization (NGOs), 212, 218, 221–235, 241, 248, 257–265,
13, 20, 43, 107, 112, 113, 118, 173, 283, 309, 316, 319, 321, 322, 324–327,
195, 197, 258, 277, 283, 292, 294, 339, 345–347, 365–380
297, 305, 335, 336, 365, 371, 379 Population density, 59, 61, 73, 75, 151, 152,
North, D.C., 29 156, 158, 166, 183–184, 187–188, 195
Nussbaum, M., 29, 135 Porter, M.E., 331, 340, 341
Nutrition, 5, 8, 9, 12, 15, 31, 73, 105–112, Position, 3–5, 13, 28, 30, 31, 33–36, 70, 118,
120, 141, 222, 223, 226, 229, 231, 233, 222, 242, 283, 297, 318, 320, 332, 333,
235, 252, 277, 281, 322, 333, 342, 344, 346, 354
348, 357, 360, 362, 363 Poverty, 1, 27, 41, 61, 70, 85, 101, 117, 133,
151, 169, 182, 205, 222, 239, 258, 269,
316, 331, 353, 365
O alleviation, 170–173, 175, 178, 206, 210,
O’Connor, A., 29 258, 264, 283, 291–292, 321, 340–341
Oportunidades, 19, 312, 321–323 reduction, 2, 7–9, 12–14, 18, 21, 36, 43,
Opportunity cost, 10, 18, 50, 152, 156, 157, 86, 89–93, 96, 102, 118, 154–158, 163,
166, 295 194, 205–218, 239–243, 247–253, 281,
Orphans, 119, 152, 158, 159, 162, 250, 278, 283–290, 297, 316, 317, 319, 320, 322,
282, 320 324, 337, 347, 353, 358, 363, 365, 378
Osmani, S., 208 trap, 7, 17, 34, 62, 89–91, 96, 190, 253, 292,
Ostrom, E., 35 320, 325
Prahalad, C.K., 331, 332, 337–339
Price elasticity, 12, 230, 231
P Principles, 15, 42, 44–45, 55, 61–63, 225, 242,
Pakistan, 8, 15, 94, 170, 175, 176, 304, 277, 280, 308, 312, 327–328, 341–343
306–309, 312, 357–363 Priority setting, 152
Pangaribowo, E.H., 12 Private sector, 2, 43, 118, 151, 160, 162, 163,
Park, R.E., 33 274, 333, 334, 340–341, 363–364
Participatory mapping, 264, 368 Probit model, 126, 129
Participatory process, 206, 260, 264, 265 Productive safety net program (PSNP),
Participatory wealth ranking (PWR), 270, 275–276, 287, 365
261–265 Pro-poor growth, 207, 208, 212, 213, 215–217
Pastoralist, 17, 60, 128–129, 271, 272, Proximate cause, 8, 11, 183, 184
278, 279, 287, 291–293, 297, PSNP. See Productive safety net program (PSNP)
377, 380 Public services, 8, 12–14, 94, 242–247, 251,
Patterson, K.P., 271 252, 271, 275, 296–297, 305, 308
Periphery, 31, 33, 278–280 Punjab, 15, 358–363
Pingali, P.L., 10, 18, 29 Putnam, R., 29
Polanyi, K., 29, 317, 321 PWR. See Participatory wealth ranking (PWR)
388 Index

R Socially inclusive policy, 215–218

Räder, C., 9 Society, 2, 3, 6, 8, 13–15, 20, 33, 42, 47, 51,
Rainwater, L., 29 53, 55, 56, 61, 63, 70, 85, 89, 143, 144,
Ravallion, M., 207, 208 173, 183, 185–186, 189, 209, 227, 232,
Reardon, T., 190 241, 248, 270, 278, 282, 297, 317, 319,
Regional inequality, 12 333–337, 341, 343, 344, 354, 355, 366
Remoteness, 8, 60, 94, 152, 332 Socio-cultural context, 9–10, 28
Requier-Desjardins, M., 186 Socio-ecological systems, 6, 19, 35, 59–63, 81
Resettlement, 13, 274–276, 278, 279, 292 Sociology, 31, 44
Resilience, 2, 17, 60–63, 105–106, 162, 171 Socio-religious group, 206, 209–218
Rhee, S.S., 45 South Asia (SA), 3, 6, 9, 70, 77, 86–89, 96,
Robinson, J., 29 154, 169–178
Roland, G., 29 Sphere of life, 45, 70, 80, 81
Ruijs, A., 29 Spreckley, F., 342
Rural SSA. See Sub-Saharan Africa (SSA)
development, 15–16, 249, 297, Stonequist, E.V., 33
356, 357, 363 Subjacent poor, 7, 88, 91, 96
factories, 358 Sub-Saharan Africa (SSA), 3, 6, 8–10, 18, 70,
poverty, 15, 171, 210, 212, 248, 251, 73, 75, 77, 80, 87–90, 95, 96, 151–167,
269–297, 353–364 169–178, 194, 196
Ryan, W., 29 Summer, H., 183
Systems, 3, 28, 42, 57, 70, 107, 118, 137, 152,
171, 183, 209, 222, 240, 270, 304, 318,
S 332, 357, 366
Sabates-Wheeler, R., 276
Safety nets, 13, 42, 102, 104–107, 110–114,
135, 174, 177, 225, 242, 259, 260, 274, T
287, 316, 376 Tabor, S.R., 232
Salvati, L., 192 Targeting, 2, 42, 87, 101, 162, 186, 209, 232,
Sawit, M.H., 232 240, 260, 273, 304, 318, 332, 363
School feeding, 104, 106–111 Taxation, 16, 223, 307–309, 322, 325, 336,
Segmentation, 53–55 348, 358, 363
Sen, A., 29, 144, 206 Technology priorities, 160–162
Shame, 10, 133–145 Tewodaj, M., 289
Share cropping, 370, 372–374, 380 Thompson, R., 356
Shared taxes, 308 Thorat, S., 11, 19
Shareholder, 15, 335, 341, 354, 355 Tobin, J., 29
Shaw, W., 29 Transfers, 2, 6, 14, 15, 19, 42, 46, 55, 104,
Shepherd, A, 8 111, 114, 143, 227, 233, 247, 249, 250,
Shmueli-Goetz, Y., 137 252, 253, 259–261, 263, 274, 275,
Small, M.L., 29 304–305, 308–309, 311, 312, 315–328,
Smith, A., 29 362–364
Social Trauma, 10, 133–145
business, 334, 343–344, 348 Trust, 15, 41, 45, 142, 310, 316, 325–326,
entrepreneurship, 341–344 354–355, 358, 360
insurance, 12, 13, 242–247, 252 Tsakloglou, P., 207
marginality, 10, 31, 33, 34, 59, 60, 63, 134, Tucker, C.J., 188
143, 206
networks, 227–228
protection, 1–2, 14, 15, 197–198, 233, U
240–243, 247, 252, 312, 315–328 Uchida, H., 73
relief, 241, 247, 252 Ultra-poor, 7, 13, 88, 91–97, 119, 127–129,
safety net, 13, 104–106, 111–114, 174, 154, 258–264, 317, 319, 327, 332
225, 242, 376 Ultra-poverty, 7, 9, 88–92, 95–98, 118, 121,
value, 135, 332–344, 347, 380 127, 129, 154, 296
Index 389

Unconditional cash transfer, 14, 227, 233, Weber, M., 29

316–317 Weiskopf, D.A., 28
Underlying cause, 11, 118–119, 173, 183, 184, Welfare programs, 320
188, 296 Well-being, 3–5, 10–11, 15, 18, 31, 33, 35, 36,
Urban poverty, 92, 214–215 59–61, 70, 71, 89, 118–122, 126,
Utting, P., 336 142–145, 169, 171–175, 177, 178, 182,
183, 189, 198, 199, 206, 221, 231, 247,
270, 274, 275, 280, 285, 289, 297, 339
V WFP. See World Food Programme (WFP)
Valentine, C.A., 29 Workneh, N., 281
van Soest, D.P., 191, 198 World Food Programme (WFP), 102–114
Vatn, A., 35 Wright, R.D., 33
Vira, B., 177 Wright, S.N., 33
von Braun, J., 29
von Thünen, J.H., 29, 32
Vosti, S., 190 Y
Vulnerability, 2, 21, 63, 102, 104, 108, Yashiro, M., 10
169–172, 177, 181–182, 192, 258, Yield gaps, 159–160, 162
260–261, 278, 279, 281, 378 Yitbark, T., 295
Young, A., 144
Yunus, M., 331, 334, 343, 344
W
Washington Consensus, 317, 324
Wealth-ranking, 368 Z
Weber, B., 29 Zhu, L., 12, 13, 19