Made in Africa: Industrial Policy in Ethiopia

Arkebe Oqubay
Print publication date: 2015
Print ISBN-13: 9780198739890
Published to Oxford Scholarship Online: August 2015
DOI: 10.1093/acprof:oso/9780198739890.001.0001

Introduction to Industrial Policy in Ethiopia

Chapter:
(p.1) 1 Introduction to Industrial Policy in Ethiopia
Source:
Made in Africa
Author(s):

Arkebe Oqubay
Publisher:
Oxford University Press

DOI:10.1093/acprof:oso/9780198739890.003.0001

Abstract and Keywords

Ethiopia is a low-income country, landlocked, and located in a ‘bad neighbourhood’,
and has the second largest population in Africa, with all the challenges and
opportunities that entails. It has recorded one of the fastest economic growth rates in
the early twenty-first century. In obvious ways, Ethiopia is a clear example of the
‘African Renaissance’, having undergone dramatic political and economic
transformation in the past twenty-five years. This is the context within which research
behind the book was carried out, drawing on quantitative data (primary and
secondary) and on qualitative interviews. The chapter introduces the reader to
industrial policymaking in Ethiopia, and the research design and methodology
required to understand the policymaking process. The methodology draws heavily on
structuralist development economics and political economy.
Keywords: Ethiopia, African Renaissance, economic growth, structural
transformation, data, structuralist development economics, political economy

1.1 The Lion awakes

In March 2014, Time magazine published an article entitled ‘Forget the BRICS: Meet
the PINEs’. PINE is an acronym for the Philippines, Indonesia, Nigeria, and Ethiopia,
accounting for some 600 million people. The author notes that for the last fifty years
Africa has generally stood on the sidelines as Asia and others in the developing world
have made enormous welfare gains. Now, at last, the continent is beginning to make
gains. And nowhere is this truer than in Ethiopia. Once synonymous with
impoverishment, the country has enjoyed strong management and may be on a new
course. The author concludes by musing whether we are not seeing the emergence of
Lion economies in Africa, the analogue of Asia’s Tigers of the late twentieth century.
In the same month, the African Union (AU) and UN Economic Commission for
Africa (UNECA) launched a joint annual report at a conference of African finance
ministers in Abuja, Nigeria.1 The ministers and those attending the conference, and
the authors of the report, showed strong interest in Ethiopia’s attempt at instituting an
activist industrial policy.
Ethiopia is Africa’s second most populous country and occupies a highly sensitive
geopolitical position, and its economic performance does indeed deserve attention.
The country has made spectacular leaps on multiple development fronts in recent
years. For instance, Ethiopia—whose economy is not mineral dependent—has
recorded double-digit economic growth for a decade, quadrupling its gross domestic
product (GDP) per (p.2) capita.2 Life expectancy increased by fifteen years from
forty-eight years in 1992 to sixty-three years in 2012, and poverty was halved in two
decades.3 This much-discussed performance is partly the result of the country’s
distinctive development path and bold experiment in industrial policy, which is unlike
any other in Africa. Few other countries on the continent have even tried to
implement an active ‘developmentalist’ industrial policy.
Despite hiccoughs and ongoing challenges, Ethiopia’s momentum seemingly persists.
Indeed, in March 2014, the Ethiopian prime minister inaugurated a new winery and
vineyard at Ziway, a rapidly transforming town 160 kilometres south of the capital.
The French company Castel, the largest wine producer in Europe, will export most of
the output of the new enterprise, and has committed to triple capacity. Of Castel’s 750
employees, only one is not Ethiopian. This company has also built three breweries
over twelve years, and has the largest share of the growing domestic market. At the
opening of the Castel winery, Prime Minister Haile Mariam declared his
government’s commitment to making Ethiopia the largest wine producer on the
continent within a decade, and promised to provide full support to industrialists to
this end.4 This was a serious affirmation, and the government’s record in delivering
on its promises has so far been good, as seen in the floriculture sector.
A short distance from the Castel winery lies the biggest concentration of flower farms
in the country. For instance, floriculture’s largest exporter, based in Ziway, employs
more than 10,000 workers. In less than a decade, Ethiopia has emerged as one of the
leading players in global floriculture. The sector has generated more than one billion
dollars in export earnings over seven years, a substantial amount for a new industry
by Ethiopian standards.5 This success is associated with the Ethiopian government’s
design and (p.3) execution of a successful industrial policy. Not far along the main
road to Addis Ababa, there are a number of new and established tanneries owned by
foreign and Ethiopian investors. Closer to Addis Ababa, the eye-catching first
industrial park built by a Chinese developer, the Eastern Industrial Zone, dominates a
landscape that includes many other warehouses, shops, and factories. All these lie
along a road that at the turn of the millennium ran through almost entirely
undeveloped country. Within twenty-four months, this industrial zone will have close
to 50,000 employees. The world’s largest manufacturer of women’s shoes, the
Huajian Group, started production in January 2012 and currently employs 3,500
people. In the near future, Huajian will build a special zone for footwear production
that will employ more than 30,000.6 Unilever, a leading European manufacturer, is
also establishing a factory in this industrial zone.
In the capital, near the ring road, the largest Turkish textile and garment plant
employing 8,000 workers is expanding. Many Turkish companies are relocating to
Ethiopia, and see the country as the best destination for their investment. According
to a representative of the Turkish Federation of Industrial Associations, ‘the Ethiopian
government not only invites investment, but also provides full support to the
firms’.7 Pittards, a firm from the United Kingdom (UK), established its first glove
factory in Ethiopia in January 2012 and exports gloves to Wells Lamont in the United
States of America (US), part of Warren Buffett’s holding company, and to Japan. In
two years, Pittards has expanded production to three factories. According to its young
Ethiopian general manager, worker skills and productivity are catching up to
international industry standards. In discussing the major obstacles, she highlighted
logistics and cumbersome customs procedures, shortage of foreign exchange, acute
insufficiency in local financing for expansion, power interruptions, and the supply
chain. A globally experienced Korean industrialist also recently pinpointed customs
procedures and the slow turnaround at the port of Djibouti as the single most
important constraints on doing business in Ethiopia. While many industrialists voice
these complaints, most are optimistic these issues will soon be addressed. Many
acknowledge the government’s efforts (p.4) in supporting industry, but some warn
that the constraints are choking manufacturing.
There are also dramatic changes in infrastructure under way—in power generation,
railways, and roads. The Renaissance Dam, the biggest hydropower project in Africa
(and the thirteenth or fourteenth largest in the world) and comparable to the Hoover
Dam in the US, is being built with finances entirely mobilized from local sources. It
is a reflection of and contributor to strong transformative economic growth under
tight macroeconomic constraints and a symbol of the government’s recognition of the
political imperative of transformation. It is also a rallying point for a form of
‘nationalism from above’ (Anderson 1991) in the context of federalism and an
ongoing ‘national’ issue (Markakis 2011). As the late Ethiopian Prime Minister Meles
Zenawi announced, ‘No matter how poor we are, in the Ethiopian tradition of resolve,
the Ethiopian people will pay any sacrifice.’ He continued: ‘I have no doubt they will,
with one voice, say: “Build the Dam!”’ This is just one example of how growth and
transformation as a shared national project have been crafted into a ‘rhetorical
commonplace’ (Jackson 2006), helping to shape and justify policy and resource
allocations. Moreover, Ethiopia began construction of the largest electric railway
network in Africa in 2010. In May 2014, a new toll-based six-lane Addis Ababa–
Adama expressway was opened to relieve the intense pressure on the old road to the
port at Djibouti.
As already noted, the country has become a destination for investment from Europe
and emerging economies, not only China and Turkey, but also India. Deborah
Brautigam, writing in The Guardian (30 December 2011), summed up differing
international perceptions of Ethiopia as follows: ‘To the Chinese, Ethiopia, with a fast
growing economy and 90 million consumers, looks like good business;…to the West,
Ethiopia typically conjures up images of drought and starving children; we want to
save Ethiopia.’ Brautigam adds that while ‘China sees Ethiopia as a land of business
opportunities…the African country remains in charge of any deals.’ Ethiopia’s rapid
economic growth and poverty reduction has attracted attention from friends, sceptics,
and staunch critics. For instance, a critique by The Economist Intelligence Unit (2012)
gives a careful picture of Ethiopia’s development path: despite the lack of
commercially exploitable hydrocarbons reserves found elsewhere on the continent,
Ethiopia has achieved export-driven economic growth rates that are the envy of other
African countries. It adds, ‘The economic results of this (p.5)state-led development
model have been impressive, and proponents of the difficult-to-define Beijing
Consensus have cited Ethiopia as a successful example…Ethiopia has grown more
quickly than almost any country on the continent while rejecting the advice from the
IMF and others to open up the economy quicker than it would like.’8
Although many recognize the recent achievements, few agree on the drivers of
growth. An Asian diplomat indicated that the secret has been that ‘first, Ethiopia
enjoys solid political stability. Second, Ethiopia has found its development path and
the government is committed to its implementation. Third, Ethiopia’s growing market
and big population has been a plus.’9 On the other hand, Western diplomats and
representatives of international finance institutions question Ethiopia’s path of
economic development, and argue that state-led development is not sustainable and
that the state is crowding out the private sector.10 They recommend slowing growth
and arresting public investment as well as rapidly liberalizing financial and
telecommunications markets. Ideology has been of no interest to foreign and local
industrialists though, and they are busily seizing new investment opportunities.
This rosy picture of a ‘rising Ethiopia’, however, reveals little about either the actual
details of the execution of industrial policy or the full spectrum of challenges faced
by low-income countries in general. It does not reflect the daunting task for such
countries in industrial catch up and economic transformation. Indeed, given the
extraordinary significance of Ethiopian policy and performance, especially in the
industrial sector, it is remarkable how little careful empirical work has been done on
them. One way to understand the successes of and challenges facing Ethiopia’s catch
up and transformation is through in-depth study of the leading industrial sub-sectors,
for instance, leather, garments, and floriculture, credited with the rapid transformation
of the economy over the past decade.

(p.6) 1.2 Has industrial policy’s moment arrived in Africa?

Given the dominance of mainstream economic thinking, debate on industrial policy
has remained a ‘restricted zone’ until very recently. In 2011, United Nations
Conference on Trade and Development (UNCTAD) and United Nations Industrial
Development Organization’s (UNIDO) annual report focused on Africa’s
industrialization, illustrating the challenges it faced. In 2012, a World Bank
publication, Light Manufacturing in Africa, was commissioned by the former chief
economist Justin Lin. This argued that industrialization is possible in Africa, but fell
short of supporting industrial policy as the main vehicle in catch up and economic
transformation. Peculiarly, it also failed to mention the necessity for an activist
‘developmental’ state. Two years later, UNECA’s economic report suggested that
perceptions on African industrialization had started to change, although it appears
that perspectives on industrial policies continue to diverge.
Although growth has been uneven among African countries and typically not
sustained, more African countries have in recent years experienced longer growth
spurts. In some countries, oil, gas, and mineral resources have driven growth and this
growth has most often not been inclusive or shared. This trend differs from the East
Asian experience of rapid industrialization and economic transformation. Scholars
and policymakers are concerned that Africa’s growth is insufficient to improve living
standards for a rapidly growing population. At a more fundamental level, there are
concerns that in many countries policies and political conditions are not in place to
sustain rapid growth, to achieve industrialization and economic transformation, and
to ensure a more equitable development structure. There are, in short, compelling
grounds for debate on African industrialization, structural transformation, and
industrial policies.
Is ‘Africa rising’? Are ‘lions on the move’? Does recent rapid growth in Africa reflect
an African version of the Asian Tigers? Perceptions of Africa’s economic
development have swung dramatically from doom-and-gloom generalizations about a
‘growth tragedy’ or former UK Prime Minister Tony Blair’s ‘scar on the conscience
of the West’ towards a sometimes hyperbolic excitement about recent growth. It is
important to examine how far that growth is underpinned by, or is being converted
into, deeper, more lasting structural change. Most people agree that such
change (p.7) requires industrialization and rapid technological innovation, which
enables countries to move out of subsistence production to more dynamic industrial
production. Historically, industrial policy has been central to successful
industrialization. However, mainstream thinking has been wary of active industrial
policy (see, for example, Ocampo, Rada, and Taylor 2009). Although the idea of
industrial policy has become more fashionable recently, it is still unclear what this
means in practical terms for Africa. This is the issue this book seeks to address.
1.3 Alternative analytical perspective on late industrialization
The book examines conventional views of and contemporary debate on issues
associated with industrial policy. It also deploys an alternative, predominantly
structuralist, development economics approach, inflected with political economy, to
growth and structural transformation in developing countries and draws on a tradition
influenced by Kaldor, Thirlwall, and especially Albert Hirschman. Finally, it makes a
novel contribution to current debates. This work is based on the premise that
economic performance is the outcome of policies and political economy. Economic
transformation and structural change is the essence of economic development.
Industrial development, and development of the manufacturing sector in particular, is
believed by many (though not all) to be the prime driver of economic transformation
and sustained growth (Kaldor 1967; Pasinetti 1981; Thirlwall 2002; Rodrik 2008c;
Reinert 2010; Chang 2003a, 2003b; Amsden 2009).
Ocampo et al. (2009: 7) define economic structure as:
the composition of production activities, the associated patterns of
specialization in international trade, the technological capabilities of the
economy, including the education level of the labour force, the structure of
ownership of factors of production, the nature and development of basic
state institutions, and the degree of development and constraints under
which certain markets operate (the absence of certain segments of the
financial market or the presence of a large underemployed labour force, for
example).

According to Hirschman (1958: 6), ‘[I]n general economic development means

transformation rather than creation ex novo: it brings disruption of traditional
ways…, in the course of which there have always been many losses; old skills
become obsolete, old trades are ruined…’. Structural (p.8) change can be defined in a
simpler way as ‘those changes in the composition that are permanent and irreversible’
(Pasinetti 1993: 1). This suggests that policies should be designed and measured with
the aim of bringing economic transformation and structural change. Because not all
sectors have the same economic or change-promoting characteristics, a sectoral
approach is called for, which also means that linkage effects should be maximized to
generate new activities and induce further investments. It also means that learning
and emulation should be enhanced to bring sustained productivity growth in line with
international competition (Lall 1992, 2000b, 2003; Reinert 2009; Amsden 1989, 2001).
Emulation, moving up the productivity ladder, and advancing towards economic
activities with increasing returns are required for successful catch up. Central to this
endeavour is the presence of an effective developmental state that invests heavily in
physical infrastructure, skills development, and direct credit and other incentives to
pioneer firms, so they can succeed in the marketplace.
Many developing countries have been able to achieve rapid economic growth and
improve global competitiveness. By contrast, economic growth has lagged in most
African countries; or at any rate, it has proven difficult for African countries to
convert episodes of rapid growth into growth sustained over twenty or thirty years.
Economic development in Africa is of utmost importance to Africans, a daunting
challenge to African policymakers, and a black box for researchers. One idea in the
literature on growth rates is the ‘Africa dummy’, a catch-all phrase for the residuals
that cannot be ‘explained’ by other variables in cross-country growth regressions.
Various possibilities are then explored or suggested for what may account for this
dummy—including geographical variables, ethno-linguistic fractionalization scores,
long-run history, institutional development, and so on. However, the Africa dummy
literature has been criticized for being analytically, methodologically, and empirically
fragile (Jerven 2010a, 2010b, 2011).

1.4 Methodology based on original research

There are three main considerations in setting out the following methodological
framework. First, studies on industrial policymaking in sub-Saharan Africa are scant,
making it an under-researched topic. This (p.9) is because such studies have been
sidelined as irrelevant and harmful by mainstream orthodoxy and the Washington
Consensus since the early 1980s. The lack of research has limited wider appreciation
of policymaking, policy learning, and development of policy capabilities. Second,
most studies of industrial development in Ethiopia have focused on firm-level
quantitative data, which are often incomplete and conflicting. Such data are important
in researching an industry’s patterns, but fall short of providing a comprehensive
picture on policymaking, let alone a political economy perspective. Third, the
research approach adopted to date is problematic in that many studies are conducted
too quickly, with an overdependence on secondary data only supported by one or two
research instruments. This has led to incomplete and, in some situations, misleading
findings. Few studies have provided any detail or real depth of understanding of
industrial performance and policy in Ethiopia. As a result, the issues are typically
frozen in polarized and shallow debates.
These approaches have to be viewed in the context of the aforementioned ‘Africa
dummy’, hopeless ‘Afro-pessimism’, and the view of Africa as ‘exceptional’ or
homogenous. A typical recent example is the Neopatrimonial School, an over-
simplistic view that blames African culture for the ‘failure’ to reform economic and
political systems as prescribed by international financial institutions in the 1980s and
1990s. These incomplete perspectives on Africa have compounded the
methodological impediments (Padayachee 2010; Mkandawire 2013) and fail to
generate understanding of African social realities. This has undermined data
reliability and diminished research outcomes and their usability. Mkandawire (2013:
52) stresses that ‘economic policymaking is a highly complex process involving
ideas, interests, economic forces and structures’ that cannot be reduced to a single
explanation. A methodology is needed that recognizes such complexity and considers
broader political economy factors rather than simplistic, uniform diagnoses. In the
Ethiopian context, scarce availability of research and constraints on data collection
compound the methodological problems. Reliable and timely data are not easily
available in most organizations in Ethiopia, and many studies are inconsistent as a
result.
The research for this book was primarily qualitative, while also drawing on
quantitative data (both primary and secondary). A comparative design was adopted,
whereby three different sectors were compared within one overarching industrial
policy. Based on empirical evidence, this study (p.10) investigates causal factors and
their relationships, but does not pretend to weigh with mathematical precision the
relative contribution of specific factors. In view of the limitations of existing data and
studies, the study has relied on extensive original data sources and adopted a more
comprehensive data-collection system. A census using a qualitative and quantitative
questionnaire was conducted in 150 firms, with a 90 per cent response rate. This
complemented in-depth and qualitative interviews involving 200 firms, intermediate
institutions, government agencies, and policymakers. Site observations of more than
fifty factories in differing industries yielded unique insights. A review of some 1,000
primary documents was undertaken, including previously inaccessible documents.
Secondary sources were also extensively consulted. The field study involved 1,300
person days, making this the first in-depth study of industrialization and industrial
policy in Ethiopia (beyond several rapidly conducted consultancy reports).
Appropriate analytical tools have been carefully utilized, generating better
understanding and new findings. These are discussed throughout the remaining
chapters. The author hopes this project will encourage scholars to undertake more
extensive studies. Much of this evidence would have been more difficult, even
impossible, for ‘outsiders’ to assemble.

1.5 Structure of book

The main argument in the book is first, that industrial policy (despite claims to the
contrary) can work and indeed thrive even in low-income countries such as Ethiopia.
Nonetheless, the book cautions that industrial policymaking is a work in progress in
many such countries, and demonstrates the colossal challenge of catching up and
industrialization in twenty-first-century Africa. Second, the book argues that the state
in developing countries can and must play an activist and developmental role beyond
being merely a ‘facilitating’ actor; that is, being little more than a servant of
comparative advantage. Policy independence is an important ingredient. As
Mazzucato (2013a: 5–6) highlighted:
When not taking a leading role, the State becomes a poor imitator of
private sector behaviours, rather than a real alternative. It is a key partner of
the private sector—and often a more daring one, willing to take the risks
that business won’t.…The State cannot and should not bow down easily to
interest groups who approach it (p.11) to seek handouts, rents and
unnecessary privileges like tax cuts. It should seek instead for those interest
groups to work dynamically with it in its search for growth and
technological change.
Third, the book argues that the outcome of industrial policy is typically uneven in
different sectors. Overall, what matters for the evolution and effectiveness of
industrial policy is the way three factors interact—industrial structure, linkage
dynamics, and, broadly, politics/political economy. Fourth, firms from low-income
countries face huge challenges in competing in a globalized economy, and more
effective industrial policies and instruments that facilitate growth and structural
transformation need to be designed and executed.
Based on rigorous and original research in Ethiopia into cement, an import-
substitution industry; leather, an export-oriented light industry; and floriculture, the
book reviews the constraints on, as well as lessons for, Africa’s industrialization and
industrial policymaking. The book focuses on the design and implementation of
industrial policy in Ethiopia, and on how and why policy outcomes are shaped by
different factors in different industries. In view of the renewed interest in industrial
policy in Africa and internationally, the book will be a valuable addition to ongoing
discussion of the ‘African renaissance’, and provide arguments and evidence for the
possibilities of industrial policy in Africa.
The book is organized into eight chapters. Following this introductory chapter, the
second chapter, ‘Climbing without Ladders: Industrial Policy and Development’,
presents a literature review outlining the theoretical framework (structuralist
tradition, catch up, political economy), empirical evidence, and the sub-Saharan
African context. This chapter also outlines the influence of industrial policy. Such
policy has always been a bone of ideological contention, and currently dominant
viewpoints are strongly averse to industrial policy in developing countries, and have
been better at identifying its failures than its evident successes. The third chapter,
‘Setting the Scene: Ethiopia’s Industrial Policies and Performance’, introduces the
main policies and policy instruments developed in Ethiopia in recent years relevant to
the selected case studies. This is done both to clarify the context of policymaking in
different sectors and to identify national-level patterns in policymaking institutions.
The next three chapters provide sector-specific analyses. Thus, Chapter 4, ‘Cementing
Development? Uneven Development in an Import-Substitution Industry’, explores
the cement industry as one of three case (p.12) studies. Cement production is a
strategic import-substituting industry in many countries, driven by growth of the
domestic market. The industry has served as a binding agent of economic
development and transformation in multiple ways. In Ethiopia, it has undergone
major changes in the period under consideration, growing faster than in most
developing countries. This chapter examines the industrial structure, linkage effects,
policy instruments, and institutions of the sector. In sharp contrast to the overall
domination by multinationals of the African cement industry, domestically owned
firms continue to dominate the industry in Ethiopia.
Chapter 5, ‘Beyond Bloom and Bust? Development and Challenges in Floriculture’,
discusses the floriculture sector’s performance, structure, linkages, and
industrialization, and policymaking relating to it. Floriculture shares many
characteristics with manufacturing, and this sector has been an economic success
story that has attracted international interest and policy debate. It emerged in Ethiopia
in 2003 and has since shown sustained growth, making the country the world’s
fourth-largest cut flower exporter. The chapter offers an alternative explanation for
the drivers of growth in the industry, arguing that government policy was critical in
this regard. Furthermore, early policies that helped bring about a successful launching
became inadequate as the industry matured and encountered new challenges. New
policies were thus needed.
Chapter 6, ‘Curing an Underperformer? Leather and Leather Products’, focuses on
the same issues as the previous two chapters, but in relation to leather and leather
goods. Unlike cement and floriculture, the performance of this sector has been
characterized by erratic and sluggish growth. The puzzle is that there is a century of
manufacturing experience and a plentiful endowment of livestock—Ethiopia stands
first in Africa, and among the top ten globally in this regard. Industrial policy relating
to leather and leather products has been unable to reverse this poor performance, and
fully exploit the potential for linkage effects and insertion into the global value chain.
However, recent developments have begun to yield more investment, better quality,
and more exports of higher-end products. The causes and factors are examined, to
derive new insights.
Chapter 7, ‘Failing Better: Political Economy and Industrial Policy in Ethiopia’,
discusses the findings from the case studies more thematically and comparatively,
and presents a comprehensive synthesis. Despite growth in all three sectors, outcomes
were indeed uneven. As noted earlier, what matters overall is the interaction among
industrial structure, (p.13) linkage dynamics, and politics/political economy. This has
significant implications for policymakers.
The last chapter, ‘Lessons from Industrial Policy in Twenty-First-Century Africa’
briefly draws the empirical contribution and analysis together and highlights some
policy and research implications for Africa. Industrial policymaking in Ethiopia is a
work in progress, but industrial policies can work and thrive in a low-income African
country, and the state can and should play an activist developmental role, with policy
independence an important ingredient.11 However, the book also highlights how great
the challenge of catching up and industrialization is for twenty-first-century Africa.

1.6 Conclusion
Few books have been written on the political economy of industrial policy in Africa,
even fewer by Africans. This book provides a unique perspective on why policy
outcomes have been uneven in different industries, a major challenge to
policymakers. It also provides new data and perspectives on the industrial structure of
different sectors and firms. Finally, the study is unique and comprehensive in
covering the macro- (national policymaking process), meso- (industrial policies in
different sectors, institutions, and in-depth industrial structure), and micro-levels. The
author, as both researcher and a senior policymaker, reflects on the challenges and
lessons of policymaking.
The book is a necessary corrective to the over-aggregated and typically superficial
hyperbole on an African economic renaissance. Beyond the restrictive policy
guidance by international financial institutions and bilateral donors, and beyond the
more recent flag-waving for industrial policy, it is important that the scope and
content of industrial policy are informed by empirical evidence and careful,
pragmatic analysis. This book shows that low-income developing countries should
not simply follow ‘comparative advantage’, as some would prefer. But equally it
shows just (p.14) how difficult pursuing an effective industrial strategy can be
(Rodrik 2008a).
Finally, this book highlights how, despite enormous challenges, Africa can catch up,
and contests the standard pessimism that Africa is a hopeless continent. The
distinguished German political economist, Friedrich List (1841: 123) once said, ‘no
nation has been so misconstrued and misjudged as respects its future destiny and its
national economy as the United States of North America, by theorists as well as by
practical men.’ Now, the US is the leading economic power of our time. History
offers many examples of economic miracles occurring in unexpected places. Indeed,
given the role of interventionist states in successful industrialization there, East Asia
might have provided a more important lesson for Africa. Even so, the Ethiopian story
shows what is possible, and the aspiration of Africans to catch up. With long-term
national development vision, highly committed political leadership, and strong
institutions, countries can shift from a relatively agrarian to an industrial society.

Notes:
(1) See UNECA-AU 2014.

(2) The average GDP growth rate for eleven years (2003–14) was 11 per cent, while
the annual average growth rates for agriculture, industry, and service sectors were 9
per cent, 13.8 per cent, and 12.2 per cent respectively. The average annual growth
rate of industry has increased to 20 per cent during the four years (2011–2014) of
Growth and Transformation Plan (MOFED 2014).
(3) The average life expectancy for sub-Saharan Africa was fifty and fifty-six years
for years 1992 and 2012 respectively. See World Development Indicators updated on
30 January 2015, and UN-DESA (2013) <http://data.worldbank.org/region/sub-saharan-
africa>

(4) Speech of the Prime Minister on 23 March 2014, press release from the Prime
Minister’s Office.
(5) Ethiopian Revenue and Customs Authority (ERCA 2012a) and the National Export
Coordinating Committee (NECC 2012).
(6) <http://www.chinadailyasia.com/business/2014-01/27/content_15115269.html>; See also
‘Ethiopia becomes China’s China in Global Search for Cheap Labour’
<http://www.bloomberg.com/news/2014-07-22/ethiopia-becomes-china-s-china-in-search-for-
cheap-labor.html>
(7) Meeting at Prime Minister’s Office in April 2014.
(8) <http://country.eiu.com/article.aspx?
articleid=659462850&Country=Ethiopia&topic=Economy>.

(9) Discussions with Chinese Ambassador in March 2014.

(10) For instance, <http://www.imf.org/external/pubs/ft/scr/2012/cr12287.pdf>;
<http://www.bloomberg.com/news/2013-10-18/imf-says-ethiopian-economic-growth-may-slow-
without-policy-shift.html>.

(11) This is associated to the broader notion of ‘policy space’ that refers to the
‘various tensions between national policy autonomy, policy effectiveness and
international economic integration’ UNCTAD (2014: vii). Globalization, market
internationalization, and legal agreements (multilateral, regional, and bilateral) create
obligations that undermine the scope of national policy.

Climbing without Ladders

Industrial Policy and Development
Chapter:
(p.15) 2 Climbing without Ladders
Source:
Made in Africa
Author(s):

Arkebe Oqubay
Publisher:
Oxford University Press

DOI:10.1093/acprof:oso/9780198739890.003.0002
Ethiopia, like several other African economies, has experienced rapid economic
growth in recent years. But to what extent is that growth underpinned by, or being
converted into, deeper, more durable structural change that lays a foundation for
lasting improvements in the welfare of the majority of the population? Most analysts
agree that such change requires industrialization and rapid technological innovation,
which enables countries to move out of subsistence production into more dynamic
activities. And historically, industrial policy has been central to successful
industrialization. However, mainstream thinking has been extremely wary of active
industrial policy. Although industrial policy has become more fashionable recently, it
is still unclear what this means in practical terms for Africa. This chapter reviews this
debate in the light of a long tradition of economic history and thinking. It emphasizes
the relevance of a structuralist school of thought in development economics.
Keywords: Ethiopia, Africa, economic growth, structural
change, welfare, industrialization, industrial policy, structuralist development economics

2.1 Introduction
All advanced capitalist nations have used industrial policies and state interventions to
establish their first mover status, to consolidate their advantages, or to force the pace
in catching up on those ahead of them. Industrial policies have been the rule rather
than the exception. The UK and the US are typical examples of the frontrunners of
the nineteenth and twentieth centuries, with the US having played a successful catch
up game itself. For instance, the eighteenth US president, Ulysses Grant (1868–76),
once pointed out:
For centuries, England has relied on protection, has carried it to extremes
and has obtained satisfactory results from it. There is no doubt that it is to
this system that it owes its present strength. After two centuries, England
has found it convenient to adopt free trade because it thinks that protection
can no longer offer it anything. Very well then, gentlemen, my knowledge
of our country leads me to believe that within 200 years, when America has
gotten out of protecting all that it can offer, it too will adopt free trade
(cited in Frank 1967: 164).

And, of course, industrial policy continues to be the rule. As is increasingly

acknowledged, the US owes many of its successes to state support for technical
innovation (Mazzucato 2013a, 2013b).
Current orthodoxy in development economics, however, favours a minimalist state
and short-term intervention. For example, Collier (2007) argues that ‘quite possibly
the easiest way for the state to “do no (p.16) harm” in this situation is for it to be
small, and concentrate on essential public services.’ By contrast, commitment to an
activist state is virtually a sine qua non for structuralist development economists
(Hirschman 1958; Ocampo 2005; Ocampo et al 2009; Chang 1994; and Amsden 1989).
For these economists, there is a distinction between tailoring interventions to the
elusive idea of ‘market failure’ and intervening to create and shape market
institutions and transactions according to criteria less informed by the quasi-religious
faith in competitive perfection displayed by neoclassical economics.
Despite periodic growth episodes, African countries have not succeeded in catching
up with advanced capitalist countries. In this context, and after decades of economic
stagnation and political instability, Ethiopia has embarked on a course of economic
revitalization. The country has adopted home-grown economic policies,
experimented with ethnic federalism, and broadly followed a line derived more from
pluralist economic traditions than from the ‘mono-economics’ often regarded as
orthodoxy. Arguably, Ethiopia’s economic expansion between 2003 and 2014 is
attributable to an aspiring developmental state and the adoption of activist economic
policies. However, Ethiopia’s industrialization and industrial policymaking are still a
work-in-progress.
What is industrial policy and what are its foundations and rationale? Can industrial
policies and state activism be relevant in the contemporary world? This chapter
explores industrial policymaking by outlining theoretical foundations and provides
empirical perspectives on industrial policy relevant to understanding the prospects for
(and process and outcomes of) such policy in Ethiopia. After introducing the concept,
the first part of the chapter discusses the structuralist tradition, classical political
economy, and heterodox or pluralist perspectives. Linkage effects, both from
industrial and policymaking perspectives, are also highlighted in this chapter.
Linkages have direct relevance to policymaking. Just as there is policy learning-by-
doing as well as industrial learning-by-doing, so there are linkage-like dynamics in
applied industrial policy as well as among economic activities. This idea
acknowledges that within the government apparatus, through shifts in one area
calling forth new initiatives in other areas or levels of government, bottlenecks and
obstacles may generate a ‘uneven development’ of policy capability. The last section
of the chapter presents empirical perspectives on Africa.

(p.17) 2.2 Industrial policy

Industrial policy has generated considerable controversy and is vaguely defined,
being frequently endowed with different meanings depending on context and the
person who defines it (OECD 1975; Johnson 1984; Chang 1994; Rodrik 2008c). One
definition is ‘a policy aimed at particular industries (and firms as their components)
to achieve the outcomes that are perceived by the state to be efficient for the economy
as a whole’ (Chang 1994: 60). This definition, despite its emphasis on selective
targeting, strategic orientation, and efficiency, omits (or keeps implicit) any structural
change and the various stages in catching up. Chang’s definition resembles Johnson’s,
which also emphasizes a strategic orientation: ‘above all…industrial policy means the
infusion of goal-oriented strategic thinking into public economic policy’. For Johnson
(1984: 8), industrial policy is ‘the initiation and coordination of government activities
to leverage upward the productivity and competitiveness of the whole economy and
of particular industries in it.’
Others, like Amsden and Chu (2003), Ocampo et al (2009), and UNCTAD-UNIDO
(2011), offer definitions more explicitly oriented towards structural change and
catching up. According to UNCTAD-UNIDO (2011: 34), industrial policy means:
…government measures aimed at improving the competitiveness and
capabilities of domestic firms and promoting structural transformation.
Industrial policy involves a combination of strategic or selective
interventions aimed at propelling specific activities or sectors, functional
interventions intended at improving the workings of markets, and
horizontal interventions directed at promoting specific activities across
sectors.
The emphasis is on ‘restructuring of production and trade towards activities with
higher technological content’ and promoting ‘innovative activities that generate
domestic spill over’ (Ocampo et al. 2009: 152–3). New activities include new
industries, new products, new markets, new technologies, and new institutions. While
the prime focus is on manufacturing, high value-added products such as horticulture
are not excluded (Ocampo et al. 2009). In this definition, industrial policy is viewed as
a vehicle for structural change, that is, constant upgrading to higher productivity
activities with increasing returns and the centrality of technological development
(Amsden 2001; Reinert 2009; Rodrik 2011). This (p.18) definition calls for an activist
state, a perspective rejected by neoclassical orthodoxy, and captures the long-term
orientation of industrial policy.
The institutionalist perspective argues that is vital for the state and industrial policy to
govern the market (Chang 1994; Wade 1990; Amsden 1989). For instance, Amsden
(1989) emphasizes not simply getting relative prices right but ‘deliberately getting
prices wrong’ in order to foster industrialization and structural transformation. Indeed
later, Amsden (n.d.) also argued that getting property rights wrong had been equally
important in many successful cases of structural change. The mechanisms involve
market control, protection of foreign investors, picking winners (chaebol, etc.),
subsidizing capital, tariff protection, etc. This argument runs contrary to the whole
continuum of positions supported by neoclassical economics, from the more extreme
neoliberal or ‘market fundamentalist’ tendency to the more moderate views based on
‘market failure’ analysis. The core argument in economic orthodoxy is that state
intervention and industrial policy distort resource allocation, where optimal outcomes
would flow from free markets and free trade regimes (see Bhagwati 1989a, 1989b).
Such economists ardently believe that industrial policies only ever cultivate
unproductive rent-seeking, and that picking winners is impossible
(Krueger 1974, 1980, 1990). Neo-utilitarian and public choice models consider
governments to be inherently ‘rent-seeking’ and view public officials as rational
(selfish) maximizers. Free markets, trade openness, economic liberalization, and
maximal (rather than optimal) competition on the grounds of comparative advantage
are advocated as ideal mechanisms for efficient industrialization. Industrial policies
are often also misrepresented or simplified as import-substitution industrialization, or
as any form of state intervention (Noland and Pack 2003; Weiss 2013; Peres 2013;
Warwick 2013).
This book proposes an operational definition of industrial policy, namely ‘a strategy
that includes a range of implicit or explicit policy instruments selectively focused on
specific industrial sectors for the purpose of shaping structural change in line with a
broader national vision and strategy.’ The definition clearly underlines the need for
selectivity, both in terms of specific industries and targeted interventions, and
emphasizes structural changes within the broader framework of the national
economy. The following sections explore the theoretical influences on industrial
policy, that is, the structuralist perspective, the theory of catch up, infant industry
theory, and political economy perspectives.
(p.19) 2.3 Structuralist and catch up perspectives on industrial policy
Industrial policy is primarily underpinned by a structuralist perspective and tradition
that upholds the manufacturing sector as a particular engine of growth and driver of
structural economic transformation, and as being the best prospect for catching up by
late (or late-late) industrializers.

2.3.1 Structuralist perspectives on industrial policy

The structuralist approach to economic development holds that economic growth
alone does not necessarily produce structural economic transformation and sustained
economic development. Economic development is a process of fundamental
structural changes and economic transformations embedded in sectoral shifts;
sustained productivity increases as a result of constant technological development,
accompanied by fundamental shifts in work occupations; and institutional and
political economy transformation (Chenery 1960; Chenery, Robinson, and
Syrquin 1986; Reinert 2010; Thirlwall 1980, 2002, 2011; Thirlwall and Bazen 1989;
Ocampo et al. 2009; Tregenna 2013; and UNCTAD 2011). Structural transformation
involves the shift of resources from low to high productivity sectors and activities,
and the fostering of a process of reallocation of productive factors among industry,
services, and modern agriculture (Ocampo 2008; UNCTAD 2006; Tregenna 2013). This
process is characterized by continuous diversification into new economic activities
with stronger domestic economic linkages and a higher quality path dependency, and
focuses on the development of domestic technological capabilities (Rodrik 2008c).
It is important to clarify the different emphases within the structuralist approach.
First, there are the arguments associated especially with Kaldor, whose propositions
or growth laws were derived from stylized facts suggesting that manufacturing has
special growth-promoting qualities. Second, there is Thirlwall’s (2002) view that
because rapidly growing developing countries face a structural balance of payments
constraint, they need to focus on rapid growth of export revenue. Third, there is what
might be called the UNCTAD strain of structuralism drawing on the Prebisch-Singer
hypothesis (of a secular decline in the net barter terms of trade for primary
commodities vis-à-vis manufactured goods) and its implications. Meanwhile, there
are others who argue against relying (p.20) exclusively on primary commodity
exports because of the relative volatility of their export prices. There are also critics
of the Prebisch-Singer hypothesis who argue that it has encouraged many years of
excessive and damaging export pessimism among the governments of developing
countries.

Manufacturing as the Engine of Growth and Transformation

Among advanced countries and newly industrializing countries (NICs), economic
growth has been associated with increased income per capita and industrial
development (Amsden 2001; Rodrik 2011; McMillan & Rodrik 2011). Tregenna
(2008a, 2008b, 2012, 2013) distinguishes between industry and manufacturing,
emphasizing that manufacturing specifically has special properties as an engine of
growth: ‘dynamic economies of scale in manufacturing; strong backward and forward
linkages between manufacturing and other sectors of the domestic economy; strong
properties of learning-by-doing; innovation and technological progress; and the
importance of manufacturing for the balance of payments’. Thirlwall (2002: 41–2)
expounds on the strong positive link between manufacturing and growth by referring
to Kaldor’s growth laws:
The first law is that there exists a strong causal relation between the growth
of manufacturing output and the growth of GDP. The second law states that
there exists a strong positive causal relation between the growth of
manufacturing output and the growth in productivity in manufacturing as a
result of static and dynamic returns to scale. This is also known as
Verdoorn’s law. The third law states that there exists a strong positive
causal relation between the rates at which the manufacturing sector
expands and the growth of productivity outside the manufacturing sector
because of diminishing returns in agriculture and many petty service
activities which supply labour to the industrial sector. [My emphasis]

According to Kaldor (1966, 1967), it is impossible to understand the development

process and growth without a sectoral approach, and without differentiating between
increasing returns in manufacturing and diminishing returns in agriculture and mining
(which are land-based). Similarly, Young (1928: 539) emphasized that economic
progress partly depends on the increasing returns realized by ‘progressive division of
labour and specialization of industries’. Young drew on Adam Smith’s insight that
‘division of labour depends upon the extent of the market, but the extent of the
market also depends upon the division of labour’. Historical facts indicate that
growth of the manufacturing sector at earlier stages of (p.21) development depends
on the transformation and increased productivity of agriculture, while at later stages it
is fed by resources from exports, suggesting the loop between exports and growth
(Thirlwall 2002). Ocampo et al. (2009) reinforce this precept by inferring that
economic structure and structural transformation is associated with:
…the composition of production activities, the associated patterns of
specialization in international trade, the technological capabilities of the
economy, including the educational level of the labour force, the structure
of ownership of factors of production, the nature and development of basic
state institutions, and the degree of development and constraints under
which certain markets operate…economic development is a process of
structural transformation,…the reallocation of productive factors from
traditional agriculture to modern agriculture, industry and services…
shifting resources from low-to-high productivity sectors…a capacity to
diversify domestic production structure: that is, to generate new activities,
to strengthen economic linkages within a country and to create domestic
technological capabilities.
These insights, rooted in economic history, also caution against the pretence that
African economies could become advanced capitalist societies without having to pass
through a phase of (environmentally often damaging) industrialization. That
argument, that industrialization is no longer fundamental to economic development,
is also a product of the misinterpretation of experiences in countries such as
Singapore, India, and Switzerland. Singapore and Switzerland, in fact, have very high
value added manufacturing compared to almost all other countries. And India’s much
vaunted services-based growth reflects a swing from a trade deficit in services in the
early 2000s to a very modest service trade surplus. Further, a huge amount of service
sector activity and employment provides services to manufacturing. There is really no
convincing evidence that a low-income country can generate sustained rises in living
standards without manufacturing. That is why many regard the fall in the share of
manufacturing in developing countries, including in Africa, in recent years as
‘premature deindustrialization’. This is what Rodrik (2011) refers to as ‘growth-
reducing structural change’.

The Strategic Importance of Exports

Some economists emphasize the strategic importance of exports to growth and
structural change. For many countries, the export growth rate sets the threshold for a
country’s rate of growth. The composition (p.22) of exports is cardinal to
industrialization and economic transformation, and can determine future trajectory.
Lall (2000b: 1) emphasizes that ‘export structures, being path-dependent and difficult
to change, have implications for growth and development’. Thirlwall in particular
(2002) has argued that sustained growth and structural change in developing countries
is threatened by a structural balance of payments constraint, posing critical policy
challenges for governments. As a consequence, rapid growth of export earnings
sustained over a long time cannot be relied upon if they based exclusively on primary
products or low-end exportable products, partly due to the volatility of demand (and
prices) for such commodities. Hence, the least developed countries (LDCs) face a
steep uphill struggle and, as Thirlwall (1980, 2011) argues, structural balance of
payments constraints. The dynamic growth of exports depends on diversified
manufacturing goods, with a constant upgrading of the technological features of
manufactured goods (Lall 2000b).
Related to this, structural development economists like Raúl Prebisch (1950)
associated with the Economic Commission for Latin America and Caribbean
(ECLAC) argued earlier that the structure of the world economy is tilted against
developing countries. Low-income countries export primary commodities, for which
there is a low income elasticity of demand. In Engel’s Law, as a person grows richer,
they spend a shrinking proportion of their income on basic needs like food, and this is
also applicable to countries. Meanwhile, the state of world trade dictates that low-
income countries import manufactured items (for which there is a high income
elasticity of demand). One consequence of these, and other, structural features of
trade and production is that the terms of trade would systematically turn against
exporters of primary commodities. Therefore, to protect low-income countries, there
is a strong argument for protective trade policies and nurturing infant industry.1

2.3.2 Late Development, Catch up, and Industrial Policy

Throughout the history of capitalism, the level of industrial development has been
uneven across countries and regions. Empirical evidence (p.23) strongly suggests that
this unevenness is an essential characteristic of capitalism and industrialization
(Gerschenkron 1962; Reinert 2010; Chang 2003a, 2003b; Schwartz 2010). Although all
agree on this unevenness, different schools provide differing advice regarding the
‘right course’ to achieve (industrial) development or catch up. Mainstream
economists argue that developing countries should follow the same (imagined) path
as the forerunners or advanced industrial countries, and that there is no ‘short-cut’ to
economic development. Others, like the ‘preconditions’ theorists, argue that some
preconditions (such as natural resources, capital and saving, entrepreneurship and
managerial abilities, skills or human capital, and institutions) are the key
determinants of economic development (Hirschman 1958, 2013; Adelman 2013).
However, according to Hirschman (1958), these resources are available in great
abundance among a disguised, hidden, unemployed, reserve army of entrepreneurs.
Furthermore, the proponents of catch up theories argue that in ‘the more backward
countries…the great spurt of industrial development occurred despite the lack of
these prerequisites’ (Gerschenkron 1962). Gerschenkron argued that latecomers have
the advantage of backwardness, as they can be motivated to learn from forerunners.
Abramovitz (1994) also emphatically highlights that ‘being backward in level of
productivity carries a potential for rapid growth’. According to Gerschenkron (1962:
45), however, ‘the higher the degree of backwardness, the more discontinuous the
development is likely to be’. Further, the implication for structuralist development
economists is that the ‘advantages of backwardness’ do not generate unconditional
convergence, but require direct intervention.
Latecomers have a proposed advantage in that they can learn from forerunners
(Amsden and Hikino 1994). Latecomers have a compelling reference in the
forerunners and their experience, although they have to develop in the face of a
formidable competitive pressure from the forerunners. These latecomers are
recommended to use institutional innovations to create, mobilize, and concentrate
resources with maximum speed and effect. By contrast, forerunners had to develop
without any reference point, although they were rewarded with super profits
(Shin 1996). The US, a latecomer, was able to catch up with and forge ahead of
Britain, the leading industrial economy in the nineteenth century. Germany, France,
Sweden, and (later) Finland were also able to catch up in a similar way. The source of
the latecomer advantage was the opportunity to follow a more deliberate and less
spontaneous process, in the (p.24) footsteps of the economic giants. Moreover, the
process was facilitated by multiple contacts between the latecomer and forerunner.
More recently, Japan, Korea, Taiwan, and currently China have been able to catch up
without having the full complement of commonly identified prerequisites for
advancement.
In response to uneven development, different states tend to follow different
development paths that combine a ‘Ricardian strategy’ and a ‘Kaldorian strategy’
(Schwartz 2010). The former is based on the export of primary commodities, or even
low-cost and low-quality manufactured products, and fundamentally follows
comparative advantage. This strategy may be relatively easy to follow, is vulnerable
to deteriorations in terms of trade and to highly volatile world market prices, and may
not succeed in the end. By contrast, a Kaldorian strategy is export-oriented, based on
manufactured products with increasing returns, and maximizes Verdoorn effects.
Such a strategy focuses on rapid technological developments and aims to develop
competitive advantage. A Kaldorian strategy relies on ‘a set of interrelated
phenomena like increasing returns to scale, learning by doing, imperfect competition,
and economies of speed to generate growth’ (Schwartz 2010: 60).
Development is thus about breaking the vicious cycle by focusing on the binding
agents, the essential underlying structures, and the strategic issues of development, all
of which will allow dispersed, hidden, and unutilized resources to be fully mobilized
for development. The pace of development depends not on prerequisites but on the
nation’s ability and determination to organize for development, which is created in
the very process of development (Gerschenkron 1962). One of the state’s primary
roles is creating the motivation to develop with a basic ‘growth perspective’, which is
rooted in the belief that a society has ‘to move forward as it is, in spite of what it is
and because of what it is’ (Hirschman 1958: 10). ‘Such a view’, Hirschman wrote
in Development Projects Observed (1967: 5), ‘stresses the importance for
development of what a country does and of what it becomes as a result of what it
does, and thereby contests the primacy of what it is, that is, of its geography- and
history-determined endowment with natural resources, values, institutions, social and
political structure, etc.’ This can best be done by focusing on clearing blockages and
releasing potentials, and aiming at inducement or learning ability as the key link or
binding constraint. This, in turn, depends on instilling a ‘growth-perspective’ or ‘the
desire for economic growth and the (p.25) perception of the essential nature of the
road leading to it’ (Hirschman 1958: 10). Examples of societies that have made
advances in particular periods led by just such a growth perspective include South
Korea after the Korean War, the US after the American Civil War, Europe after the
Second World War—indeed, often but not always after wars. (Although the civil war
ended in 1949, China’s fastest growth occurred after the adoption of its Openness
Policy in 1978). Often but not always, advancement is also closely tied to nationalism
or to external and internal threats (Doner, Ritchie, and Slater 2005).
According to Hirschman, ‘the complementary effect of investment is therefore the
essential mechanism by which new energies are channelled toward the development
process’, and the process is one ‘where one disequilibrium calls forth a development
move which in turn leads to a similar equilibrium and so on ad infinitum’
(Hirschman 1958: 72). This gives rise to the critical role played by linkage effects and
latitudes for performance discussed in the section 2.6 of this chapter. Hirschman
(1958) emphatically states that the major obstacle to change is the negative perception
or image of change, which thwarts the process of mobilization for development. Such
fear of change is often tied to particular interests that are threatened by change. In
other words, there is a political economy dimension to this.

2.4 Theory of infant industry in classical political economy

The theory of infant industry can be traced back to Alexander Hamilton (1755–1804).
It was he who first proposed this concept, while Friedrich List (1789–1846) is
considered to have laid its theoretical foundations (List 1841, 1856; Chang 2003b;
Reinert 2009, 2010). The theory is aimed at refuting David Ricardo’s (1772–1823)
theory of comparative advantage and free trade. According to Ricardo, international
trade should be free, with each country specializing in commodities with the least
relative costs of production, hence its comparative advantage (Ricardo 2004[1817]).
The Washington Council for International Trade (2013) defines comparative
advantage as ‘the ability to produce goods at a lower cost, relative to other goods,
compared to another country’ (www.wcit.org). The comparative argument states that
developing countries should produce raw materials and exchange the materials for
manufactured (p.26) goods from the industrial countries. Moreover, it argues that free
trade and the market mechanism should be left to function without policy
intervention. The theory predicts that this would benefit both industrial economies
and agricultural exporters. Essentially, the theory is sensible on its own terms, but
desperately static and unrealistic (Singh 2011).
By contrast, the theory of infant industry is based on the assumption that the
manufacturing sector should play the key role in the economy, and that its promotion
requires jumping ahead of current comparative advantage, thus necessitating
protection of infant industries, the use of industrial policies, and an indispensable role
for the state. These facets are extensively discussed in the fundamental works of
Hamilton (1934) and List (1841). In the early days, when the US was predominantly
agrarian and had little industrial base, policymakers did not understand the benefits of
manufacturing. President Jefferson maintained that the US should depend on
agriculture, as it could rely on European imports for industrial goods (Goodrich 1965).
However, Hamilton argued that manufacturing was the key to wealth creation (in
contrast to mining, for instance), and that it would also help in the development of
agriculture through increased demand, productivity improvements, cheaper supplies
of industrial products, and a secured market. Hamilton highlighted the overall
benefits in terms of increased labour productivity, full employment, and increasing
returns in manufacturing. Like Hamilton, List argued that manufacturing was the
primary source of wealth, and a diversified economy based on manufacturing and
agriculture has stronger stimuli for growth than an economy based solely on
agriculture.
The rationale for protecting infant industry is based on the often prohibitive risks of
acquiring finance and new skills, the lack of inducement by entrepreneurs, and the
formidable competition from forerunners. The competitive obstacle posed by
predominant economies such as Britain was profound and nearly insuperable. List
(1827: 32) remarked:
 It takes a long time until the labourers are experienced in the different
workmanship and accustomed to it; and until the necessary number for
every business is at all times to be had…In the old manufacturing
countries, we observe quite the contrary…The old country, as long as it
preserves its freedom, its vigour, its political power, will, in a free
intercourse, ever keep down a rising manufacturing power.…A new
country is moreover, the less able to contend against the manufacturing
power of the old country, the more the interior market of this
old (p.27) country is protected by duties, and competition in the new
country is supported by drawbacks, and by an absence of duties in the
foreign markets.

List contends that ‘even if there were not capital and skill enough in the country, they
could be drawn from abroad by political measures’. He pointed out that Britain rose
to become the leading industrial power by adhering to three mechanisms: ‘First, to
prefer constantly the importation of productive power to that of commodities; second
to maintain and carefully protect the development of productive power; third, to
import only raw materials and agricultural products, and to export only manufactured
articles’ (List 1856: 297). In response to the assertion that protectionism stifles
domestic competition and consumer interests, Hamilton argued that once the
protected infant industry is able to establish itself, more manufacturers will
participate in domestic industry, weakening monopoly and pushing down prices.
The high cost of acquiring new skills and the hesitation by industrialists in the face of
perceived risk makes state promotion of infant industry an absolute necessity. The
novelty of the industry and the need to build a national industry are the rationale for
this support. Both Hamilton and List provide insights into the design of industrial
policy. Hamilton emphasizes the importance of understanding the successful
instruments countries can use. List distinguishes between inducements and
restrictions, between targeted and general instruments, technological transfer, and the
importance of a more refined tax system. He contends that the selection of industries
should be based on their contribution to the nation’s economic and defence interests,
and on technological considerations (and even specifies five criteria). Moreover, he
recommends the alternating, and sometimes combined use of protection and
promotion, and a package for each industry, including incentives for importing
locally unavailable raw materials (duty-free privileges) and a bounty for local
production of raw materials. The foreseen implementation problems and need for
prevention of potential abuse are also stressed.
In his Report on Manufacturers (5 December 1791), Hamilton further advocated the
importance of inducing foreign capital and investments, foreign technology, and
foreign experts and skilled labour to US manufacturing industry. He called for the
establishment of a planning board, and the allocation of funds for advancement of
technology, rewarding invention, inducing specialists, and importing foreign
technologies. This (p.28) was justified on the grounds that ‘in countries where there is
great private wealth, much may be effected by voluntary contributions of patriotic
individuals; but in a community situated like the United States, the public purse must
supply the deficiency of private resources. In what can it be so useful, as in
promoting and improving the efforts of industry?’ (Hamilton 1934: 276).
List stresses that protecting infant industry ‘can be operative for good only so far as it
is supported by the progressive civilization and free institutions of a nation, we learn
from the decay of Venice, Spain and Portugal, from the relapse of France…and
history of England’ ([1841] 2005: 132). List links policy effectiveness to ‘progressive
civilization’, which is similar to the widespread political commitment to growth that
Hirschman, among others, was interested in (though arguably also foreshadowing the
‘preconditions’ approach favoured by proponents of the more recent ‘good
governance agenda’). He also argues that the same policy may be effective in one
institutional context but less so in another. In this, he anticipates the arguments of
more recent industrial policy advocates and analysts, as well as economic historians
like Peter Hall (1987).
Both Hamilton and List argue that infant industry cannot be developed without strong
state leadership, maintaining that state intervention emanates from the unqualified
sovereign power of the government based on the constitution (Hamilton 1934;
List 1841). In his letter to George Washington, Hamilton (1934) notes that:
…every power vested in a government is in its nature sovereign, and
includes…a right to employ all the means requisite and fairly applicable to
the attainment of the ends of such power, and which are not precluded by
restrictions and exceptions specified in the Constitution, or not immoral, or
not contrary to the essential end of the political society. The principle, in its
application to government in general, would be admitted as an axiom…(p.
xiii)

List emphasized that the source of state intervention lay in the state’s sovereign
power and stressed that government ‘has not only the right, but also it is its duty, to
promote everything which may increase the wealth and power of the nation, if this
object cannot be effected by individuals’ (Earle 1986: 247; Austin 2009: 81). US
industrial policy was based on Hamilton’s principles and this policy was evident until
the 1950s. Goodrich (1965: vii) confirms that the American government
took (p.29) ‘deliberate action to promote industrialisation and economic growth’. The
purposeful industrialization of Japan after the 1860s by the Meiji dynasty was based
on very much the same belief.
In line with this, Mazzucato (2011, 2013a) argues that ‘market failure’ (the ideal of
perfect competition from which the idea of ‘failure’ is derived) is a myth, and that the
US remains one of the most interventionist of entrepreneurial states. The state has
played an active role in creating and shaping new products, new industries, many
leading high-tech firms (such as Google, Intel, Apple), and new technologies, such as
the internet, biotechnology, nanotechnology, space technology, and most new
medicines. The Economist (2013: 56) summarizes Mazzucato’s views:
 Ms Mazzucato says that the most successful entrepreneurial state can be
found in the most unlikely place: the United States. Americans have
traditionally been divided between Jeffersonians (who think he governs
best who governs least) and Hamiltonians (who favour active government).
The secret to the country’s success lies, she thinks, in talking like
Jeffersonians but acting like Hamiltonians.

Of Britain, O’Brien (1991: 33, cited in Ocampo et al. 2009) argues:

For more than a century, when the British economy was on its way to
maturity as the workshop of the world, its governments were not
particularly liberal or wedded ideologically to laissez-faire…the
Hanoverian governments…poured millions into strategic objectives which
we can see (with hindsight) formed pre-conditions for the market economy
and night-watchman state of Victorian England.

Two centuries after these initial debates, and after the so-called Asian miracle, the
debate between orthodox and heterodox scholars continues. A recent exchange
between Ha-Joon Chang and Justin Lin on industrial policy typifies the state of the
argument in the early twenty-first century. The two development economists differ on
the role of the state and whether industrial policy should be based on comparative
advantage-following (CAF) or comparative advantage-defying (CAD) principles (Lin
and Chang 2009; Lin 2009). Justin Lin, while arguing for a CAF and ‘facilitating’
state, cautions against efforts to promote new comparative advantages, competitive
edges, and the constant upgrading of industries. His emphasis is adamantly on
‘a facilitating state—a state that facilitates the private sector’s ability to exploit the
country’s areas of comparative advantage…the key is to make use of the country’s
current comparative advantage—not in the factors of production that it may have
some day, (p.30) but in the factors of production that it has now’ (Lin and
Chang 2009: 2). In contrast, Chang argues that nations should focus on creating and
developing their comparative advantages as well as exploiting existing comparative
advantage, and he argues for a much more interventionist or activist state. Despite the
appeal of this approach—based especially on economic history—Chang arguably
fails to address how developing countries can effectively deploy most of their
available productive forces in the development of a few high-tech industries, given
technological backwardness.
Another contribution on the theoretical niceties of comparative advantage and their
practical implications for policy comes from Ajit Singh. He (2011: 13) argues that
reliance on existing comparative advantage is insufficient and that ‘rather than close
integration with the world economy, developing countries should seek strategic
integration that enables them to integrate up to the point where it is in their interests
to do so. This was the strategy followed by the East Asian Miracle countries.’ Singh
(2011: 13) highlights four areas in which trade openness is beneficial: ‘relatively
specialized resources’, ‘diffusion of knowledge’, ‘sufficient competitive pressure’,
and in accelerating a Schumpeterian ‘process of creative destruction’. This echoes,
while going somewhat beyond, Pasinetti’s (1981: 259) argument that ‘the primary
source of international gains is international learning (not international trade), where
firms in one country are challenged by lower priced products from abroad. They will
either learn how to cut down costs or close down. Some of them, at best, may learn
and survive.’ Rather than having to choose between mutually exclusive positions
(CAF versus CAD), it may be possible to pick up on Singh’s more pragmatic and
eclectic approach. Such an approach is neatly captured in Schwartz’s (2010) analysis,
theoretically informed and based on economic history, and introduced above. He
argues that many industrializing countries have relied on a combination of relying on
Ricardian (CAF) practices at an early stage of development and gradually shifting
towards Kaldorian (more CAD) strategies (Schwartz 2010). This approach is
embedded in dynamic comparative advantage, which is driven by policy to catch up,
based on the belief that endowments are endogenous and can be altered (Lall 2005;
Johnson 1984). Competitive edge was created by getting involved in new activities
(learning by doing), and by continuously adjusting policies to develop new and
higher level productive activities that lead to economic transformation (Rodrik 2011;
Hirschman 1958; (p.31) Amsden and Chu 2003). This approach also builds on the
belief that technological development is the key determinant of successful
industrialization.

2.5 Political economy perspectives and dimensions of industrial policy

Perhaps the key political economy issue is the distribution of particular material
interests among groups and the way these can create obstacles to structural change
that promotes growth. This section highlights the role of the state and the political
aspect of industrial policymaking, and the productive use of rents to accumulate
technological capabilities and promote continuous productivity growth. Key issues in
industrial policymaking include the complementary relationship between export-led
and import-substitution industrialization. Industrial policy also involves the
enhancement of policy capabilities, policy instruments, and institutions; and utilizing
linkage effects, including the productive use of latitude for performance standards.
All of these are discussed below.

2.5.1 Political economy approach to industrial policy

An industrial policy is not a technical exercise, but a political process in which
political and economic factors interact. Policy formulation, implementation, and
outcomes depend on the compatibility of state–society relationships and the political
support (or quiescence) of economic actors (Hirschman 1963; Hall 1987; Di
John 2009). Understanding the political process, the polity, and the political
dimension of each economic policy is likely to be a key determinant of industrial
policy outcomes. The state is a political entity and its actions have to be explained
within the dominant polity, power balance, or political settlement. Specific social
interests drive its actions, and its support is based on the dominant political forces and
their interests (Hall 1987; Di John 2009; Hirschman 1963).
According to institutionalist perspectives of political economy, industrial policies are
constrained by socioeconomic structures conveyed through political processes and
determinants. National economic policies are determined ‘first, by what a
government is pressed to do and, secondly, by what it can do in the economic
sphere…the former defines what is (p.32) desirable in a democracy and the latter
defines what is possible’ (Hall 1987: 232). The key political determinants are the
organization of capital (the relations between finance and industrial capital); of labour
(labour markets, organized labour, and labour relations); of the state (the internal
organization of state apparatus); of its political economy, which includes electoral
practices and political parties; and the nation’s position in the international economy
(Hall 1987). According to Hall, an institutional political economy approach
underscores organizations as key variables in policymaking, which is a political
rather than a technical exercise. Policies are responses to pressures from specific
social groups, whose interests are significantly influenced by the economic and
political structure. Economic policies are often responses to contesting pressures and
demands, with major distributive effects (Hall: 1987). For instance, President
Jefferson rejected Hamilton’s policy proposal, because American manufacturers were
too weak to exert meaningful pressure in the early days. The proposal was put into
action in 1812, but only after war became imminent and US manufacturers were
strong enough to exert pressure (Goodrich 1965). In Latin America, by comparison,
given the insignificant share of industrial exports relative to the export of primary
products by powerful land aristocrats (Hirschman 1968), manufacturers were in no
position to exert enough pressure on the state to adopt favourable policies. The
success of a policy ultimately depends on the state’s having a clear vision for the
sector; the will and capacity to enforce the policy; and the ability to mediate tensions
among economic and political actors, while maintaining its autonomy
(Cramer 1999a). Cramer (1999a) shows how political factors were one of the key
determinants in the industrial processing of primary commodities (or not). It should
also be noted that what matters is not the amount of state intervention, but the type
and nature of state intervention and leadership (Amsden 1989; Chang 1999;
Johnson 1982). This perspective is echoed in recent work by Buur (2014), Buur et al.
(2012) and associates, including Whitfield (2011) (also see Chapter 7).
In France, state intervention took the form of nationalized state-owned banks and
industrial corporations, and the active role of state institutions (Commissariat Général
du Plan, CGP; Ecole National d’Administration, ENA; INSEE, the national agency
for the collection and analysis of economic data) in planning. In Germany, privately
owned investment banks created and guided investments in the railway and steel
industries. They also shaped so-called ‘cooperative managerial capitalism’
(Hall 1987; (p.33) Chandler 2004). In Japan, the Ministry of International Trade and
Industry (MITI) played a vital role, and bank-corporate organizations (Zaibatsu,
Keietsu) were a key feature, while in Korea the activist state, chaebol, and state-
owned banks and state-owned enterprises (for instance, the steel manufacturer
POSCO) played a significant role in the country’s catch up process (Amsden 1989;
Johnson 1982).
The developmental state is based on development projects that have the potential to
organize society around its vision, and the autonomy (referred to sometimes as ‘hard
state’) to insulate itself against the narrow interests of specific actors and to secure
the support of all social forces (Evans 1995; Woo-Cumings 1999). Kohli’s ‘cohesive
capitalism’ has some of the attributes of a developmental state (Kohli 2004). In Korea
and Taiwan, successful industrialization was supported by land reform that unlocked
the potential of rural farmers, secured support, and weakened the forces maintaining
the preindustrial status quo (Wade 1990; Amsden 1989). Kohli argues that Japanese
reforms and power relationships with local social forces positively influenced the
catch up and developmental agenda (Woo-Cumings 1999). This configuration was
conducive to industrialization, and demonstrated the difference in the capability of
the Japanese state in comparison to Latin America (Kay 2002). Notice should also be
taken of the role of horrific war in the dramatic social, political, and institutional
change in South Korea. War is a not uncommon precursor of fundamental social
change that can be associated with long-run development. Thus Cramer (2006) argues
that war has often provided an ‘enabling environment’ for economic development.
Meanwhile, Ocampo et al. (2009) highlight the role of military expenditure in driving
the expansion of British capitalism in the late seventeenth and early eighteenth
centuries.

2.5.2 Heterodox views on developmental states

The most successful industrial catch up of the twentieth century was in East Asia,
most notably in Japan, Korea, and Taiwan, and more recently China. This was
spearheaded by what have come to be labelled developmental states. Arguably, many
of these developmental interventions were driven by pragmatism and political
economy contexts and compulsions, and facilitated by the possibility of
experimentation during what Amsden (2007a) has called the First American Empire
(1945–79), whose (p.34) watchword was ‘do it your way’.2 By drawing on the early
development insights of economists such as Gerschenkron and Hirschman, a
relatively coherent body of analysis of developmental states and their rationales
emerged. According to Hirschman:
If we were to think in terms of a ‘binding agent’ for development, are we
simply not saying that development depends on the ability and
determination of a nation and its citizens to organize themselves for
development?…By focusing on determination, for instance, we are taking
hold of one of the specific characteristic of the development process in
today’s underdeveloped countries, namely the fact that they are latecomers.
This condition is bound to make their development into a less spontaneous
and more deliberate process than was the case in the countries where the
process first occurred. (Hirschman 1958: 7–8)

Such a deliberate process of development and catch up cannot be left to market

forces, and calls for an activist or developmental state that has a developmental
agenda as its central goal.
Orientation and Basic Features of Developmental States
The developmental state has been more formally conceptualized and empirically
underpinned by Johnson (1982) in his study of MITI and the Japanese Miracle. Hall
(1987) also shows how the postwar French government played a developmental role
(as does Meisel 2004). In contrast to states with conventional regulatory regimes, in
developmental states the developmental orientation dominates and produces a
different business–government relationship. This is further substantiated by other
development economists, including Amsden (1989), Wade (1990), Chang (1994), and
Evans (1995). Such states are also referred to as hard states and their policies as state
development capitalism, although neither illustrates the nature of developmental
states accurately. Various other names have been coined by different scholars, such as
‘activist state’, ‘promotional state’, Polanyi’s ‘transformative state’, Hirschman’s
‘mid-wifery’ role, and ‘plan-rational capitalism’ (Polanyi 1944; Hirschman 1958;
Johnson 1982). In Johnson’s view, ‘the issue is not one of state intervention in the
economy…all states intervene in their economies for various reasons…the question is
how the government intervenes and for what purposes’ (Johnson 1982: 17–18). From
an empirical perspective, it may be sensible to view (p.35) the developmental state
along a ‘predatory-developmental state continuum’, with the anti-developmental
parasitic state at the one end and the fully transformative and developmental state at
the other.
Mkandawire (2010: 59) argues that ‘developmental states are not an end in
themselves, but an instrument for attaining particular goals—in this case catching up,
rapid economic transformation and growth. So what matters is the collective
aspiration and intent to develop.’ According to Chang (1999), a developmental state
takes long-term growth and structural change seriously, manages the inevitable
conflicts during the process of such change (with a firm eye on the long-term goals),
and engages in institutional adaptation and innovation to achieve those goals. He
(Chang 2003a) stresses the state’s role as entrepreneur and conflict manager. Zenawi
(2012: 169) emphasizes the ‘single-minded pursuit of accelerated development’, the
shared vision of the development project, and autonomy as the basic characteristics of
developmental states. Evans (1997) also highlights the necessity of ‘embedded
autonomy’. He continues that:
the character of the business community can be reshaped by state policy.…
In short, either autonomy without embeddedness or embeddedness without
autonomy is likely to produce perverse results. Without autonomy,
embeddedness becomes capture. Without embeddedness, joint projects that
engage the energy and intelligence of business cannot be constructed. The
state’s contribution to transformation depends on combining the two.
(1997: 74)

In the case of almost all late-developers and late-industrializers that succeeded in

catching up, the state played a visible transformative role. In the Japanese context, for
instance, Johnson (1982: 19) stresses that ‘the state itself led the industrialization
drive, that is, it took on developmental functions’. Kohli (2004: 9) argues that ‘the
way state power is organized and used has decisively influenced rates and patterns of
industrialization in the global periphery.’ Cohesive-capitalist states [developmental
states] ‘have proved to be the most successful agents of deliberate industrialisation in
peripheral countries’. It can be concluded that states with the political intent and
determination to catch up can play a transformative role.
Developmental states, despite their peculiar nature or orientation, should not be
understood as a pure prototype, as no two development paths are identical and none
can repeat itself. It is true that the (p.36) ‘developmental state’ is a classificatory
artifice that, first, tends to be demonstrated by outcomes and, second, gathers under
one rubric a wide variety of actual experiences, institutions, and policies. For
instance, Hobday (2013: 151–2) identifies great variety among the Asian Tigers in
terms of policymaking, capital ownership, industrial structure, and institutional
innovation. Most often the developmental state is a post-hoc simplification of what
was at the time a less clear-cut, more experimental reality, with many mistakes as
well as evident successes. There is also a tendency to use the term loosely and
assume that developmental states can be built regardless of the specific country’s
political economy and history. Doner et al. (2005), for instance, emphasized that
threats, resource scarcity, and systemic vulnerability can play a part (Doner et
al. 2005; Mkandawire 2001; UNECA-AU 2011). Despite some common views, there is
ongoing debate on the context that facilitates the emergence of such states, their
viability in Africa, the role played by democracy, etc. These debates revolve around a
changed international environment less conducive to developmental states and what
some regard as Africa’s near-impossible chances to succeed. The neopatrimonialism
school argues that African states are unable to develop into developmental states.
These views, even when advanced by people considering themselves progressive,
reflect what Hirschman (1991) referred to as ‘reactionary rhetoric’, and in particular
the ‘futility’ and ‘perversity’ variant of such rhetoric.3

The Developmental States of Late Twentieth Century Asia

The developmental state of the twentieth century has been the object of much study
and debate, particularly in relation to the newly industrializing countries (NICs) of
Asia. The developmental state associated with NICs exhibits different features that
reflect their transformative nature. Based on patterns of development and government
behaviour, adherents of the developmental state school have identified the main
features of a developmental state. First is the presence of development-obsessed
political elites, under fully democratic or authoritarian political systems. While the
Korean developmental state, engineered by General Park, was (p.37) authoritarian,
Japan was different, in that a dominant-party system prevailed.4 In Taiwan, despite
the Kuomintang’s long rule, a multiparty system gradually strengthened. Second, ‘the
states know what to do and are capable of doing it. There is then a cognitive question
and a capacity question’ (Herring 1999: 307). This is not dissimilar to Amsden’s (n.d.)
identification of the importance of the ‘right’ role model in the industrial policy
learning process and of a certain level of skills and knowledge. Having an appropriate
role model, but without skilled labour and know-how, cannot be effective; nor can
having excellent skills but an entirely inappropriate role model. Such states are
purposive and have a ‘narrowly-defined national economic goal’ or ‘national
projects’, which are infused with nationalism, as was the case with Japan, Korea, and
Taiwan. Johnson (1982: 19) adds that ‘the very existence of an industrial policy
implies a strategic, or goal-oriented, approach to the economy’. Third, such states
mobilize the society around the national goals and hegemonic purpose. Under
government leadership, a strong alliance between state and industrial class or the
private sector is built around national goals. Such national and social mobilization
requires ‘pilot agencies’ (such as MITI in Japan and Kuomintang in Taiwan).
Although an efficient bureaucracy is indicated as one condition, the vitality of
institutions is also enhanced during the implementation of the ‘national project’. In
Peter Evans’ interpretation (1995), a bureaucracy that is developmental should be
‘embedded’ in society, connecting society with the state (what Buur et al. (2012), refer
to as an ‘embedded and mediating bureaucracy’).
Fourth, such states have ‘autonomy’ and the ability to take decisive political action,
which assumes a broad political base that may arise from the legitimacy of the
economic growth and transformation. A number of developmental states have
implemented land reform, for instance Taiwan, Korea, and Japan, which strengthened
their political base, partly by removing or diluting the influence of large landowners,
who are typically not well disposed towards industry. Fifth, such states have the
political will and capacity to channel developmental rents from less productive
activities to more developmental ones. Wade (1990: xviii) notes that such transfers are
‘often in the form of transfers from unproductive groups to productive groups’ and
‘sometimes in the form of policies to convert unproductive groups into productive
ones’. However, one should (p.38) be cautious about the tendency to make neat
textbook classifications of developmental states. As Fine (1996) argues, Korea and
others did not see themselves as developmental states at the time. The term was
invented later and imposes more order and tidiness on the reality than was the case.
Resource mobilization (through domestic saving and revenue collection) and
channeling into productive investment was an additional feature. Taxation and the
capacity to mobilize resources are believed to indicate state resilience or fragility and
reveal the degree of authority and legitimacy (Di John 2006, 2008). Di John (2006: 1)
argues that taxation could be the ‘principal lens in measuring state capacity, state
formation and power relations in a society’. In addition, industrial peace was
considered important to rapid industrialization and catch up. National consensus,
combined with improved living conditions, appears to maintain industrial peace,
though this has sometimes involved brutal pacification rather than liberal consensus.
For instance, in Korea, stable industrial relations were maintained through a
combination of coercive measures and wage increases following productivity
increases. Seguino (2000) argues that the early industrialization of East Asian
countries relied on exports produced by a high ratio of women employees, whose
wages were artificially repressed.
In conclusion, this section links the older political economy to the idea of the
industrial policy embedded in twentieth century developmental states. It also shows
that there are varying definitions of what this might entail and a range of features and
attributes. The implication is there may be no single blueprint transferable to an
African context, but there may be important features to identify and try to adopt and
adapt to circumstances on the ground.

2.5.3 Financing Industrialization, Managing Rents, and Supporting Technological

Upgrading
Mobilization and apportionment of financial resources during catch up by late
developers is a strategic issue. The importance of resource mobilization and the role
played by investment or development banks during catch up have long been
emphasized in classical works by Gerschenkron (1962) and others. Development
banks have long existed, and are considered the flagships or conduits of
developmental states (Amsden 2001; (p.39) Diamond 1957; Schwartz 2010;
Aghion 1999; Diamond and Raghavan 1982; UNCTAD 2008). Ocampo (2008: 132)
regards them as ‘a major institutional innovation in support of [industrial] policies’,
and emphatically stresses that development banking is a ‘major instrument that has
not been limited by international agreements’. In recent years, the Brazilian
development bank, BNDES, has become one of the largest such banks in the world.
Development banks are defined as institutions established to ‘supply capital and
enterprise in order to speed up the process of development’, and are seen as catalysts
for investment in the private sector (Diamond 1957; Diamond and Raghavan 1982;
UNCTAD 2008). Empirical evidence from elsewhere (for instance, Europe, East Asia,
and Latin America) reveals the roles played by development banks in
industrialization (Diamond 1957; Schwartz 2010; Aghion 1999; Amsden 2001).
Although orthodox development economists have typically looked down on such
banks, other development economists characterize them as ‘flagships’, the ‘nerve
system’ or as a ‘conduit’ for developmental states. The history of development banks
shows their diversity of ownership, purpose, and operating patterns, as well as
performance (Diamond 1957). BNDES, for example, has played a very different role
in Brazil’s structural change as compared to South Africa’s Industrial Development
Corporation. Development banks are also different in their focus on long-term loan
capital, in contrast to equity finance or development corporations (Diamond 1957;
Diamond and Raghavan 1982; UNCTAD 2008; Amsden 2001; Aghion 1999). In
addition, Amsden (2008 1: 11) notes that ‘as development banks imposed operating
standards on their clients, they themselves tightened their own monitoring skills and
procedures’.
In reference to Korea, Amsden (1989, 2001) emphasizes that banks played their role by
applying a ‘reciprocal control mechanism’, by targeting strategic industries,
monitoring their loans, imposing export target conditionality, and lending at lower
interest rates. This is necessary because rents are to be used to foster learning and
catch up, and this depends on disciplining the private sector. Khan and Jomo (2000: 5,
74) define rent as:
 …an income which is higher than the minimum which an individual or firm
would have accepted given alternative opportunities…Rents include not
just monopoly profits, but also subsidies and transfers organized through
the (p.40) political mechanism, illegal transfers organized by private
mafias, short-term super-profits made by innovators before competitors
imitate their innovations, and so on.…Rent-seeking is the expenditure of
resources and effort in creating, maintaining or transferring rents.

Khan and Blankenburg argue that managing rents to promote learning and
technological capability depends primarily on political constraints
(Khan 2000a, 2000b, 2006; Khan and Blankenburg 2009).
An equally important principle is Amsden’s ‘reciprocity principle’ noted above.
Based on the principle of a reciprocal control mechanism, incentives and ‘rents’ were
linked to learning and performance (Amsden 1989). The purpose of rewarding
development rent to industrialists in priority sectors is to reduce the disadvantages of
competing with well-established industries. Amsden (2008: 108–10) emphatically
argues that ‘what lay behind successful post-war industrialisation was a monitored
system of controls on subsidies. Neither import substitution nor export-led growth
were free at all…Performance standards were thus an antidote to abuse and
inefficiency in government intervention.’ According to Amsden (1990), ‘in all late-
industrializing countries—Japan, Korea, and Taiwan included—not only have
governments failed to get relative prices right, they have deliberately got them wrong
in order to stimulate investment and trade’. However, one of the hazards of allocating
rents to firms, Amsden indicates, is that the state will be pushed into rent seeking, and
firms may not have the incentive to improve their productivity.
Ocampo et al. (2009: 156) also argue that ‘incentives should be matched by
performance standards’ and that these should be ‘granted on [a] temporary basis and
dynamically adjusted to move forward in the structural transformation process’.
Incentives and protection instruments need to be monitored constantly, as their
effectiveness in serving industrial policy can fade over time. The allocation of rent, in
the form of subsidy, represents the socialization of the risks faced by firms
participating in a new industry or producing new products, but has to be linked to
performance (Johnson 1982). For instance, South Korea successfully linked incentives
to export targets (SaKong and Koh 2010; Studwell 2013; Rhee, Ross-Larson, and
Pursell 2010; Amsden 1989; Amsden and Hikino 1994). Most development economists
maintain that the East Asian Tigers were more successful than Latin American
countries because the former succeeded in disciplining the private sector. In Korea,
‘the reciprocity principle…operated in almost every industry’ and ‘in return for
protection of the domestic (p.41) market, the government required the enterprises to
export…part of their production’ (Amsden 2001: 149, 151). This is related to linking
incentives to performance, with constant adjustments to new and higher standards;
and states had the power and wit to withdraw incentives that were ineffective, that is,
incentives were time-bound (Ocampo et al. 2009).
Productive rent management and the reciprocal control mechanism are inseparable
from the development of technological capabilities. Such capability is defined as ‘the
ability to use that capacity efficiently’ (Lall 2004). According to Amsden (2001),
technological capabilities can be classified into production capabilities, investment
capabilities, and innovation capabilities. The emphasis on technological development
has increased in the last century with the increased pace of technological progress
(Rodrik 2011; Amsden 2001; Lall 1999). Technology and innovation are the key drivers
of productivity and economic advancement. Manufacturing has greater scope for
contributing to increased productivity, and firms are the main embodiment of this
technological advancement. Contrary to the neoclassical assertion that technology is a
freely available item, in developing or late industrializing countries, technological
supply faces profound constraints. Circumventing them typically takes state
intervention.
The most effective industrial policies have had technological development as a
central component. The depth and scope of technological development may differ in
accordance with the structure of the industry and level of development. Even so,
technological development is crucial to industrial diversification and to industrial
deepening or upgrading. Interventions include developing technological
infrastructure, skills formation, and promoting in-firm technological capabilities
(Lall 2004; Amsden 2001; Rodrik 2011). Depending on the context, policies may
emphasize ‘learning-by-doing’, or/and ‘innovation’ (Amsden 2001).

2.5.4 Complementary Export-Led (ELI) and Import-Substitution Industrialization

(ISI)

Focus on Eli
Industrial policies and industrialization patterns are often depicted as export-
led or import-substituting, orientations that are presented as exclusive rather than
complementary. Hirschman (1968) objected to this dichotomy, as both strategies are
significant and mutually reinforcing. (p.42) For most successful late industrializers,
ISI has preceded exports, and has continued to grow along with the export industry
(Amsden 2009; Lall 2000a; Rodrik 1997; Rodrik 2012).5 Imports and import
substitution are also inseparable, and imports can play a creative role by signalling
demand for domestic manufacturing. Imports are also conduits for technological
development, as domestic producers may improve the quality of their products to
match imported goods. There is a need to strengthen indirect exporters, as these serve
as a conduit for exports and domestic production.
Nevertheless, it is important to distinguish the strategic importance of export-oriented
industry. The industrialization process has to be eventually oriented towards
competition in the international market. Exports will increase the dynamism of ISI by
augmenting demand and economies of scale and furnishing the foreign currency
requirements of domestic industry. Exports play a key role in tackling market-size
limits, improving balance-of-payments, setting high product and productivity
standards, and ensuring a sustainable source of capital formation. Exports are the
only source of autonomous demand (Thirlwall 2002), and play a key role in
disciplining domestic industry and speeding technological development (Lall 2000a).
But it has also been recognized that domestic industrialists are often not enthusiastic
about embracing international competition, while overvalued exchange rates are often
also favourable to domestic ISI industrialists, but can impose structural limitations on
the prospects for outward-looking industry (Hirschman 1968; Thirlwall 2002).
Hobday (1995) and Schmitz argue that latecomer firms face two critical challenges:
being dislocated from major export markets (marketing gap) and from sources of
technology (technological gap). According to Schmitz (2007: 420), ‘most developing
country firms face the problem of being latecomers: they have to make a rapid
transition from patterns of demand and competition typical of a domestic economy, to
global standards of competitiveness.’ Schmitz (2007: 419) presents this dilemma
taxonomically and emphasizes that the key task is combining challenges to firms
(setting targets, exposing them to foreign competition in order to have access to
support) and supporting firms (access to subsidies, technological (p.43) support, or
protection from foreign competition in the domestic market). Different combinations
of support and challenge achieve different outcomes. The Washington Consensus
approach offers low support, while posing a high challenge (international openness).
An active industrial policy combines high support with high challenge. Protectionism
under the old ISI combines high support and low challenge. No policy provides both
low support and low challenge. Based on the relative gaps, Schmitz suggests four
strategic avenues: first, attracting foreign direct investment (FDI) where marketing
and technological gaps are wide; second, integrating into global value chains led by
global buyers where marketing gaps are wide but technological gaps are narrow;
third, opting for joint ventures and licensing where the marketing gap is narrow and
technological gaps are wide; and, finally, exporting domestically designed products
where technology and marketing gaps are narrow. Although this approach is helpful
in designing export promotion strategies, in reality, the situation is more complex.

Attracting and Managing Fdi

Almost all countries use incentives to attract FDI. FDI incentives may be defined as
‘measurable advantages accorded to specific enterprises/groups of enterprises by a
government to encourage them to behave in a certain manner’ (UNCTAD 2000: 11).
Technological development, the dynamic effect of domestic capital, and spillover
effects can be important in attracting foreign investment (Lall 2000a, 2000b). Foreign
investment can be used as a catalyst ‘to enable and to embolden a country to set out
on the path of unbalanced growth’ (Hirschman 1958: 205, emphasis in original). In
addition, it helps by bringing in ‘missing’ factors of production, and may complement
those available locally in the early stages of development of a poor country
(Hirschman 1958, 1963). Foreign capital may also serve as a stopgap measure and
pressure reliever until the image of domestic investors improves
(Hirschman 1958, 1971). To the extent that a sector’s or industry’s organizational
capacity, and the institutionalization of the relationship between that sector or
industry and the state are important, it may be that in some cases foreign investors in
manufacturing bring greater organizational capacity to a sector (Burr 2014; Buur et
al. 2012).
Incentives are secondary to the fundamental determinants such as market size, access
to raw materials, availability of skilled labour, and infrastructure development
(UNCTAD 2000: 11). This includes tax (p.44) incentives. According to UNCTAD
(2000: 19–23), tax incentives are classified into:
reduced corporate income tax rate, loss carry forwards, tax holidays,
investment allowances, investment tax credits, reduced taxes on dividends
and interest paid abroad, preferential treatment of long term capital gains,
deductions for qualifying expenses, zero or reduced tariffs, employment-
based deductions, tax credits for value additions, tax reductions/credits for
foreign hard currency earnings.

Depending on the condition of the latecomer, the role of FDI may be viewed as
indispensable, but not necessarily as a supply of capital. For instance, FDI played a
relatively small role in Japan and Korea, where domestically owned firms were the
key players. FDI played a much more dominant role in Singapore, China, Taiwan,
and Vietnam. The political choice was based on the available avenues and on political
economy factors. Domestic markets played more dominant roles in Japan’s
industrialization, and lesser roles in Taiwan and Korea. However, in the long-term,
domestically owned enterprises are likely to play a more decisive role in
industrialization than foreign-owned firms. They are better motivated to reinvest their
profits within the country, easier to control and ‘discipline’, and arguably are more
prone to invest in research and development, to increase local content and local
knowledge, and to display better risk-taking and entrepreneurship (Amsden 2007b).
The obvious implication is that what matters is less the level and amount of FDI than
its composition, where it is directed, whether states succeed in maximizing domestic
gains from FDI, and how transfer pricing, employment generation, training, and
technical and know-how transfers are managed.

Role of Isi
Many scholars agree that import-substitution is a key transmission belt for promoting
industrial development and acquiring foreign technology. One of the main policy
issues is how to induce ISI firms to export, and so to become competitive. As noted
by Hirschman (1968: 25), ‘many industries…started out producing for the home
market and eventually spilled over into foreign market;…prior, successful acceptance
of a manufactured commodity in the home market has even been considered to be a
prerequisite for successful exporting.’ The possibility of successful exporting also
depends on the nature of the economic actors, a ‘cohesive, vocal, and highly
influential national bourgeoisie’ being more effective than (p.45) traditional
importers. Foreign companies with export experience are also likely to invest in the
manufacturing sector.
Hirschman (1958) argues that the process of import substitution is dynamic and
complex and needs flexible treatment. At the early stages of development, imports
serve as a catalyst for domestic industry through demand-formation, demand
reconnaissance, and reduction of uncertainty, and by reducing selling prices because
of bias against domestic products (Hirschman 1958). With a successful policy, an
import will eventually culminate in what Hirschman calls ‘import swallowing’. This
has to be accompanied by a differently designed protection instruments during the
prenatal stage and when the infant industry is born. Hirschman insists that:
during the prenatal stage, the opposite of the infant industry treatment is
called for if the confinement is to be accelerated. In fact, if it is desired to
prepare the ground for the creation of a particular industry, then it might be
advisable to restrict other imports so as to channel import demand
artificially toward the commodity whose eventual domestic production is to
be fostered’ (Hirschman 1958: 124).

This process will at some point result in ‘import-swallowing’ and the strengthening of
domestic industry (Hirschman 1958: 120–4).
Understanding the dynamics of a specific country’s ISI requires identifying the key
drivers behind it and its probability of success. According to Hirschman (1968, 1971),
there are four distinct origins of ISI: wars, balance-of-payment difficulties, domestic
market growth, and official development policy. Import substitution is more likely to
succeed and be sustainable when it is driven by expansion of the domestic market.
Domestic market growth with a gradual increase of income, coupled with a deliberate
policy to identify bottlenecks and promote investment linkages, is likely to lead to
successful import substitution. ISI fails when the prime motivation is balance-of-
payments constraints or war, Hirschman argued. The failure of import substitution in
many Latin American and African countries from the 1960s to the 1990s depended on
the weaker drivers behind import substitution than the fact of ISI itself.

2.5.5 Linkage Effects as Prime Conceptual Framework

Linkages are a central concept in the history of thought about industrialization, and in
the analytical approach adopted in this book. In particular, (p.46) if linkages matter to
the momentum of structural change, then successful industrialization may involve the
selection of, and targeting of support to, those sectors and activities likely to have the
most, or strongest, linkage effects (Hirschman 1958, 1967, 1986, 1992; Ocampo et
al. 2009). According to Hirschman (1981: 76), ‘a linkage exists whenever an on-going
activity gives rise to economic or other pressures that lead to the taking up of a new
activity.’ Hirschman (1958) also adds that linkage effects may broadly include the
transformation of subsistence goods into commodities, backward linkages (including
the development of the means of production), forward linkages, consumption
linkages, and fiscal linkages (see also Sender and Smith 1986). Backward linkages
involve signalling a lack of—and a potential for—the production of inputs to an
existing economic activity. Forward linkages are created where one activity leads to,
or compels the initiation of, a new activity that uses the output of the original
production as a direct input. Weak linkages among economic activities are typical of
underdeveloped economies.
A linkage approach to economic development led Hirschman (1958: 5) to argue, in the
fundamental proposition of his unbalanced-growth theory, that ‘development depends
not so much on finding optimal combinations for given resources and factors of
production as on calling forth and enlisting for development purposes resources and
abilities that are hidden, scattered, or badly utilized’. A linkage or linkage effect is ‘a
more or less compelling sequence of investment decisions occurring in the course of
industrialization and, more generally, of economic development’ (Hirschman 1992:
56). The linkage concept is thus ‘devised for a better understanding of the
industrialisation process’, for ‘detecting how one thing leads (fails to lead) to another’
(Hirschman 1992: 74). This then leads to a useful policy and investment criterion
based on giving priority ‘to investments with strong linkage effects’, where there is
an absolute necessity for ‘pressure mechanisms’ or ‘pacing devices’. Hirschman
(1981: 75) emphatically notes that:
Development is essentially the record of how one thing leads to another,
and the linkages are the record of how one thing leads to another from a
specific point of view. These on-going activities, because of their
characteristics, push or, more modestly, invite some operators to take up
new activities…Entrepreneurial decision making in both the private and
public sectors is not uniquely determined by the pull of incomes and
demand, but is responsive to special push factors, such as the linkages
emanating from the product side.…the linkage effects of a
given (p.47) products line [are defined] as investment-generating forces
that are set in motion, through input-output relations, when productive
facilities that supply inputs to that line or utilized its outputs are inadequate
or non-existent. Backward linkages lead to new investment in input
supplying facilities, and forward linkages to investment in output-using
facilities.

The idea of linkages is clearly related to but not identical with other concepts. Input–
output relationships are at the heart of backward and forward linkages, but Hirschman
himself was wary of efforts to limit the analytical significance of linkages by
reducing it entirely to input–output table measurements, not least because many
linkage dynamics are not easily quantified. Linkages have something in common also
with the more recently developed idea of supply chains in economic development
(and globalization). Whereas supply chains also involve input–output relationships,
supply chain analysis tends to emphasize the vertical integration of such chains and,
in some iterations, the power relations reaching down through them. Linkages do not
always unfold through such explicitly managed chains and can be more spontaneous
(and more elusive). Further, while linkages may involve, for example, taking up
opportunities for processing domestically available primary commodities or for
minerals beneficiation, they are not restricted to this or to a given supply chain. They
may operate within or across sectors, opening up the scope of both vertical and
horizontal effects.
An example might be the South African firm Bell Equipment. This began in the
1950s, producing equipment for timber and sugarcane cutting, in what was a
backward linkage effect from the sugar and forestry sectors. But Bell’s experience
allowed it to develop the capability, first, and later the ability to identify other
possibilities, notably the production of large earth-moving equipment (articulated
dump trucks) for the global mining sector. Others, meanwhile, have identified a range
of backward, forward, and ‘side’ linkages deriving from the Finnish timber and
forestry industry (see, for instance, Jourdan et al. 2012). These include the
development of specialized chemical and biological inputs, machinery and
equipment, and specialized services (backward linkages); the production of wood
pulp, wood products, including furniture, and construction materials, round wood,
and paper and cardboard, including 25 per cent of the world’s supply of art paper
(forward linkages); and the development of process automation, logistics and
marketing, energy production, and environmental industries (‘side’ linkages). The
concept can expand in different directions, and may (p.48) include fiscal linkages
(even the World Bank and International Monetary Fund now advocate greater
mobilization of fiscal linkages from resource rents in low-income countries) and
spatial linkages where, for example, agriculture can hitch a ride on transport links
developed primarily to serve mining or energy mega-projects.
Hirschman (1967: 5) emphasizes that ‘some projects and technologies have a special
vocation for inducing certain types of learning, attitude change, and institutional
reform (and not others).’ Stimulating linkages that involve manufacturing may be
especially important. This is both because of the special characteristics of
manufacturing in the larger growth process (see above) and because, as has often
been argued, there are more linkages associated with manufacturing than with, for
example, agriculture. This raises a fundamental point about linkages, which is that
they do not necessarily unfold automatically in the cumulative causation process
envisaged by Hirschman. Rather, linkages typically depend on policy. This can be the
macroeconomic policy environment: for example, the compulsion of a potential
linkage may be outweighed, for an entrepreneur, by either inflationary uncertainty or
foreign exchange shortages or rationing. But there are other more direct ways in
which policy matters to unlocking linkage effects. Fiscal linkages from a mining
project may be virtually non-existent or significant, depending on the details of
contracts between states and mining companies. Local content stipulations in
contracts can increase the scope for backward linkages; just as insisting on an
obligatory payroll expenditure on research and development can stimulate linkages.
Both Chile and Norway have implemented policies on backward linkages from the
natural resource sector. By contrast, the demise of the mining research agency
COMRO in South Africa shows how policy can undermine or weaken the scope of
linkage effects in an economy (for instance, see Jourdan et al. 2012). Even in
countries with mineral resources, as in countries without such endowments, steel is
often imported at high cost to nascent manufacturing. The costs and scale
requirements of a steel industry can undermine the development of a linkage to
production. But some countries have reacted to this challenge by establishing state-
owned enterprises to produce low-cost steel for the emerging manufacturing sector
(including POSCO in Korea, the Japan Iron and Steel Company, CSC in Taiwan, Bao
Steel in China, and Rautaruuki in Finland).
An issue complementary to linkages is latitude for failure, another possible source of
signalling available to policymakers. Competition is (p.49) one ‘disciplining’
mechanism that can help narrow the latitude for failure, although it is rarely close to
‘perfect’ and may be best thought of in terms of optimal rivalry and competition (e.g.,
among chaebol in Korea) rather than maximal competition. Competition is neither
the only nor necessarily the most effective means of promoting learning, adaptation,
or productivity. Clearly, the discipline of export competition can narrow the latitude
for poor performance in many industries. In the absence of export discipline, policies
should promote the intensity of competition and domestic rivalry. Narrow latitude can
impel performance, a compulsion to deliver, and put pressure on firms and state
institutions. According to Porter (1998, 2008), two central concerns underlie the choice
of competitive advantage. The first is the industry in which the firm competes, and
the second is its position in the industry. An industry is a group of competing firms
producing similar products or services.
The dynamic potential of linkage effects may also not be released because of lock-in
or path dependency. According to Mahoney (2000: 507), path dependency
‘characterizes specifically those historical sequences in which contingent events set
into motion institutional patterns or event chains that have deterministic properties’.
Path dependency is a formidable problem because of the deliberate actions of actors
to maintain the status quo and to protect their narrow group interests, because of
‘cognitive blindness’, and because of simple inertia among firms, stakeholders, and
government agencies. Again, this serves to emphasize the significance of designing
policy that encourages linkage effects rather than waiting for them to unfold
automatically.
A final point about a linkage approach and its policy significance is that there can
also be employment linkages between different activities. Expanded flower
production in a country like Ethiopia (see below, Chapter 5) not only generates direct
employment but can stimulate additional employment: indirectly, through, for
example, the creation of cardboard packaging plants, logistical services, and cold
storage and air freight facilities; and at another remove, through induced demand for
labour in the services springing up (bars, hotels, construction, motorized rickshaw
transport, etc.). More broadly, indirect employment linkages are a function of
productive linkages: induced employment effects are external effects of investment in
manufacturing other than these linkages—they are effectively a result of Keynesian
multiplier effects (Lavopa and Szirmai 2012). In line with the arguments developed
above, (p.50) manufacturing matters here. Manufacturing generates direct labour
demand, and given that most manufacturing is characterized by relatively high
productivity, this often means relatively decent jobs at reasonable pay. But it is also
known that manufacturing usually has only limited overall employment effects,
accounting for a small proportion of total employment. However, the critical feature
of manufacturing is that it may have relatively high scope for generating indirect and
induced employment across sectors. According to Lavopa and Szirmai (2012: 5), ‘the
evidence suggests that one job created in manufacturing will create a larger number
of jobs in other sectors than one job in any other part of the economy’.

2.6 The African context: why industrial policy and industrialization have failed

2.6.1 The State of African Economies and the Role of Industry

Half a century after independence, Africa’s industrialization and exports lag far
behind the rest of the world. The manufacturing sector in Africa represents 10.5 per
cent of GDP, with small and medium-sized firms dominating (UNCTAD-
UNIDO 2011). Close to 70 per cent of African manufacturing is resource based and
low technology. A significant amount of manufacturing is ‘informal’ (UNCTAD-
UNIDO 2011; Page 2011, 2013; Page and Söderbom 2011). The continent’s share of
world manufacturing value added (MVA) decreased from 1.2 per cent in 2000 to 1
per cent a decade later (UNCTAD-UNIDO 2011). Primary commodities continue to
dominate exports, with overdependence on a few commodities, exposing African
countries to vulnerability and external shocks (Cramer 2012; Sender and Smith 1986;
UNCTAD-UNIDO 2011; Soludo, Ogbu, and Chang 2004; Lall 2000b). Africa’s share of
global manufacturing and industrial exports has been persistently insignificant
(UNIDO-UNCTAD 2011). Between 2001 and 2006, sub-Saharan Africa’s (SSAs)
share of manufacturing exports was 0.558 per cent, and only 0.225 per cent when
South Africa is excluded (Jomo and Arnim 2012: 511). As Table 2.1 below shows, the
low share of manufacturing in the economy indicates low development and lack of
structural transformation (Szirmai, Naudé, and Alcorta 2013; Page 2011, 2013).
(p.51)
Table 2.1. Regional structure of production, 1950–2005
Region 1950 2005

Agriculture Industry Manufacturing Services Agriculture Industry Manufacturing Service

Africa 43 22 11 34 28 27 10 45

Asia 49 14 10 36 14 33 22 53

Developing 37 22 12 42 16 31 15 53
countries

Advanced 16 40 29 45 2 27 16 71
economies

Note: Gross value added as percentage of GDP at current prices, regional average.
Source: Szirmai, Naudé, and Alcorta (2013), Table 1.2, pp. 11–12.
Africa’s economies continue to face fundamental structural constraints, including
lack of economic diversification, the prevalence of unregulated and often low-
productivity economic activities, low productivity in industry and agriculture, and
poor infrastructure (World Bank 2010; Commission for Africa 2005). Africa is
estimated to lose an estimated 1 per cent of GDP to poor infrastructure alone, which
is much higher than in other regions (World Bank 2010). This situation, and a weak
regional market, undermine industrial competitiveness. Khan (2012: 439) argues that
employment intensity and the labour force participation rate in SSA were much lower
than in Asia. Moreover, low saving capacity, weak investment, and limited capital
have been additional constraints (UNCTAD-UNIDO 2011).
Since 2000, many African countries have begun to register economic growth,
primarily from resource extraction and commodity price booms. This growth
continues to be led by the very factors responsible for previous African growth
episodes, which did not, however, engender sustained rapid growth and structural
change—inward capital flows, cheap capital, and surging commodity prices. Overall,
Africa’s economic growth has been slower than Asia’s or Latin America’s, but the
continent has also experienced periods of faster economic growth and economic
success (Jerven 2010a, 2010b), for example, during its so-called Golden Age (1960–
75). Countries such as Botswana and Mauritius more recently, and Côte d’Ivoire in
the 1970s, were commended for their somewhat activist industrial policies (Bhowon,
Boodhoo, and Chellapermal 2004; Stein 2006; Soludo, Ogbu, and Chang 2004).
Supported by cheap borrowing, capital flows, and high commodity prices, these
activist states played an important role (p.52) in achieving economic growth spurts in
their countries (Ocampo et al. 2009). However, nowadays African governments are
mostly advised against adopting an ‘activist’ stance on the basis of previous growth
collapses. Neopatrimonialism, invoked increasingly since the early 1990s, is said to
explain why African states can be neither developmental nor implement activist
industrial policies (see Altenburg 2011; Chabal and Daloz 1999). A recent piece of
neoclassical advice to African countries has been to focus on factor endowments and
light manufacturing industry (Dinh et al. 2012). This counsel tends to come out of an
economic tradition that was also at the heart of the Washington Consensus. Yet
Africa’s slowest growth was recorded precisely during the 1980s and 1990s, the very
period when African governments were preoccupied with implementing the
Washington Consensus prescriptions.
Recent empirical studies and contributions to the debate on Africa’s industrialization
highlight the political constraints on, and the acute deficits in, industrial
policymaking. Cramer (1999a), drawing on a case study of Mozambican processing of
primary commodities, cautions against overgeneralization but highlights that the key
constraint on Africa’s industrialization has been ‘political rather than purely technical
or economic…the state lacks the capacity or will to produce a coherent and emphatic
analysis and policy package for industrial sectors.’ Warren-Rodriguez (2008)
highlights that a ‘deteriorating policy and economic environment’ was the major
factor in undermining technological development in Mozambique. As a result, the
country failed to optimize its industrial drive through accumulation of technological
capabilities and manufacturing skills. ‘Wrong policies’ have not only led to lack of
industrialization, but have in some cases generated deindustrialization (decline in
manufacturing share). Tregenna (2008a, 2008b, 2012) provides a comprehensive
structuralist explanation based on extensive empirical evidence from South Africa.
While emphasizing the ‘special characteristics of manufacturing’, Tregenna (2008a)
highlights recent deindustrialization and the weakness of industrial policymaking in
South Africa. Moreover, he (2012) argues that South Africa’s dependence on
‘domestic demand expansion’ as the main source of growth has contributed to weak
subsector dynamics, including sluggish technological change. Industrial policy
appears to have been unable to effectively promote export capacity and global
competitiveness.
Deindustrialization has not been limited to South Africa, but has been a feature of
SSA countries at lower levels of development too (Jalilian
and (p.53) Weiss 1997, 2000), with negative consequences for long-term productive
gains. The overall policy environment and structure of African economies in the early
twenty-first century remain fundamentally unchanged, although there are some
positive departures. For instance, despite limits to Mauritius’s transformation and
upgrading (the so-called ‘Mauritius Miracle’, it is one case of successful
industrialization and industrial policy (Stiglitz 2011; Rodrik 1997; Bhowon et al. 2004;
Ancharaz 2003; Rodrik 2012). (It is surely curious, and linked to the fatalism common
to many economists, that the reflex reaction to any remotely successful case of
industrialization is to dub it a ‘miracle’.) Brautigam and Tania (2009) provide
explanations for Mauritius’s achievements, highlighting a visionary elite and societal
consensus on national vision, transnational networks (such as export-processing
zones, EPZs), and systematic vulnerability. Its per capita income rose from $400 in
1968, the year of independence, to about $15,000 in 2012, and the economy
diversified from monocrop dependence (sugar) into industrial and services
sectors.6 Mauritius developed EPZs, protection tariffs, and incentives; and
innovatively adjusted to changed domestic and international variables. This was
accompanied by a positive social and political framework, specifically a sound
democratic and welfare system (free education, free health service).
Tregenna more recently (2013) notes that between 2000 and 2007, eleven African
countries showed an increased share (at least by 1 per cent) of manufacturing in GDP,
eight of which had actually experienced deindustrialization from 1990 to 2000.
However, not all industrial policy instruments have been successful across the board.
Stein (2012: 322–39), for instance, maintains that despite their wide adoption, EPZs
failed in many African countries because they were not part of a broader industrial
policy and industrialization drive. He points out that of the 3,500 zones worldwide
employing 66 million people in 2006, only ninety-one (2.6 per cent of those in
developing countries) existed in twenty SSA countries, employing only about one
million people. Most EPZs failed as most were initiated for the short-term goal of
benefiting from the African Growth and Opportunity Act (AGOA) or multi-fibre
arrangement.
Despite the recent global economic crisis, international factors have been broadly
favourable to African countries. These factors include high (p.54) commodity prices,
cheaper borrowing costs, rising foreign investment inflows, increased labour cost in
emerging countries, and growing South–South economic ties in investment, trade,
and financing. For instance, FDI to Africa has increased fivefold in recent years
(AfDB 2011). Brautigam (2011) and Jianhua (2013) highlight the increasing Chinese
presence in Africa and its implications for African industrialization. Domestic
conditions, such as growing investments in human capital (education and health),
political stability, and state institutions have also shown positive developments.
Regional cooperation is increasing (AU, NEPAD, and sub-regions), while regional
conflicts show a declining trend (HSRG 2012).7

2.6.2 African Development and Industrial Policy Literature

A review of African economic development and industrialization raises fundamental
issues about analytical perspectives (Sender and Smith 1986; Cramer 1999b;
Riddell 1990; Jerven 2011). Most available literature is presumptuous, riddled with
gaps, and metes out standard pronouncements. Common diagnoses (for instance, the
African ‘growth tragedy’) and the resultant conclusions are not founded on reliable
data or appropriate methodology, and have often reflected political bias. Riddell
(1990) stresses how scarce data on industrialization in Africa are, one instance of the
larger issue of ‘poor numbers’ on African economic activity (Jerven 2011). One of the
major problems in the African political economy literature is the preoccupation with
the source of the ‘Africa dummy’, a residual in growth equations that cannot be
otherwise explained by neoclassical growth models. This quest involves the
elaboration of econometric models with ever-expanding variables, most of whose
assumptions, data, and technical methodologies are not context appropriate
(Jerven 2011; Sender 1999, 2003; Oya 2012). Many of these ingenious statistical
exercises, as well as many other analyses, treat the continent as a homogeneous
entity, despite the huge diversity between and within its fifty-four countries. This
diversity is partly highlighted by the different growth and economic development
patterns between states (see, for instance, Sender, Cramer, and Oya 2005). Such an
oversimplified approach, however, has led to unrealistic expectations and
disappointing outcomes.
(p.55) The second major problem with existing research on African political economy
is myopic Afro-pessimism, as a result of which misrepresentation and ideologically
driven bias have significantly clouded reality. This mindset presumes an African
growth tragedy and then attributes this to cultural factors and ethnicity, among others
(Easterly 2002; Easterly and Levine 1997). This outlook is aggravated by market
fundamentalism, which blames state intervention for Africa’s underdevelopment and
lagging industrialization. It is difficult not to see in this the contempt Hirschman
(2008) believed characterized many mono-economic approaches to explaining the
state of Africa.
2.6.3 Background to Africa’s industrial policymaking
Industrial policymaking has gone through different phases in Africa. The first phase,
between the 1960s and 1980s, was driven by vigorous industrial promotion in many
countries aiming for self-reliance and import substitution. African nations embarked
upon this quest after learning that political independence does not automatically
guarantee economic independence and a better quality of life. Nonetheless, this
industrialization drive and the resultant economic growth could not be sustained
because of balance of payment problems, debt burdens, and the inability to build
internationally competitive industries. The final straw was the 1970s oil crises. Poor
policies pursued by many African governments during that period—among them,
failure to invest in export sectors, neglect of high productivity agriculture,
mismanagement of macroeconomic policy and state-owned enterprises—contributed
to the crisis.
This period also witnessed some less-than-successful policy experiments, often
coloured by populism, such as Ujamaa or ‘African socialism’ in Tanzania. The
Zambian mismanagement of copper deposits, war and policy crises in Mozambique,
and macroeconomic policy struggles in Ghana and elsewhere are other examples
(Kitching 1982; Sender and Smith 1986). Kitching (1982) argued that populist and neo-
populist ideology lay behind experiments such as Ujamaa, a widely influential
initiative that led to a bias in favour of a non-industrial path to development, an
option that is not viable in theory or practice. In his book Manufacturing Africa,
Riddell highlights the neglect of industrialization policy between the 1970s and
1990s, and contrasts this with the focus on Africa’s industrial development during the
1960s. These decades were characterized by (p.56) neglect of the manufacturing
sector, the weak structure of manufacturing, low and falling manufactured exports,
lack of interlinkages within manufacturing, and weak forward linkages.
The Structural Adjustment Programme (SAP) phase (1980–2000) was characterized
by significant policy changes, which resulted in slower industrialization and
deindustrialization (Riddell 1990; Padayachee 2010; Tregenna 2013). Most African
countries were forced to undertake reforms focused on economic liberalization,
privatization, and structural adjustment (Rodrk 2014a). Jomo and Arnim (2012: 511)
underline how ‘economic liberalisation has brought economic stagnation, de-
industrialisation and agricultural decline, rather than structural change induced by
productivity gains and stronger domestic demand from increasing incomes’.
Industrialization in many countries continued to play a minor role, and the economic
reforms promoted by International Financial Institutions (IFIs) did little to promote
new industries (Watanabe and Hanatani 2012). Under pressure from the IFIs, most
public enterprises were privatized, although this did not lead to significant
productivity gains. Trade was liberalized and domestic firms were unable to compete
in the international market without protection measures (Riddell 1990). This painful
‘adjustment process’ also had social costs such as layoffs, and the state’s role was
weakened.
Agriculture, the traditional economic mainstay, was plagued by problems and failed
to increase productivity and transform. Support to farmers and state interventions
supposed to improve agricultural productivity were largely ignored (Oya 2010, 2011;
Sender 2003; Sender, Oya, and Cramer 2006). Many African countries abandoned
industrial policies, and the opening of financial markets to foreign banks weakened
domestic capital. There is also evidence of a loss of policy capability (Warren-
Rodriguez 2010; Palma 2003, 2009, 2011, 2012). Once these reforms were considered
outdated, the poverty reduction strategy and Millennium Development Goals were
introduced, also narrowly focusing on social goals and poverty reduction rather than
addressing fundamental growth and economic transformation. Indeed, the poverty
reduction and social goals are more likely to be achieved through successful
economic development and structural changes (Ocampo et al. 2009; Amsden 2009).
Despite the fragmented use of industrial policy instruments, industrial policy has
been based on a neoclassical vision of markets. There are two ways in which this may
be the case. One is that for a long time, industrial policy has essentially been
about not interfering in markets, and not (p.57) promoting industry, etc. The other is
the newer neoclassical version, where a moderate effort to facilitate industrialization
is accepted, but not more activist and higher risk interventions. Tregenna (2013)
underlines the importance of increasing labour productivity in manufacturing,
highlighting the dynamic role it played in East Asia. This is a dilemma for many
African countries, as low wages are not an advantage until productivity reaches
international standards (Schwartz 2010). These observations call for new approaches
and policy perspectives that can lead to economic transformation and for more fitting
industrial development perspectives for Africa. One of Riddell’s conclusions from the
case studies in seven African countries is that ‘there does not appear to be one
particular road to industrialization or mould into which either these or other SSA
countries could or should be made to fit’ (Riddell 1990: 47). He adds that:
the long-term prospects for the development and deepening of the
manufacturing sector in SSA, in general, and for the seven case-study
countries, in particular, will be critically determined by the nature of the
policy environment, the incentive system in which manufacturing
enterprises operate, and by policies and stimuli targeted specifically at
firms within the manufacturing sector (Riddell 1990: 51).

2.7 Summary and Conclusions

The theoretical and conceptual constructs used in this book are summarized as
follows. First, mainstream economic theories emphasize that there is no case for
active industrial policies and that free markets and trade are superior to
interventionism in terms of economic growth. At best, the ‘new development
economics’ (Fine 2006) builds on neoclassical foundations to argue that ‘market
failures’ are fairly widespread and do justify limited corrective state intervention,
typically to be tailored to ‘state capacity’. From this perspective, even where there is
a case for industrial policies (because of market failures or ‘imperfections’), there are
still grounds for extreme suspicion of state intervention on the basis of rent seeking
and political economy considerations. As Mazzucato (2013a: 21) put it: ‘Economists
willing to admit the State has an important role have often argued so using a specific
framework called “market failure”. From this perspective the fact that markets are
“imperfect” is seen as the exception, which means that the State has a role to play—
but not a very (p.58) interesting one…’ (my emphasis). Neoclassical economists also
typically reject claims that there are sector-specific dynamics in development.
Second, the theory of infant industry has significant policy implications, as it gives
prominence to manufacturing and the promotion of new industries. The theory refutes
international free trade and posits that protectionist policy is necessary to develop
new industries. It also accepts the legitimacy of the state’s role in developing these
capabilities. These principles and concepts are still relevant in the contemporary
world (despite World Trade Organization [WTO] rules, phyto-sanitary standards,
etc.).
Third, the structuralist and catch up perspectives provide the basis for industrial
policy and a strong rationale for deliberate intervention to generate dynamic gains
from industrialization. This perspective also calls for a sectoral approach that is based
on manufacturing’s greater scope for increasing returns in comparison with mining
and agriculture. The dynamics of sectors can be best explained by Hirschman’s
linkage effects (with varying spillover effects).
Fourth, the role of industrial policies is to ‘stimulate the sectors with increasing
returns while shifting resources from elsewhere in the economy’ and to foster
economic transformation by promoting the ‘ability of an economy to constantly
generate new dynamics’ (Ocampo et al. 2009: 10). This perspective also stresses that
each country follows its own development path with specific characteristics that are
usually not replicable. Industrial policies are basically dependent on the dynamic
nature of the state as well as the broader political economy, in particular the
relationship of the state with the private sector and other social groups.
Fifth, overall, it is clear that Africa has not been successful to date in industrializing
or in industrial policymaking. Furthermore, all the conventional policy prescriptions
from the 1980s to 2000s have not achieved structural transformation in Africa.
Hence, active industrial policies for achieving this end have to be explored. This book
aims to contribute in this regard by using comparative case studies from Ethiopia.
This chapter has also highlighted the dimensions of industrial policy, in particular the
strategic orientation towards exports and accumulation of technological capabilities
and the productive use of rents through reciprocal control mechanisms and
performance standards. In addition to policy capabilities and institutional innovations,
linkage effects (including latitude for performance standards) can serve as an
important (p.59) conceptual framework and policy guide. This analytical framework
provides the foundation for exploring industrial policy and varying outcomes across
three sectors in Ethiopia. A thorough review of the literature on industrial policy over
time indicates that all countries can make the transition from agriculture to industry,
but the catch up can vary among countries depending on their peculiar national
situations, political economy, history, and the international environment.

Climbing the ladder!

What does this extensive literature review tell us about the challenges of industrial
policy and late industrialization, and what are the key drivers? The empirical record
discussed above shows clearly that catching up and structural transformation do not
depend on ‘laissez-faire’ and are not easy. As Friedrich List warned, forerunners ‘can
do nothing wiser than to throw away the ladder’ and to ‘preach to other nations the
benefits of free trade’. This still holds true in the twenty-first century. Despite the
importance of the changing global context, the virtues of free trade do not determine
the absolute destiny of nations.
As Hirschman emphasized, development is determined neither by scarcity of
resources nor by various ‘prerequisites’. Structural transformation and catching up go
far beyond fixing ‘market failures’, and depend on a developmental perspective and a
transformative state, which are in turn reliant on the political economy of each
country. Industrial policies in developing countries should thus address three
important aspects that lead to successful catch up: ‘innovation in new economic
activities or new ways of doing (in a Schumpeterian sense); linkages (Hirschman);
and surplus labour (Lewis)’ (Ocampo 2007: 1). Countries have to find their own paths
of development that reflect their own peculiarities and ‘advantages of late
development’. What implications does this discussion have for Ethiopia? The
following chapters discuss how climbing the ladder has been undertaken in twenty-
first-century Ethiopia, showing both what is possible and the formidable challenges.

Notes:
(1) For a critique of the empirical record on the disadvantage in the terms of trade for primary
commodities vis-à-vis manufactured goods and on the effects this notion has on policymakers and
advisers in Africa, see Sender and Smith (1986).

(2) As opposed to the Second (ongoing) American Empire’s motto, ‘do it our way’.
(3) According to the futility thesis, progressive proposals for social change may be
accepted as highly desirable but dismissed as impossible. The perversity thesis
suggests that similarly agreeable proposals should not be attempted, because they will
probably lead to perverse, disagreeable outcomes.
(4) See also Selwyn (2012) on relationship with labour.
(5) Rodrik (2012: 169) highlights that ‘…the overall record of ISI was in fact rather
impressive. Brazil, Mexico, Turkey, and scores of other developing nations in Latin
America, the Middle East, and Africa experienced faster rates of economic growth
under ISI than at any other time in their economic history.…Industrialization drove
this performance’.
(6) Unless specified, the currency is in US$ throughout the book.
(7) Human Security Research Group.

Setting the Scene

Ethiopia’s Industrial Policies and Performance
Chapter:
(p.60) 3 Setting the Scene
Source:
Made in Africa
Author(s):

Arkebe Oqubay
Publisher:
Oxford University Press

DOI:10.1093/acprof:oso/9780198739890.003.0003
Chapter 3 maps out the history and foundations of policymaking in Ethiopia.
Economic growth has been impressive although structural transformation lags far
behind, and manufacturing has played an almost insignificant role in the Ethiopian
economy. The Ethiopian government has certainly engaged in policy experiments and
largely maintained its autonomy. This chapter suggests that Ethiopia in recent years
has exhibited features of ‘developmental state’. This chapter introduces the main
industrial policies and policy instruments developed in Ethiopia in recent years. This
is done both to clarify the context of policymaking in different sectors and to identify
national-level patterns in policymaking institutions. Thus, as well as providing a
valuable reference resource, this chapter sets out the broader framework relevant to
understanding the industrial policy in the different sectors addressed in this book.
Keywords: Ethiopia, developmental state, policy
instruments, policies, policymaking, institutions, industrial policy, cement, flowers, leather

Prior experience of manufacturing is regarded by some (for instance, Amsden 2001)

as important to the prospects for industrial policy. Ethiopia does have a history of
manufacturing, though little is known about the mid-twentieth-century
deindustrialization of the country. The British removal of Ethiopian manufacturing
equipment and factories built by Italians was justified on grounds that Ethiopia was
‘over capitalized’ and that ‘Ethiopians have no mechanical aptitude’. According to
Richard Pankhurst (1996: 43), who chronicled these events:
the Italian fascist surrender at Gondar, on 27 November 1941, marked the
end of the British East African Campaign. Five days later the British
military authorities made their first detailed proposals for dismantling of
Italian Assets in Ethiopia…Ethiopia at the end of the Italian occupation
 struck them as better equipped than several long established, but little
developed, British colonies or dependencies from which they concluded
that the country had been highly, and artificially, industrialised.1

This chapter discusses Ethiopia’s recent experiences with industrial policies designed
to recreate industrial capacity and competitiveness.
Ethiopia is one of the poorest LDCs, although the country has maintained its
independence throughout its long history, dating back to the early Aksumite
civilization of the first millennium BC. The country was ruled for (p.61) almost half a
century by Emperor Haile Selassie, an absolute monarch who was toppled in 1974 in
a popular revolt. A military junta, the Derg, took power in 1975, and established
totalitarian rule and a USSR-type command economy. Throughout much of the
twentieth century, Ethiopia was identified with famine, conflict, and war, but by the
early twenty-first century it was becoming better known for its aspirations to become
one of the greatest success stories in Africa. The 1983–85 famine in Ethiopia affected
close to eight million people and led to more than 400,000 deaths (some estimates put
this figure at one million). It was described by the BBC as ‘a biblical famine’ and ‘the
closest thing to hell on earth’.
Less than a decade after the great famine, things started to change. By the early
1990s, the country was emerging from totalitarian rule and three decades of war, and
confronted a rapidly declining economy and widespread marginalization of its
diverse population as a result of highly centralized Derg rule. Totalitarian rule had
ended in 1991 following the defeat of Africa’s biggest army by the Ethiopian Peoples’
Revolutionary Democratic Front (EPRDF), which led the liberation movement
(EPRDF 2011b; Balema 2014). This was followed by a referendum in Eritrea (the first
in Africa), which led to Eritrea’s peaceful independence. The EPRDF government
then established a federal system, and introduced significant administrative and fiscal
decentralization. It also pursued an agricultural development-led industrialization
(ADLI) strategy, which focused on reducing poverty and stimulating the economy
(FDRE 1994, 1996). The strategy assumes that initial take off depends on stimuli from
agriculture in terms of growth in demand, supply of foreign exchange for machinery
imports, and inputs for factories.
The government has since adopted a Growth and Transformation Plan (GTP) that
aims for rapid growth and structural change and the promotion of industry as the
leading sector of the economy by 2020 (MOFED 2010). The overriding objective of
GTP is to make Ethiopia a lower middle-income country by 2025. This ambitious
vision and strategy entails pursuing an appropriate industrial policy, mobilizing the
population around the vision, and effective state leadership. Achieving these
objectives involves a challenging combination of ideas, policymaking and
implementation institutions, and a coalition of interests sufficiently robust to facilitate
policy implementation and to lend legitimacy to the state’s strategy.
Ethiopia has now been able to reduce the number of people living in poverty from 54
per cent in 1992 to 27 per cent in 2012, while life (p.62) expectancy has increased
from forty-seven to fifty-nine years (by comparison the average across SSA is five
years, from fifty to fifty-five) (World Bank 2013). However, these indicators do not
tell the whole story, nor are they a strong basis for celebration. There is a need to both
extract ‘lessons’ and acknowledge and engage with the massive challenges and
constraints Ethiopia still faces. A constructive approach is to focus on policymaking,
on understanding the factors that positively contributed to outcomes, and on
emerging challenges in order to sustain the gains made.

3.1 Genesis of Ethiopian industrialization

Ethiopia emerged as a united nation and state in the late nineteenth century, and
Ethiopia’s industrial development can be traced to the early twentieth century
(Zewde 2002a, 2002b). Like other African nations, Ethiopia has gone through stages of
industrial development. Manufacturing accounted for about 1.5 per cent of GDP
under imperial rule in the 1950s. The sector grew to 5 per cent in the 1970s, and was
dominated by foreign owned firms and ISIs.2 In the 1970s, there were about 300
foreign firms, accounting for more than 75 per cent of the industrial sector. The
Confederation of Ethiopian Labour Unions (CELU) had about 50,000 members. The
manufacturing sector was limited to the production of consumer goods, and its
growth was constrained by the small market size and insignificance of the private
sector, as well as its weak forward and backward economic linkages. Land ownership
was exploitative and limited to aristocrats, the royal family, and the church.
Consequently, agriculture stagnated until the mid-1970s. This was a feudal economic
system, rooted in autocratic rule and based on a parasitic aristocracy. The period was
characterized by slow economic growth, which led to the popular revolt in 1974 that
ended both Haile Selassie’s reign (begun in 1934) and the monarchy.
During the Derg period (1974–91), all means of production were put under state
control and administered under a socialist economic management system
(Balema 2014; Tekeste 2014). Industrialization was guided by a command economy
and import substitution, and economic activities were largely designed to support the
Derg’s war machine. For instance, Asco (a shoe factory) was entirely given over to
the production (p.63) of army boots. Garment and food processing factories had also
to service the Derg’s 600,000 soldiers. Armament factories were also established,
with loans and technical assistance from Eastern Europe and North Korea. ISI was
low technology and focused on light industry, as were state-owned enterprises. The
policy had an anti-export bias, which also inhibited private sector involvement.
Private sector investment, such as it was, was limited to small enterprises such as
grain mills. Moreover, labour and population mobility was tightly restricted.

3.1.1 Industrial policy post-1991: Challenges and opportunities

The chief aim of this chapter is to outline in detail the process of industrial
policymaking in Ethiopia post-1991, in the context of dramatic political changes
under the EPRDF-led government. During this period, the government has been
heavily engaged in major state-building as a precondition for transforming the
country and achieving middle income status by 2025. Phase one covered the period
up to 2002. It was a period of transition from war to peace, from command economy
to market-based economy, and from totalitarianism towards a more democratic
system (Ottaway 1999). The Transitional Government of Ethiopia (TGE), a
provisional coalition, introduced a new constitution in 1995. This stipulated a
multiparty political system, separation of powers within government, and federalism.
Major political changes and economic rehabilitation and restructuring followed. The
federal system established followed new political and fiscal models (Balema 2014).
The foundations for a market-economy were laid. Recovery and rehabilitation of
infrastructure and basic services were undertaken. The revitalization of agriculture
was key to political stabilization, the activation of productive forces, and the
stimulation of the economy. These policies were also preparing Ethiopia for industrial
development. Domestic and foreign investment started to expand slowly. Later, the
growth momentum was disrupted by the costly Ethio-Eritrean war of 1999–2000. In
addition to the loss of life and physical resources, GDP growth slowed to 1.6 per cent
in 2002 and –2.1 per cent in 2003 (MOFED 2010). Thus, although the economy grew
at about 5 per cent per annum during this period, this growth was erratic.
The second phase, from 2003 to 2012, is characterized by rapid growth and economic
development. Average annual economic growth was 11 per cent between 2004 and
2012, almost double the SSA average for the same (p.64) period
(MOFED 2012a, 2012b). This period featured a clearer national vision, with national
strategies and policies in key sectors, which were influenced by East Asian economic
growth and developmental practice. The industrial development strategy focused on
labour-intensive, export-oriented sectors, such as textiles and leather. Investment
promotion became more focused, and Ethiopia attracted more FDI than in years past.
The flow of FDI has increased from almost nil in 1992 to more than half a billion
dollars in recent years. Corporate governance systems in public enterprises and
investments in the commanding heights were further consolidated, while many state-
owned enterprises (SOEs) were privatized to strengthen the newly flourishing private
sector. Moreover, federalism and decentralization deepened, and new urban
development reforms were initiated, increasing industrial and agricultural growth.
Support to small farmers grew, including the establishment of market institutions and
the development of rural infrastructure. Overall, infrastructure and human
development became a national priority.
After 2010, a five-year GTP began to be implemented, with rapid growth in
manufacturing and structural transformation of the economy as its central aim. For
instance, the target growth rate for manufacturing is 20 per cent per annum. Exports
are planned to grow fivefold, with an increased share for industrial exports. The GTP
also aims at the development of 2,000 km of railway lines and a fivefold growth in
power generation (from 2,000 MW to 10,000 MW) to help the process of
transformation. Moreover, the domestic savings rate has to more than double in five
years. The GTP emphasizes linkages within the economy, primarily between
manufacturing and agriculture, and the creation of a single economic space.
The GDP growth rate has been 10.6, 11.4, and 8.7 per cent in 2010, 2011, and 2012,
respectively (MOFED 2013a). Per capita GDP reached about $558 in June 2013 (more
than four times higher than in 1991, when it was $120). The ratio of domestic saving
to GDP increased from 5.2 per cent in 2010 to 17.2 per cent in 2011. However,
bringing down inflation to single-digit figures has been more challenging than
government expected. Public investment has also shown significant growth from
barely Ethiopian Birr (ETB) 1 billion or $0.2 billion in 1991, to ETB 53 billion in
2011 ($3.5 billion) (MOFED 2012a). Capital expenditure (public investment) was 22
per cent of the total federal budget in the early 1990s and exceeded 56 per cent in
2011. This contrasts starkly (p.65) with many African countries. Ethiopia’s budget
allocation has focused on poverty reduction, agricultural development, infrastructure,
and human development (MOFED 2013a). Military expenditure had been
significantly cut to 1.1 per cent of GDP in 2011, down from 6.5 per cent under the
previous regime in 1990. The government has used SOEs to undertake massive
investments in energy, telecom, and rail networks (MOFED 2012a, 2013a).
Despite government efforts to put the economy on a high growth path, the share of
manufacturing in GDP remains low and the 10 per cent annual growth in national
industrial output has not exceeded GDP growth, a necessary condition for change in
the composition of the economy. The sector also continues to be characterized by
many structural constraints, such as low productivity, low value, and lack of
international competitiveness. Export volumes have increased, although the export
structure has not fundamentally changed. There has been diversification, but within
primary commodities, which has reduced the heavy dependence on coffee (from two-
thirds in the 1990s to about one-quarter in 2012). Foreign currency shortages have
become pervasive, despite the increased share of exports from 3 per cent of GDP in
1992 to 17 per cent in 2011. This can be contrasted, for instance, with Mauritius’s
exports, which accounted for 59 per cent of GDP in 2011. Ethiopia’s imports of
goods and services reached 32 per cent of GDP in 2011, thus highlighting the severity
of the balance of payments problem.
One major constraint on rapid growth and catch up is the inability to mobilize
resources and concentrate on investment. Resource mobilization requires social
mobilization, institutions, culture change, and increased growth dynamics. Ethiopia’s
gross savings are among the lowest in Africa, despite recent improvements. Arguably,
the government’s approach appears to resemble the Keynesian perspective, where
investment is expected to drive savings up, rather than, or as well as, the other way
round. The East Asian developmental states (notably Korea and Taiwan) have been
effective in resource mobilization (Amsden 2001; Wade 1990). The powers to levy and
collect tax and revenues are defined in the Ethiopian Constitution as well as in
various federal and regional proclamations. However, Ethiopia’s central government
tax revenue as a share of GDP is also among the lowest in Africa. As a result, revenue
and tax reform has become a major government priority. Tax reforms include
improving the tax information system (introducing sales register (p.66) machines,
automating import–export customs, and assigning tax identification numbers to
taxpayers), public tax education and mobilization, enforcement, and strengthening the
tax authority. Despite efforts to increase awareness of taxpayers’ obligations,
corruption and contraband crimes remain major challenges. Progress has been made
since 2005, with government revenues boosted from ETB 11.2 billion in 2006, to
ETB 59 billion in 2011, and ETB 84 billion in 2012. Tax revenue has gradually
increased from 5.6 per cent of GDP in 1992 to 11.5 per cent in 2011.3

3.2 Patterns of industrial development in Ethiopia

The Ethiopian manufacturing sector has two distinct features. First, there is a low
level of industrialization in terms of the sector’s share in GDP, export earnings,
industrial intensity, and competitiveness. Second, the industrial structure is dominated
by small firms and resource-based industries (in particular the food industry), and
concentrated around the capital city. These features are explored below.

3.2.1 Low industrialization

Ethiopia’s low level of industrialization is well documented by government agencies
and in international reviews. Industry is composed of manufacturing, mining, and
construction. First, the share of industry (13 per cent) and of the manufacturing sector
(5 per cent) in the economy is very low, and has shown little change between 1992
and 2013, even exhibiting some decline (see Figure 3.1). It is much lower than the
10.5 per cent average share of manufacturing in the economies of SSA countries
(UNCTAD-UNIDO 2011). In 2009, African industry accounted for 40.7 per cent of
GDP. The corresponding figure for East Africa is 20.3 per cent for industry and 9.7
per cent for manufacturing.
While manufacturing industry has been growing rapidly, it slightly falls short of the
growth rate of the overall economy (MOFED 1999; Tekeste 2014). The average annual
growth rate of the manufacturing industry was 11 per cent between 2004 and 2014,
compared to the 10.9 per cent of the (p.67)

Figure 3.1. Contribution to GDP by economic activity at constant prices

Source: Unpublished data of National Economic Accounts Directorate, MOFED
(December 2012 and October 2013)

Figure 3.2. Real GDP growth rates 2004–13

Source: MOFED, December 2011 and October 2013
average GDP growth rate during the same period. Recent estimates show that growth
of the manufacturing industry has started to accelerate, growing by 13 per cent
between 2011 and 2014. This rate of growth is, however, still insufficient to generate
the necessary shifts in the structure of the economy, in view of the considerable
shares of the agricultural and service sectors (see Figure 3.2).
Second, the share of manufactured goods in Ethiopia’s export earnings in recent years
has been very low (below 10 per cent), and was based on low technology goods such
as leather and leather goods, textiles, and other agro-industry products (see Table 3.1).
The export earnings share (p.68)
Table 3.1. Share of manufactured goods in Ethiopia’s export earnings (in $ millions)
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Export 37 39 84 67 82 75 78 110 109 136 98 89 176 212

earnings (in $
millions)

Share of 8 8 19 14 15 12 11 11 10 9 7 4 6 7
manufactured
goods (%)

Growth rate – 7 116 (20) 22 (9) 5 41 (1) 24 (28) (9) 97 20

(%)

Note: Rounded to nearest single digit

Source: Unpublished data, ERCA (2012a)
(p.69) of textiles, leather and leather goods, and agroindustry was 51, 40, and 8 per
cent respectively in 2011 (ERCA 2012a). This highlights the profound structural
problems facing the economy. In SSA countries, the share of manufactured exports in
export earnings has decreased in recent years (UNCTAD-UNIDO 2011).
Third, Ethiopia’s manufacturing value added (MVA) per capita was only $9 in 2010,
in contrast to Egypt’s $177 or Mauritius’s $522. In UNCTAD/UNIDO typology
(based on five groups, namely forerunners, achievers, catching up, falling behind, and
infant stage, that is, below $20 MVA per capita), Ethiopia is in the infant stage of
manufacturing. A more comprehensive measure of industrial competitiveness is the
Competitive Industrial Performance (CIP) Index (UNCTAD-UNIDO 2011;
Lall 1996, 1999). While Ethiopia is considered one of the lowest on the index, it has
shown some improvement in ranking in recent years, from 118th in 2005 to 111th in
2009.

3.2.2 Industrial structure

The industrial structure is dominated by smaller firms, and the average size of
manufacturing firms is even smaller (Tekeste 2014; Page and Söderbom 2012).
Medium and large firms are smaller than in other developing countries
(Söderbom 2011). Most firms are concentrated in Addis Ababa and its periphery, the
result of historical factors, available infrastructure, and market concentration. The
dominant manufacturing industry is the food industry, accounting for more than one-
third of the firms and employment in the sector. Insufficient industrial inputs for
manufacturing slowed the growth of domestic linkages, as is the case in most sectors.
According to the official 2007 Central Statistics Agency (CSA) survey, close to 40
per cent of small manufacturing enterprises are grain mills, accounting for 50 per cent
of employment; 20 per cent were furniture manufacturers; followed by 25 per cent for
metal fabrication plants. Most small manufacturers indicate insufficient capital as
their biggest problem. Numbers of medium and large firms increased from 1,243 in
2005 to 2,172 in 2011 (CSA 2006, 2012c). Employment also increased from 118,000
in 2005 to 175,000 workers in 2011, primarily in labour-intensive industries such as
food products, beverages, and textiles. The first two industries employ about a third
of workers in the manufacturing sector. Firms (p.70) identified market demand as a
primary challenge, with newly established enterprises listing infrastructure
(electricity, water, and production premises) as a major problem (CSA 2010).
More than 75 per cent of factories are concentrated in the Addis Ababa ring (with 40
per cent in Addis Ababa, 21 per cent in Oromia, 13 per cent in Southern Nations,
Nationalities, and Peoples’ Region [SNNPR]). Agglomeration economies provide
benefits to producers by offering bigger and closer markets, technological spillovers,
concentration of services and infrastructure, and consumers (Marshall 1920;
Henderson 1974, 2003; World Bank 2008). Thus the spatial logic of industrialization
appears to run counter to the dispersion of economic opportunity intended by the
federal government.

3.2.3 Main industrial actors

Broadly, the key actors are in the private sector, which includes foreign-owned (FDI)
and domestically owned firms and firms owned by regional endowment funds. Their
relative roles differ from sector to sector. For instance, in banking, food, and leather
industries, domestically owned firms dominate the home market, while foreign-
owned firms dominate in beverages and floriculture. In some industries, such as
cement or construction, both foreign- and domestically owned firms have a
significant presence. For industries that require lumpy investment (such as sugar,
cement, infrastructure, etc.), government continues to be the major actor.
The first investment law, enacted in 1992, allowed the private sector to be the main
industrial actor. Government also charted a privatization programme to encourage the
sector’s development. The domestic private sector quickly capitalized on the
promised benefits, particularly in services, and to some degree in manufacturing and
agriculture. By 2009, the share of the private sector in GDP at constant prices was
estimated at 90.1 per cent. This sector, of course, includes smallholder agriculture and
non-agricultural micro and small businesses. Formal private sector corporations
accounted for about 27.3 per cent of GDP in 2009. However, the informal sector is
still believed to account for the bulk of private sector activity (Kolli 2010).
Endowment fund–owned enterprises were founded in the mid-1990s, with initial
resources provided by the liberation movement for the rehabilitation and
development of war-torn regions. These major entities (p.71) are Endowment for the
Rehabilitation of Tigray (EFFORT) and Endowment Fund for the Rehabilitation of
Amahara (TIRET) (Vaughan and Gebremichael 2011; Kelsall 2013). Endowment
fund–owned conglomerates were established in accordance with Articles 483–515 of
the Civil Code of Ethiopia, No. 165 of 1960 (EFFORT 1995). According to Ethiopian
law, ‘an act of endowment is an act whereby a person destines certain property
irrevocably and perpetually to a specific object of general interest’ (GOE 1960). These
endowment investment groups (EIGs) have primarily invested in large-scale
manufacturing sectors, such as cement, textiles, tanneries, breweries, and a malt
factory; pharmaceuticals, marble processing, and agro-industries
(EFFORT 2010, 2011; Kelsall 2013). For instance, EFFORT founded the second largest
cement factory in Ethiopia in 2000. In total, EIGs employ more than 25,000 people in
more than twenty enterprises, and their capital outlay is estimated at about $1 billion.
The role of the EIGs was highly controversial during the 1990s and early 2000s, and
remains a divisive political issue. The major criticism of EIGs is by IFIs and the
international community based on the perception of them as ‘parastatals’ and the
questioning of their party political affiliations (Altenburg 2010, 2013). These critics
argue that the entities ‘crowd out’ private sector opportunities, a view also shared by
prominent members of the local chamber of commerce (Altenburg 2010; Vaughan and
Gebremichael 2011; Hagmann and Abbink 2012, 2013; Kelsall 2013). An alternative
explanation emphasizes the large-scale manufacturing nature of the EIG investments
in sectors in which the private sector has shown little interest because of the risks and
uncertainties. There is little evidence to confirm that EIGs are party affiliated, in the
sense of receiving party donations or special favours. The charge that loan terms for
EIGs (from state-owned banks) are uniquely favourable and not subject to the same
competition policies as other businesses (Vaughan and Gebremichael 2011) is also not
supported by the evidence. In fact, EIGs have fostered industrial development while
facing region-specific constraints, such as proximity to the volatile Eritrean border,
relatively poor infrastructure, and remoteness from the Addis Ababa market. Thus,
EIGs have become major players in Ethiopia’s industrial policy and economic
transformation (Kelsall 2013). This suggests that EIGs have pioneered long-term
value creation, filled gaps where the private sector had been absent, narrowed
regional imbalances, and generated employment. Some linkages have (p.72) been
activated by EIG-owned cement, brewery, and agro-processing projects.
Government policy has focused on public investment in a few strategic areas, and
government has privatized about 300 enterprises. The central focus of the remaining
SOEs is in three broad areas: financial (four corporations in banking and insurance
services), utilities and infrastructure (six corporations in electric power;
telecommunications; road, air, and sea transport), and large-scale manufacturing
investments (three corporations in chemical and cement, metal and industrial
engineering, and sugar). Since 2010, the government has established larger
conglomerates by merging existing public enterprises and establishing new ones in
selected sectors.
In sum, Ethiopia’s pattern of industrialization has significant vulnerability, as the
export and sectoral composition is not shifting in favour of manufacturing. The major
industries are labour-intensive, manufacturing low-value products. Industrial policy
needs to move into higher gear, as does structural transformation. Thus, further
industrialization is central to the country’s path of development.

3.3 Aspiring developmental state

Drawing on discussion in the previous chapter, it is possible to summarize the main
characteristics of a developmental state as the exclusive pursuit of development
(vision and practice); public mobilization around a grand vision; and state capability,
embeddedness, and autonomy. The developmental nature of the Ethiopian
government is discussed below.

3.3.1 Evolution of political process and tenets

Political and Ideological Tenets

The historical roots and political ideals of the ruling party, in power for two decades
under a dominant-party multiparty system, have shaped the policies behind the new
economic dynamism. EPRDF broadly defines its ideological tenets as ‘revolutionary
democracy’ (EPRDF 2011b; Balema 2014). Its thoughts and strategies, articulated
as Abyotawi Democracy (‘Revolutionary Democracy’), are expounded in speeches,
publications, debates, and other communication forums. One sceptical review
condemns (p.73) this cohesive ideology as ‘vanguard party rule’, and emphasizes that
EPRDF’s ‘condemnation of neo-liberalism is rooted in TPLF Marxist-Leninist
origins’. The critique adds that ‘the ruling party refused to relinquish command over
the economy’ and promotes ‘a rhetoric that emphasizes stability as a precondition for
development’ (Hagmann and Abbink 2012: 586). The undermining of governance and
multiparty politics has been a concern for some scholars, the political opposition, and
donors. Some observers argue that this has resulted in a ‘monolithic party-state
system’ dominated by the ruling EPRDF (Clapham 2009; Lefort 2013). Nonetheless,
there has been a working relationship with donors and IFIs. Clapham (2009) states:
‘Internationally, Ethiopia has had considerable success, presenting itself as a model of
“good governance” with donor approval. Having accepted the basic tenets of
neoliberalism, it also backed the “global war on terror”, giving it scope to promote its
own agenda in Somalia, with US backing. Its cardinal problem remains the
management of diversity and opposition.’ There seems thus to be disagreement
among external observers about whether, for instance, the ruling coalition is really in
favour of ‘neoliberalism’, but what is clearer is that, for good or ill, the coalition does
not neatly meet the typical criteria for ‘good governance’.

Wartime Intellectual Formation

The developmental orientation of the Ethiopian government has been shaped by
many factors, in particular the forging of a wartime coalition when the liberation
movement was led by the EPRDF. The liberation movement’s origins can be traced
back to strongly leftist student activists of the early 1970s. EPRDF emerged as the
dominant political and military movement between the mid-1970s and early 1990s.
Land reform and the national question were among the main political issues
(EPRDF 2011b; de Waal 2012). The values and commitments stemming from the
historical experience and intellectual formation of the leadership, in addition to the
choice of federalism as a mechanism to manage long-lasting political challenges in
Ethiopia, have been key. Furthermore, the desire to maintain independence and the
freedom to make mistakes has been a strong theme among the leadership.
The ruling party and government have exhibited a long-term commitment to
sustained, rapid, and equitable growth (EPRDF 2011a; 2013a; 2013c; 2013d). Threats of
endemic famine and poverty and the risk of (p.74) Ethiopia’s disintegration under
internal and external pressure have all contributed to the commitment to this vision.
Increasing political pressure in urban centres has been an important factor for the
government’s focus on urban economic and governance issues, especially after the
2005 national election. This election and its aftermath were an important wake-up
call, perhaps even an internal threat, to the ruling party. The focus on large housing
and infrastructure developments and employment programmes in urban centres
commenced after, or partly because of, these events.
Considering the widespread and profound poverty in Ethiopia and the country’s long
history of political fragility and ethnic diversity, the issue of equitable growth has
been especially pressing. Regional equity has become the foundation of the polity,
institutionalized through a commitment to federalism. According to Clapham (2009),
the management of diversity and the difficulty of reconciling ‘autonomous systems of
power and authority within a common political structure’ remain a central challenge
for the government (Markakis 2011).
Arguably, this commitment to ‘horizontal’ equality (Stewart 2002, 2008) may have
offered at least a rhetorical counterweight to the clearly accelerating vertical
inequality accompanying and encouraged by growth. This is an interesting contrast
with Amsden’s view that promotion of large concentrated champions was easier in
more equal countries like Korea than in, say, South America. In Korea, a
homogeneous national identity appears to be a positive factor. In Ethiopia, promoting
small businesses and supporting smallholder farms has been preferred to the
dominance of a few large firms. This is justified on grounds of increased employment
generation and fairer wealth distribution. The focus on agriculture was also warranted
by the need to alleviate poverty in the countryside, where more than 80 per cent of
people live, and to use agriculture for the initial take off of industrialization.

3.3.2 Strong developmental orientation in vision and practice

An Exclusive Pursuit of Developmental Goals

There is strong evidence of developmental state behaviour in Ethiopia, as expressed
in the exclusive pursuit of developmental goals. Firstly, the record demonstrates a
political commitment to a grand vision. The Ethiopian state has fostered a national
project, the ‘Ethiopian Renaissance’, (p.75) which lies at the core of public policies.
As already noted, the government’s medium-term vision is to become a middle-
income economy by 2020–25. These public commitments and policy statements may
be nothing more than that, and many governments make rhetorical commitments that
do not correspond to reality. What is more important is whether the rhetoric is
matched by the behaviour of developmental states and translated into meaningful
interventions in investment, resource allocation, policy implementation, and
outcomes such as rates of growth of key indicators and structural change. The
government has been directly and aggressively investing in infrastructure
development and human resource development towards this end. For instance, its
interventions to increase electricity generation capacity, expand railways and roads
(over 90,000 kilometres of roads, including universal access to rural roads), and
create capacity to train half a million university students amply demonstrate its
developmental orientation (MOFED 2010).
Ethiopia has sustained double-digit economic growth, in particular between 2003 and
2013, without relying on a resource boom (such as oil or minerals).4 It is recognized
as having one of the ten fastest growing non-oil economies. The government has
reduced the country’s famine vulnerability and improved its capacity to feed its
growing population (UN-DESA 2007; CSA 2013). The number of people living below
the poverty line has declined by half. Enrolment in primary schools has increased
from below 20 per cent in the early 1990s to about 95.3 per cent in 2013. In 2012, the
total student population in all levels exceeds 30 million.
The government has emphasized the essential role of an activist state in the process
of catching up, a role further necessitated by the strong determination and vision to
develop Ethiopia. This developmental orientation is home-grown and based on
specific conditions in Ethiopia, although emulating forerunners has also played a role
(for instance, Germany’s technical and vocational education and training [TVET] and
university system, Japan’s Kaizen production system, China’s industrial parks).
Ethiopia’s rich history of independence and civilization and its mimetic interest in
finding East Asian role models have served as sources of inspiration.

(p.76) 3.3.4 Building from scratch: State capability and embedded autonomy

State Capability
Another developmental characteristic is the somewhat imprecise idea of the
capability of the state. The bureaucracy from the dictatorial regime of the Derg was
politically hostile to the new EPRDF-led government, having been used as an
instrument for war mobilization, and having served both the command economy and
predatory ends. The state machinery at the time was ill-suited to serving the new
political leadership’s developmental goals. The bureaucracy of the newly established
federal state had to be rebuilt. This implied massive bureaucratic transformation in
terms of political indoctrination (with the philosophy of the new government) and
instilling professional capabilities. The civil service reform programme was thus
initiated in the late 1990s to this end; but it was not selective in its focus and did not
succeed in generating the required transformation. Arguably, the party’s
organizational capability was strong and may have compensated for the perceived
deficiency of the bureaucracy in the short term.

Inherited Capacity of Private Agents

Although the old private sector barely survived under the Derg, some of its
participants managed to seize the opportunity to invest in many sectors after the
introduction of the market economic system in 1991. Nonetheless, the low level of
infrastructure and human development undermined the profitability of productive
investment and instead encouraged unproductive and rent-seeking activities among
all actors, including public officials and the private sector. In this, both push and pull
factors interplay. On the one hand, some of the businesses such as trading have been
speculative, earning super profits and evading tax; while this has not been the case in
some productive activities such as manufacturing. On the other hand, the private
sector faces differing and unequal constraints across sectors. And given existing
constraints, incentives push businesses and entrepreneurs to invest more in short-term
activities: they are not ‘compelled’ by incentive structures and institutions to do
things differently. It is not that they are inherently good or bad, or weak or strong.
What is needed is a policy setting that incentivizes and maximizes the dynamism of
the private sector in Ethiopia.

(p.77) Embedded Autonomy

Another developmental characteristic is the embedded autonomy of the state.
Embeddedness comes with acceptance and, hence, legitimacy. Autonomy is also
about avoiding capture by specific interests and maintaining policy independence.
Ocampo et al. (2009: 155) suggest that the ‘the nature of the partnership [between
state and private sector] will vary from country to country, depending on the
characteristics of both the private sector agents and the state’ and the purpose should
be ‘mutual learning’. The government and the political party leadership in Ethiopia
have been careful to maintain autonomy and distance from the private sector,
although this has been uneven among sectors. Such distance obviously limits mutual
learning and achieving in an environment of high rent seeking by both civil servants
and the private sector, and requires political commitment and skillful management.
The distance here may have hindered the institutionalized interaction between state
and productive private sector identified by Buur et al. (2012) as critical to productive
expansion, though this has changed in some Ethiopian sectors, as later chapters show.
During the liberation struggle and post-1991 period, the rural population continued to
offer political support, guaranteeing the legitimacy of the ruling party. However, as
the economy diversifies and new and powerful social actors with a vested interest in
the economy emerge, state autonomy becomes ever more critical. In a country where
rent seeking is widespread, the state and the private sector are disposed to be locked
into a state of mutual suspicion and mistrust. This lack of trust has been an obstacle to
a durable state–business relationship. In this context, maintaining the balance
between state autonomy and developing trust remains challenging. Structural factors
such as intense foreign exchange constraints have combined with political history to
add urgency—and hostility—to these fraught relationships in some cases. In one
respect, this also shows the inability of government policy to shape private sector
behaviour (though later chapters show how state–business relationships have
sometimes evolved in more productive ways).

Policy Independence
In addition, the government withstood pressures from IFIs and their shareholders and
has kept important parts of the economy under state ownership. This includes the
power company (EEPCO), telecom (Ethio (p.78) Telecom), the railway company, and
banks. These SOEs were used to advance broader developmental objectives and
industrial policies. The political wrangling between the Ethiopian government and the
IFIs is documented by Stiglitz (2002: 32),
When I arrived in 1997, Meles was engaged in a heated dispute with the
IMF, and the Fund had suspended its lending program…Ethiopia resisted
the IMF’s demand that it “open” its banking system, for good reason. The
Ethiopian banking system was seemingly quite efficient…The IMF was
unhappy, simply because it believed interest rates should be freely
determined by international market forces, whether those markets were or
were not competitive.

This is an example of how pressures combined with ideas to generate particular ways
of addressing difficult challenges. Liberalization of the banking sector, it has been
argued, might have detrimental political consequences, considering the low levels of
capacity of domestic banks and the government’s insufficient regulatory capacity.
The government’s approach to privatization also differed from what IFIs prescribed.
The above developmental state construct also influenced the government’s attitude
towards FDI. Government FDI policy has entailed an open policy in most sectors,
with the exception of banking and certain businesses (e.g., security).
One interesting aspect has been the growth of pressure from foreign experts. This is
not always in the form of heavy-handed IMF or World Bank demands or insistent
pressure from bilateral donors and governments. Sometimes, it is far more subtle and
in the guise of supportive offers of help and advice, arguably designed to adjust the
parameters of policy design and to introduce new or rival ‘rhetorical common places’
into discourse and thinking among Ethiopian policymakers.

3.3.5 Foundations of policymaking in Ethiopia: Strategy, institutions, practices

In broad terms, industrial development in Ethiopia between 1992 and 2013 has been
shaped by major political changes, shifts in economic policies, and the political
economy factors outlined in Chapter 1. The organizational structure through which
policy is designed, implemented, adapted, stalled, and clogged was moulded within
and is being changed by this broader context. This and subsequent sections of this
chapter map (p.79) the economic interests and principal agents, the policymaking
framework, policy instruments, and institutions shaping industrial policy in Ethiopia.
No such comprehensive overview has existed prior to this research, to the best of the
author’s knowledge. Moreover, this mapping exercise minimizes the need for
repetition in the three case studies in Chapters 4–6.
The mapping of industrial policy instruments and institutions may also further
elucidate the circumstances by identifying patterns or striking features, especially
those which have a bearing on the research. As the narrative and analysis in this and
subsequent chapters show, anti-fragility was the dominant characteristic: there were
successes, failures, and remarkable changes and adaptation under threat. Anti-
fragility applies to phenomena that benefit from stressors, volatility, and threats, and
that gain strength and reduce fragility (Taleb 2012). Industrial policy instruments have
evolved from their rudimentary beginnings into more coherent and complementary
toolkits, and from the generic to the more targeted. On the other hand, institutional
mismatches (rules failing to regulate in the ways intended) and persistent institutional
constraints have been significant and have undermined the effectiveness of policy.
Chapters 4–6 develop and compare these issues through specific sectoral examples.

Industrial Development Strategy of Ethiopia

Within the broader ADLI, the Industrial Development Strategy of Ethiopia (IDSE) is
the basis for the country’s industrial policies. ADLI became the country’s
development strategy in 1994. IDSE has been the ruling party’s guiding document
since 2000, although it became an official document only in August 2002. The
industrial development strategy aims at promoting industrial development that is
export-oriented, agriculture-led, and focused on employment generation through
labour-intensive industries. It emphasizes the need for all-round support to
industrialists, and promoting an environment conducive to private sector expansion
through macroeconomic stability, infrastructure provision, and access to loans. Also
highlighted is human resource development, improving the regulatory environment
and justice system, and combating a rent-seeking political economy. The strategy
gives due attention to the development of medium and large-scale manufacturing
firms, particularly in priority industries such as garments and textiles, agro-
processing, meat processing, leather and leather products, and construction.
(p.80) Central to the strategy is the development of small enterprises. The
commitment to micro and small enterprises (MSEs) is the political and ideological
corollary of commitment to small farmers. But in adopting this approach, government
is setting itself up for the huge challenge of coordinating and leading large numbers
of dispersed investors, producers, and traders. Indeed, the government has found it
immensely difficult to deliver on commitments to numerous, widely scattered small
farmers and businesses. This challenge is further compounded by resource and
institutional constraints, as well as the low level of industrialization and insufficient
industrial experience.
To conclude, the key features are that the IDSE policy document was originally based
on the ruling party’s white paper, and it is one of the main documents used for
indoctrination and training within the party, the civil service, MSEs, technical-
vocational institutes, and universities. Second, the document has not been supported
by research and updated in light of changing circumstances and perspectives. Third, it
has not been fully supported by any sector- or industry-based comprehensive strategy.

Sector Development Programmes and Five-Year Development Plans

Planning is considered one of the tools of industrial policy and it is used with
different levels of emphasis. In Ethiopia, the industrial development strategy was
translated into medium-term five-year development plans, including the Sustainable
Development and Poverty Reduction Program (SDPRP, 2002–05), a Plan for
Accelerated and Sustained Development to End Poverty (PASDEP, 2005–10), and
GTP (2010–15). The distinctly bold GTP reflects, first, the government’s ambition to
pursue rapid industrialization and structural economic transformation. Its target for
the annual growth rate in manufacturing was set at 20 per cent, while total export
earnings were to increase fivefold. These targets are extremely high compared to
experience elsewhere. Second, in terms of sectoral transformation, the share of
industry in GDP was to increase from 12.9 per cent in 2009–10 to 15.6 per cent in
2014–15. GTP also called for a related shift in infrastructure development. For
instance, electricity generation was to increase from 2,000 to 10,000 MW, and a new
railway network was to be developed. Third, the priority industries were expanded to
include textiles and garments, leather and leather products, sugar and sugar-related
products, (p.81) cement, metal and engineering, chemicals, pharmaceuticals, and
agro-processing (MOFED 2010; KOICA 2013).
Ambitious plans can arise from positive intentions to force the pace of change (if the
plans are shared and reinforced by incentives, sanctions, and accountability systems),
or from lack of data and poor analysis. Where key agents and interest groups are not
involved in the process, it can become especially easy to blame them for failures to
achieve targets. Where there are repeated setbacks in achieving targets, then the
tendency to pay lip service to plans will become more apparent. The planning process
should clearly define boundaries and content, as there are different interpretations of
plans. Almost all countries use planning. What differs is the purposes and
instruments. Central planning has been used in command and socialist economies,
while indicative development plans have been extensively used in Korea, Taiwan,
Japan, and France, among others. Many institutional and historical factors influence
the boundaries and content of specific indicative plans. The evidence from Ethiopia
suggests that its planning process has been subject to precisely these challenges. For
instance, SDPRP, PASDEP, and GTP are not uniform in content or approach
(MOFED 2002, 2006, 2010). Moreover, there have been repeated cases of agencies
blaming others, often sidetracking committee meetings in the process. Evidence of
lack of reliable data and analyses include, for instance, exaggerated targets for the
cement industry and low targets for alluvial gold mining. The absence of sufficient
data and sectoral understanding makes such digressions unavoidable. In short, there
can be no effective industrial policy without state capacity to gather reliable data
across economic sectors and to deploy the requisite analytical capacity to translate
evidence into concrete policies and to closely monitor outcomes.

Policymaking Hierarchy and Constitutional Framework

The policymaking framework emanates from the Ethiopian Constitution (which was
endorsed on 8 December 1994, and came into force on 21 August 1995). The new
Constitution stipulates that the federal government and states have legislative,
executive, and judicial powers (FDRE 1995). The highest federal authority, the House
of Peoples Representatives, ratifies proclamations, basic policies, and budgets, while
the Council of Ministers approves regulations, policies, and executive
directives, (p.82)

Figure 3.3. Policymaking hierarchies in Ethiopia

which are acted upon by ministries (see Figure 3.3). The Constitution also stipulates
the need for public participation in policymaking processes. Regional governments
(states) and local administrations play important roles in implementing industrial
policies by, for instance, providing land to businesses and promoting MSEs
(FDRE 2012). This entails increased institutional complexity and calls for more
effective coordination among government hierarchies in order to implement an
effective industrial policy. By contrast, Korea, Taiwan, and China have centralized
unitary systems, which have helped them in pursuing industrial policy. The challenge
for the Ethiopian government is to ensure that the policies and goals set at the federal
level are implemented with the same conviction locally and regionally.

3.4 Industrial financing: Policies and instruments

3.4.1 Ethiopian banking policy and constraints

Despite pressure from the IMF and World Bank on Ethiopia to liberalize its financial
sector (see BBC, ‘Ethiopia hits out at IMF’, 1 September 2003), as well as diplomatic
pressure from individual countries, foreign banks are not allowed to operate in
Ethiopia. The government intends to continue with this policy until domestic banks
achieve the required financial, (p.83)managerial, and technological capacity to
compete against international banks, and until institutions are developed to regulate
giant foreign financial institutions. ‘Intrinsic financial fragility’ and vulnerability, and
high dependence on foreign capital expose LDCs to external shocks (UNCTAD 2011).
Palma (2003) argues emphatically that ‘developing countries should avoid opening up
their capital market’ in order to avoid crises. Grabel (2003) notes the need for
financial architecture that harnesses economic development and equity, and warns
against uncontrolled surges of international private capital. UNCTAD (2011) also
advocates a regional developmental framework in which regional and sub-regional
development banks play a positive role (UNCTAD 2011; Grabel 2003; dos
Santos 2011).
Ethiopian policy also emphasizes the key role of state-owned banks, in particular
policy banks, in supporting the government’s industrial policies. The regulatory
arrangements, on the other hand, have two dimensions. The National Bank of
Ethiopia (NBE), the central bank of the country, regulates the financial sector as a
whole. In addition, the Public Financial Enterprises Agency (PFEA) was established
to supervise state-owned banks and financial institutions, including the Commercial
Bank of Ethiopia (CBE) and the Development Bank of Ethiopia (DBE).
Because of this policy framework, the main actors in Ethiopian banking are the
government and the domestic private sector. State-owned banks continue to dominate
the Ethiopian banking industry, with 40 per cent coverage of branches and 60 per cent
of the capital base. While CBE spearheads trade support (working capital and
international banking services), DBE provides long-term loans to priority sectors at
subsidized rates (CBE 2011, 2012). CBE was established in 1970, and had 209
branches in 2012, while DBE has thirty-two. State-owned banks combined had 273
branches and private banks 408 branches. Another state-owned bank, the
Construction and Business Bank (CBB) established in 1983 for mortgage financing,
continues to exist but has little influence in the sector. The policies pursued by the
previous regime meant there was not a single private bank until 1991. Following the
opening up of the economy in the 1990s, however, there has been rapid growth of
private banks. In 2012, there were sixteen privately owned domestic banks. In
addition, there are micro-finance institutions that focus on supporting smallholder
farmers and urban MSEs. Lending rates have remained relatively stable throughout
the period (between 7.5 per cent and 14 per cent).
(p.84) Nonetheless, the financial industry in Ethiopia remains underdeveloped
(EEA 2013) in terms of capital, savings, and institutional capacity. Specifically, the
constraints and weaknesses of the banking sector are: capital constraints due to low
domestic saving; the banks’ tendency to focus on short-term lending to the service
sector rather than long-term financing of, in particular, manufacturing; the limited
coverage of banking services; weak institutional capacity; and slow pace of
modernization (NBE 2011; MOFED 2010). Recent policies to increase the level of
savings and investment include expansion of banking outlets and improving
coverage, sale of bonds (in particular, in relation to the Renaissance Dam), savings
for housing programme, and increasing saving rates (MOFED 2013a).

3.4.2 DBE as prime policy bank

DBE was established in 1908 and was the only development bank that financed long-
term industrial and agricultural projects. The government aims to use DBE as a
policy bank to advance industrial policies, as indicated in its industrial development
strategy. DBE’s policy directive specifies that the bank should ‘provide medium and
long-term loans for investment projects, which are engaged in agriculture, agro-
processing and manufacturing industries, preferably export focused’ (DBE 2012a).
The two criteria for funding eligibility are ‘diligence or KYC (know your customer)
assessment to identify the integrity of the borrower’ and appraisal of ‘the feasibility
study submitted by the applicant’. DBE’s interest rate is lower than that of
commercial banks and market interest rates, entailing a subsidy of between 2.5 and 5
per cent. The bank requires a minimum equity contribution of 30 per cent. Loan
processing times have decreased over time, although not to the satisfaction of clients.
Sole responsibility for deciding on loans rests with the bank’s president, while the
board of director’s responsibilities are confined to policy decisions and to overseeing
bank performance. Before 2005, the board was involved in loan approval, which
resulted in conflicts of interests and lax accountability. The Ministry of Finance and
Economic Development (MOFED) and Ministry of Industry (MOI) have indirect
influence in prioritizing loans.
Since 2005, DBE loan policies, procedures, and terms have been brought into line
with government priorities. Reforming the bank took longer (p.85) than expected and
was not completed until very recently. In 2012, two-thirds of DBE loans were
channelled to the manufacturing sector, while 23 per cent went to agriculture. The
major beneficiaries in manufacturing were textiles (30 per cent), and non-metallic
industries, primarily cement (20 per cent). Industrial crops and floriculture were the
key loan beneficiaries in agriculture. In terms of concentration, 83 per cent of loans
were disbursed to fifty industrial projects in textiles, cement, and sugar (DBE 2012b).
Thus, DBE is serving the government’s industrial policy by promoting industrial and
agricultural development.
Despite the recent operational changes, DBE has been constrained in supporting rapid
industrialization in a number of ways. First, DBE’s role is restricted by its limited
capital base and by its ‘single borrower limit’. Second, bureaucratic inefficiencies and
rigid application of standards have undermined DBE’s role in effectively supporting
industrial policy. This has arisen from the difficulty of balancing transparency and
efficiency. Many firms in the leather, floriculture, and cement sectors complained
about delays in processing loans and the lack of efficiency and flexibility in
procedures. By contrast, the government and DBE’s board, wishing to ensure
accountability and transparency, strictly follow procedure. Third, DBE faces
information asymmetry problems, that is, insufficient knowledge of industries and
difficulty in approving borrowers based on their past record. Finally, the bank’s
limited institutional and human resource capacity has also undermined its
effectiveness.

3.4.3 CBE’s active role in industrial financing

CBE has played an indispensable role in promoting exports and industrial
development. Experience elsewhere shows that commercial banks may play a partial
role in investment financing, either through specialized subsidiaries or co-financing
(Aghion 1999; Schwartz 2010; Diamond 1957). Since CBE’s capital base was much
bigger than DBE’s, it co-financed large-scale industrial projects such as sugar and
cement projects. Until 2005, the bank’s loans were not in line with government
policy, and between 2000 and 2005 it accumulated bad loans because of laxity in loan
terms. The result in 2003 was that 52 per cent of loans were non-performing.
Recently, thanks to more effective management, this percentage has decreased
significantly, and the bank has made changes to its portfolio in line with government
policy. This involved shifting from (p.86) the profitable import and domestic trade
towards financing exports, agriculture, and manufacturing. Consequently, CBE’s
lending to manufacturing and agriculture increased to 25 and 23 per cent respectively
after 2002. Its lending to domestic trade declined to 10 per cent, while the import–
export trade accounted for 30 per cent of lending in 2011. Its share of export
financing increased from mere 17 per cent in 2008 to more than 80 per cent in 2012,
reaching $1 billion. While CBE has furnished working capital to manufacturers, as
well as providing an international banking service, in practice the bank discriminates
against manufacturing, as it does not recognize machinery as collateral, but favours
buildings instead.
CBE has also provided long-term loans of more than ETB 15 billion for the
government-sponsored housing programme, and more than ETB 3 billion for capacity
building in the construction industry (MUDC 2013). Coordination between CBE and
DBE has been strengthened, thereby apparently augmenting the implementation of
industrial policy. It seems that CBE has, in part, played the role of a development
bank. According to its president, a major constraint for the bank is inadequate
institutional capacity in terms of sectoral knowledge and experience.

3.4.4 Supplementary financing instruments

Resource mobilization through domestic savings and domestic revenue mobilization,
and channelling these resources into productive investment are also features of
developmental states. The government has, in addition, used other industrial
financing instruments. It established an Industrial Development Fund (IDF), intended
to finance expansion projects by public enterprises. This has allowed government to
mobilize part of the profits of SOEs and use them to finance very high priority
projects. Allocation decisions are made by MOFED and MOI based on the country’s
five-year plan. Public enterprises can retain a maximum of only 15 per cent of
corporate income and can reinvest in productivity, technology, and governance
capabilities. This funding scheme has been an innovative and important trade-off,
allowing investments in priority areas upon approval by the respective boards, and
improving enterprise performance.
The government has also permitted the channelling of external loans for investments
by public enterprises. This includes loans for expansion projects by Mugher Cement,
Sugar Development Corporation, and (p.87) Ethiopian Airlines (EAL). In addition,
government has allowed promoters of large-scale, mainly FDI projects, to access
external loans from international funding agencies. For example, Derba Cement
received loans from the African Development Bank (AfDB), European Investment
Bank (EIB), and International Finance Corporation (IFC). More recently, the
government made compulsory the buying of a 27 per cent central bank bill (the NBE
Bill) by private banks, which is then used by DBE for long-term lending and
investment in priority sectors. This has been a contentious issue for private banks,
NBE, and IFIs, on the grounds that the arrangement favours public projects. There is
an element of truth in this, as loans to the private sector diminished after 2011 due to
the strain of financing new infrastructure projects. These initiatives indicate the gap
between long-term investment requirements and the government’s quest for
alternative financing without opening up the banking sector to foreign banks. This
experience of policymaking thus shows not only successes, but also fragility and
adaptation under threat.
In conclusion, as discussion in the following chapters will highlight, the following
findings can be underscored. First, a key feature of industrial financing has been the
banks’ lack of knowledge of the industries they finance, the slow and bureaucratic
process of financing, deficiencies in prioritizing and selecting prospective industries,
and low institutional capacity. Monitoring effectiveness differs by sector, and loans
were less linked to economic performance. The banks were mainly concerned with
timely loan repayment, rather than driving long-term performance. Second,
government was reluctant to privatize banks and to admit foreign banks, a major bone
of contention with IFIs, bilateral agencies, and other countries. Third, government did
not use the banks effectively as policy banks, at least initially, and loans were not
targeted at industrialization. Fourth, this improved after 2004–05, and all government
banks were reoriented to follow government policies and priorities. DBE and CBE
have now effectively become policy banks and been instrumental in providing long-
term working capital to some key sectors. Fifth, the government even sought to
influence the behaviour of private banks (focused on trade and short-term loans) and
used indirect instruments to channel resources for financing long-term priority
investments. Lastly, the gap between credit supply and demand in priority industries
remains huge, indicating that banks need to give more attention to deposit
mobilization.

(p.88) 3.5 Investment and export promotion

3.5.1 Investment promotion instruments and organs

Foreign investment was welcomed in all sectors, with the exception of
telecommunications and finance. Since 2004, the Ethiopian government has targeted
Turkey, India, and China as sources of FDI in manufacturing. Combined with
increased labour cost in these countries, this targeted investment promotion has had
positive results. The country’s improved economic performance, expanding domestic
market, and cheap labour were key attractions. In addition, the Netherlands
government has provided incentives to floriculture firms that invest in Ethiopia, while
China has supported the establishment of an industrial park (for instance, through the
China-Africa Development Fund) as part of ‘China Goes Global Policy’ (World
Bank 2012a; Brautigam 2011). The building of new industrial parks (for example, the
Eastern Industrial Zone in Dukem, the first in Ethiopia) has attracted more
investments from China (MOI 2012c). The share of Chinese investment in
manufacturing, mostly by private sector investors, amounted to 83 per cent of
registered FDI certificates (FIA 2012a). Most of the Chinese SOEs are involved in
construction works (hydropower projects, railway and road construction) and
providing equipment for private and government projects (including cement
manufacturers and the Ethio-Telecom SOE). China has also become a partner in
financing infrastructure and in project execution (Brautigam 2011). Exim banks in
China, as well as Turkey, India, and Egypt have provided support. Between 2004 and
2010, FDI flow to Ethiopia was 2 per cent of the country’s GDP, much lower than in
most Asian countries. The equivalent flow in Vietnam was 5.7 per cent (2000–10); in
China, 3.9 per cent (1991–2010); and in South East Asia, 4.5 per cent (World
Bank 2012a). However, the volume of investment in Ethiopia has grown rapidly since
2005. In floriculture, FDI comes mainly from Europe, Israel, and India. In breweries,
West European companies are visible, while in the cement industry, textiles, and
leather and leather products, Chinese, Indian, and Turkish companies dominate
(NECC 2012).
Guiding FDI towards industrial exports, the development of technological and
marketing capabilities, and increasing local content continues to be a key challenge.
Experience in Asia and elsewhere shows that FDI does not necessarily bring net
positive effects, and strengthening the indigenous private sector is vital from a long-
term perspective (Amsden 2009).

(p.89) Investment Incentives

To create an incentive structure, incentives must be designed, granted, and
implemented, and there must be follow-up on compliance (UNCTAD 2000: 23).
Incentives that included exemption from customs duty on capital goods and related
spare parts, and a two to five-year tax holiday on profits were provided. The latter
may be extended for two years for investments in remote regions and for exporters
who export more than 50 per cent of their output. Investors are also entitled to ‘carry
forward losses’ and to use a preferred depreciation system, accelerated or straight-
line. The incentives were uniform in that they applied to most sectors, rather than just
some. The domestic private sector was quick to respond by investing in many areas,
including manufacturing. However, the incentives and institutional supports were
insufficient to direct the private sector into priority manufacturing areas, into value
addition, and into technologically advanced activities.

Genesis of Investment Administration

In view of the low levels of industrialization and private sector development, and the
intense competition worldwide to attract investment, designing appropriate policies
and incentives to attract foreign investment has become paramount. In 1992, the first
investment office was established to promote and facilitate domestic and foreign
investment. It was an autonomous body led by a board of investment and chaired by
the prime minister (FIA 2012b). The board had fifteen members, including the
Ministers of Finance, Trade, Industry, Agriculture, and Foreign Affairs, the NBE
Governor, Commissioner of Science and Technology, and the Secretary of the
Chamber of Commerce. The Minister of Planning and Economic Development (the
current MOFED) deputized as chairperson. In the same year, the proclamation was
amended and directed regional states on how to organize their own investment offices
under regional enactments. In 1996, the investment office was re-established as the
Ethiopian Investment Authority (EIA) with additional powers. It had a seven-member
board, again chaired by the prime minister. The notion of a ‘one-stop’ service was
also adopted, although it became evident that this concept was not fully understood,
so that significant adjustment at federal and regional levels was needed. Institutional
tensions among federal, regional, and local agencies have undermined the
effectiveness of the one-stop service, for instance, in the provision of land for
investment.
(p.90) In 2002, the EIA was rebaptized as the Federal Investment Commission and in
2006 it became the Federal Investment Agency (FIA). This was accountable to the
Ministry of Trade and Industry and was led by a new board of investment, comprising
government officials and private sector representatives and chaired by the Minister of
Trade and Industry. In reality, the private sector representatives were handpicked to
represent the industrial associations. With the restructuring of the cabinet in 2010, the
agency was made accountable to MOI.
A review of board of investment minutes (2008–12) shows that of a total of 446
decisions, 75 per cent related to duty-free privileges. Another 8 per cent related to
VAT and excise duty exemptions (FIA 2012c). Only seventeen decisions and
consultations (8 per cent) focused on amendments to various directives. This finding
indicates that, first, the highest political leadership (the board of investment) did not
focus on institutional constraints and development of the agency’s institutional
capacity. The repeated cosmetic changes to the institution serve as additional
evidence of this deficiency. Second, this finding also highlights the vague lines of
accountability between board and management team. The management executive
should have made the most of these decisions, but a lack of delegation hampered
efficient service delivery. Third, the finding also indicates that executive directives
lacked clarity, thereby creating transparency loopholes. The latest changes under a
new proclamation in 2013 specify expanded responsibility for the agency, including
‘one-stop service’ and ‘investment after-care’.

Constraints and Challenges

First, frequent revisions to and weak implementation of investment policies (for
instance, of one-stop service) have undermined the incentive effect, contributing to a
largely piecemeal approach and insufficient learning. Nevertheless, unlike the export-
trade duty schemes, which were not revised for a long period, these revisions do
imply some learning. Second, investment incentives were intended to encourage
investment in remote regions and to support equitable regional growth. Despite some
improvement in investment flows into major regions, investments continue to
concentrate on Addis Ababa and Central Oromia. The economies of agglomeration
appear to dictate the spatial distribution of economic activities. There is thus a
constant tension between equitable regional growth and the reality of economic
agglomeration, between interlocking (p.91) economic logics and the pressures of
politics. Investment promotion is thus in a major quandary (Schwartz 2010;
Henderson 2003; Marshall 1920).
Third, recent shifts in investment policy favour selected industries in the
manufacturing sector and the domestic private sector. Manufacturing was accorded
more generous incentives than other sectors, while incentives for established
industries (tanneries, cement etc.) were reduced (KOICA 2013). The development of
industrial parks was given priority in investment law. According to a proclamation
ratified in September 2012:
…the encouragement and expansion of investment, especially in
the manufacturing sector, has become necessary…to strengthen…domestic
production capacity;…further increase the inflow of capital and speed up
the transfer of technology into the country;…enhance and promote
the equitable distribution of investment among regions; put in place a
system of supervision to ensure that permits and incentives granted to
investors are used for the intended purposes;…. the system
of administration of investment needs to be transparent and efficient…the
establishment of industrial development zones helps, by creating enabling
and competitive conditions, to interrelate manufacturing sectors based on
value creation as well as to attract and expand investment.… [my
emphasis]

Because of lack of effective control, there were many instances of abuse of incentives
by investors. These included abuse of duty-free privileges, including selling duty-free
vehicles and goods for hotel and touring operations at market prices, and the transfer
and resale of land given under concessional terms (see various Ethiopian Revenue
and Customs Authority [ERCA] and FIA reports).
In summary, the investment agency has not been effective and is largely without
‘teeth’. It was reduced to compiling data on the number of investment licences and
capital, and providing investment certificates. The tendency was to measure
investment success by total registered capital rather than actual outcome.

3.5.2 Export promotion and trade protection

Development economists and economic historians have emphasized export-led
industrialization as a strategy for catching up (Studwell 2013; Wade [1990] 2004;
Amsden 1989; Thirlwall 2002; Ocampo et al. 2009). What matters is not openness per
se, but the manner of an economy’s insertion into international markets (Ocampo et
al. 2009). Export (p.92) promotion is vital not only to generating foreign exchange
and for balance of payments (see Chapter 2), but also for pushing the productivity and
competitiveness of a national economy. Ethiopian industrial policy documents
emphasize export-led industrialization, and refer to the success of the East Asian
experience (FDRE 2002).

Exchange Rate and Allocation Instruments

A stable exchange rate is part of a state’s macroeconomic policy for sustained
industrialization (FDRE 2002). Since 1992, the value of the ETB has been constantly
adjusted to narrow the disequilibrium. There is indeed an argument for aiming at an
undervalued currency to underpin export performance (Ocampo, Rada, and
Taylor 2009). Under the Derg’s command economy, exchange rates remained
unadjusted, resulting in an overvalued currency, with a powerful anti-export bias
(Tekeste 2014). It remained at ETB 2.07 against the dollar for almost seventeen years
(1975–91). Contrary to IFI prescriptions for full liberalization, the current
government adopted a managed floating system, which narrowed the gap between the
official and parallel market. A weekly auction of exchange rates is conducted among
local banks, and the central bank intervenes to influence rates and reduce volatility. In
2010, the local currency was devalued by about 20 per cent (Figure 3.4). The uneven
effects of the exchange rate on cement, floriculture, and leather and leather products
industries are discussed in Chapters 4–6.

Figure 3.4. ETB–US$ exchange rate (July monthly average), 1992–2012

Source: National Bank of Ethiopia, Research Department, 30 November 2012

(p.93) Foreign Currency Retention

The foreign currency retention scheme was introduced in 1996 by the National Bank
of Ethiopia (Tekeste 2014; NBE 2013). In terms of the directive, eligible exporters can
retain up to 30 per cent of foreign currency earnings, and exporters need to open two
foreign currency retention accounts for this purpose. The account holder can use 10
per cent with no time limit for various purposes, such as the import of goods and
export promotion. Moreover, the holder is free to sell the remaining 20 per cent
within one month. This system was extensively applied and not difficult to
administer. As the country’s foreign currency reserves were limited, there were
persistent shortages and the central bank made allocations based on government
priorities. This instrument was used in periods of critical shortages; and often,
prioritization suffered from subjectivity that caused dissatisfaction, and malpractice
in banks. The complaints were from international contractors (because of delayed
payment), investors (repatriation of profits), importers, exporters, and investors alike.

Export Promotion and Trade Duty Instruments

Multiple export promotion schemes, such as duty-drawback, voucher schemes, and
bonded warehouses, were used to reduce the pressure of increased working capital
requirements and to relieve bottlenecks in trade facilitation and logistics. These
schemes may be broadly classified into pre-export incentives (voucher systems,
bonded warehouse), and post-export incentives (duty-drawback). Under these
schemes, the export sector was exempted customs and indirect taxes such as VAT,
customs duty, and withholding taxes upon submission of the necessary documents.
The administration of the export promotion schemes is rife with problems. First, MOI
and other bodies have never developed reliable and timely input–output data, a basic
requirement for the system. The burden of preparing these data was transferred to
manufacturing firms, who had neither the time nor capacity to undertake the task.
Second, the system was not supported by online services and automation to speed up
processing. This problem was exacerbated by lack of coordination among
government offices. Processing involves different government offices such as the
MOI, the ERCA, FIA, NBE, banks, and shipping agencies. On top of this, there was a
lack of trained personnel, weak supervision, and corrupt practices. Third, domestic
firms supplying to exporters were unable to benefit from this system, due to deficient
design and (p.94) implementation. The system, moreover, failed to improve over
time. Manufacturers, the supposed key beneficiaries, were not involved in the design
and implementation processes. Finally, this system was further weakened by lack of
awareness and anti-export and anti-manufacturing biases among different levels of
government agency. Chapters 4–5 will further explore how these instruments failed to
support industrial policy.

Export Target-Setting and Monitoring

Developmental states have used industrial policy in a visible way to discipline the
private sector by providing rents that are linked to performance. An export target-
setting system has been used to review monthly export performance since 2010.
Experience in Korea and Taiwan shows that export discipline and monitoring was
instrumental in export-led industrialization. The export target-setting instrument has,
however, been largely ineffective in Ethiopia, thanks to the passive engagement of
exporters, and, in part, the government’s inability to provide the support required to
operationalize targets. Target-setting requires a comprehensive approach and shared
commitment (or compulsion) involving both government and industrialist. In
addition, an export award was endorsed in a regulation, but was ineffective and did
not have the intended outcome. No effort was made to target industrialists for the
award, rather than all exporters. The scheme was discontinued after some recipients
of the award were charged with tax evasion and other illegal activities.

Diminishing Role of Import Tariffs

Tariffs are used to protect domestic production against import competition, although
their application has diminished over time. Trade liberalization waves since the mid-
1980s, structural adjustment reforms in Africa, and the requirements of the World
Trade Organization (WTO) accession have shaped the rules of international trade.
Ethiopia began trade liberalization under the auspices of the IMF and World Bank in
the early 1990s. Tariff reduction was one of the components of this liberalization. The
government has adopted a gradual approach, in which seven major revisions have
been made to customs tariffs between 1993 and 2011. The upper tariff rate decreased
from 230 per cent before 1993 to 35 per cent in 2011, while tariff bands decreased
from twenty-three to six (MOFED 2011a, 2012c; ERCA 2012b). The simple average
tariff rate decreased from 41.6 per cent in 1992 to 20 per cent in 2011, while
the (p.95) weighted average rate decreased from 79.1 to 17.5 per cent in the same
period. Import bans were used in exceptional cases, notably cement; and were used to
discourage the export of raw materials, for instance, raw hides and skins in 2012
(MOFED 2013b). Nevertheless, Ethiopia’s tariffs in 2012 were still higher than those
in many SSA countries. Ethiopia, as a WTO applicant, is expected to further reduce
its tariffs.

3.6 State as direct economic actor and privatization

3.6.1 Direct role of state as economic actor

Contrary to mainstream viewpoints, activist states use SOEs as direct producers and
industrial players with multiple goals (see also Chapters 2 and 7). Ownership
structure has no relationship to inferior performance or bureaucratic inefficiencies,
and SOEs can play a key role in catch up (Chang and Singh 1997; Chandler 2004;
Chandler et al. 1997; Nolan 2012; Amsden 1989). Direct participation in economic
activities has been cardinal to the government’s industrial policy in Ethiopia, and may
have been influenced by many factors (FDRE 2002; EPRDF 2011c; EPRDF 2013e).
First, the low level of industrialization in Ethiopia and the limited capacity of the
domestic private sector (in terms of long-term, high-risk, and large-scale investments)
require an active government role to supplement the small private sector in
accelerating economic growth. Prior to 1991, the private sector had been prohibited
from engaging in industry. Second, given the low level of infrastructural and human
resource development, widespread market failures, as well as enduring perceptions of
political risk, the private sector is more likely to invest in areas of quick return, rather
than in productive sectors essential to long-term development. This has necessitated a
selective government presence in strategic areas by establishing SOEs, where
necessary, in joint developments with foreign investors. The ideological
underpinnings of this policy appear to be rooted in the government’s developmental
perspective, which underscores the state’s key and active role during catch up
(EPRDF 2011b). In addition, Ethiopia has a long history of SOEs dating back to the
pre-1974 imperial period and the Derg regime (1974–91). This history may have
shaped a domestic political outlook that is not hostile to this policy approach.
However, this is difficult to ascertain and evidence. (p.96) Recent policy has focused
on rationalizing major public enterprises by merging them into new entities. For
instance, Metal Engineering Corporation (METEC) was founded in 2011 from former
defence engineering plants. A new Ethiopian Shipping and Logistics Corporation
(ESLC), established in 2012, is a merger of Ethiopian Shipping Lines, Ethiopian
Maritime and Transit Enterprise, and the Dry Port Enterprise. A new Chemical
Industry Corporation (CIC) will focus on fertilizer, chemical, and cement enterprises.
However, it is not clear whether this reflects a coherent policy focused on the
development of national champions as key drivers of industrialization.

3.6.2 Privatization as complementary policy

Since 1980, privatization became the most generally prescribed cure for the ills of
SOEs worldwide (Cramer 1999c).5 This was based on the ideological belief that
government is a problem and should play a minimal role (Stiglitz 1998; Cramer 1999c;
Bayliss 2006; Bayliss and Fine 2008). Privatization became part of the IFIs’ structural
adjustment programmes (SAPs) that were imposed on almost all African countries as
a precondition for accessing loans. According to Stiglitz (1998), privatization should
not be seen as an end, and regulatory mechanisms should be put in place for a
positive outcome. The justifications for privatization included operational
improvement, efficient allocation of resources, reduced government spending on
SOEs, strengthening the capital market and FDI inflows, and technology and skills
transfers. Ethiopia’s privatization strategy diverged from IFI prescriptions and was
the subject of intense dispute. Late Prime Minister Zenawi (BBC 2003) remarked that
‘the IMF has been pressing the government to sell these state firms, but we resisted
these measures which would have resulted in the collapse of our business’. This
opposition has been recorded often enough.6 In terms of
institutional (p.97) development, privatization proceeded through a taskforce under
the prime minister’s office, assisted by multiple technical teams. An autonomous
agency, the Ethiopian Privatization Agency, was established in 1994, and was led by
a board of directors appointed by the prime minister. Another agency, the Public
Enterprises Supervisory Agency, was also established to supervise the public
enterprises that were made autonomous. Faced with operational ineffectiveness and
labour instability, both agencies were later reorganized as one agency, the
Privatization and Public Enterprises Supervisory Agency (PPESA). Under the Derg’s
command economy, all enterprises were structured as bureaucratic corporations. The
Federal Democratic Republic of Ethiopia (FDRE) Proclamation 25/1992 disbanded
these corporations, and public enterprises were to be independent legal entities
organized as profit centres, and to be led by their own boards of management
appointed by the PPESA (FDRE 1992).
Several conclusions can be drawn from Ethiopia’s privatization programme. First, in
parallel with privatization, government has also continued to invest, solely or in
partnership, in areas where it believes the market system would not invest. Saigon-
Dima Textiles is an example of the joint development between the government and
Turkish investors, which was initiated by the Turkish Party. The government has
pursued a gradual and pragmatic approach. This is consistent with Janos Kornai’s
argument that privatization is ‘the prime economic task of a change of economic
system’, and that a gradualist approach leads to organic development of the private
sector, by encouraging new domestic private sector investors to enter (Lindbeck 2007;
Kornai 1990). Some now argue that it was too slow (it took two decades), reflecting
lack of government commitment to privatization and of private sector development.
This conviction is further underscored by the government’s embarkation in recent
years upon large-scale investments in sugar, fertilizer production, metal, and
engineering. The sequencing of the privatization programme was arguably logical,
given institutional capacity, the objective of promoting the domestic private sector,
and building confidence/credibility. Privatization began with smaller firms (such as
those in retailing), later advancing to bigger firms as experience was gained. This
sequencing was similar to that adopted by Mozambique and Zambia
(Cramer 1999c, 2000).
Second, the lack of domestic private sector capacity in terms of motivation, finance,
and managerial capacity to buy about-to-be-privatized firms posed a major problem.
The government deliberately used the (p.98)
Table 3.2. Summary of privatized firms and buyer profile
Period Privatized Revenue (million Yearly privatized Revenue per firm
firms ETB) firms (million ETB)

Firms In % Revenue In %

Phase 1 (1994– 230 73 3,100 24 29 13.5

2001)

Phase 2 (2002– 82 27 9,600 76 8 117

2011)

Total 312 100 12,800 100 17 41

Profile of buyers

Buyer profile No. of firms Share in %

Domestic buyers 264 85

Foreign buyers 31 15

Total 295 100

Source: Unpublished data of PPESA (2012)

privatization programme to foster domestic businesses. Preferential requirements
were specified, and many auctions targeted only the domestic private sector. As a
result, 85 per cent of firms privatized were sold to domestic buyers (see Table 3.2).
The outcome was expected to be even better, particularly in terms of the domestic
private sector’s buying some of the large-scale enterprises. However, the sector’s
weak response eventually led to more opening up of privatization to foreign firms.
Consequently, foreign buyers bought many large enterprises. For instance, three
breweries (Meta, Harar, and Bedele) were sold to two European firms (Diageo and
Heineken) for about $400 million in 2012. Government revised the less favourable
valuation method, and extended payment times. Domestically owned firms were
allowed to pay up to 65 per cent within five years, and foreign buyers up to 50 per
cent within three. The interest rate for delayed payment was to be based on NBE’s
saving rate ceiling.
Third, transparency and accountability were as important as outcome, thereby
arguably narrowing the scope for abuse of resources and corruption. About three-
quarters of the firms were transferred through open tender and auction. Regional
endowment funds bought no major privatized enterprise, and there is no evidence of
widespread gains by government officials or politically connected individuals
(Kelsall 2013). In 2000, two ministers were implicated in corruption related to
privatization, and jailed.7 This is described more fully in Chapter 5. Lack of
institutional (p.99) capacity (in the form of weak coordination among government
offices, shortages of skilled personnel, lack of appropriate valuation method and
effective promotion), was a major constraint. The use of external experience and
expertise appears to have been limited, and the process shows learning by doing. This
problem was bigger particularly in the early phases of privatization, but was less
costly as the focus was on smaller firms.

3.7 Coordination and sectoral organs

3.7.1 National Export Coordination Committee (NECC)

List (1856) emphasizes that a nation’s effective political economy is dependent on the
strength of its institutions. Institutions have played critical roles in industrialization,
and have in turn evolved because of the catching up strategies adopted (Amsden 1989;
Rhee et al. 2010). In the Ethiopian context, a particular set of institutional mechanisms
was developed, including the National Export Coordination Committee (NECC),
MOI and its affiliated institutes, and SOEs such as DBE. The following analysis is
based on a few selected institutions.
NECC was established in 2003 after the endorsement of the new industrial
development strategy in 2002. Its precursors included the Export Promotion Board
(1995–98) and the Ethiopian Export Agency (1998–2002), which had been largely
ineffective (FDRE 1998a, 1998b). The NECC’s aims and mandates are promoting
exports and improving coordination among government institutions. It is chaired by
the prime minister and is composed of representatives of relevant government
ministries. Membership was limited to fewer than fifteen, although recently it has
exceeded twenty-five.
NECC meeting agendas have included performance reviews (monthly targets and
actual performance) of the overall export sector and each ministry. Discussion has
been based on the report of each relevant ministry and agency, and focuses on
constraints requiring decisions. Between 2003 and 2012, NECC had ninety meetings.
Monthly meetings have rarely been cancelled, have always been chaired by the prime
minister, and usually take almost a day. Where necessary, difficult decisions were
made by the prime minister, the highest executive decision-maker. Decision-making
was improved by the presence of most of the key actors, and (p.100) generally issues
were discussed until consensus was reached. Partly thanks to NECC’s leadership,
exports grew on average by about 22 per cent per annum between 2006 and 2012.
During the same period, merchandise export earnings more than tripled from US$1
billion to US$3.2 billion. The NECC has continued to function effectively since the
death of its founding chair, Prime Minister Meles Zenawi, who is recognized as the
architect of the country’s industrial development strategy (EPRDF 2013b).
The Korean parallel is the National Export Promotion Meeting established in 1961
and chaired by the Korean president, General Park, which met on a monthly basis. It
had 172 members, including thirty-six cabinet ministers and state ministers, fifty
presidential and ministerial staff, fifty representatives of firms and industrial
associations, eight bank representatives, and nine representatives of universities and
research institutions. Its name later changed to National Trade Promotion Meeting. Its
sole aim was export promotion, and to this end monthly targets were set and
decisions, characterized by pragmatism and flexibility, and were made on the spot
(Rhee et al. 2010). Rhee et al. attribute its effectiveness to the capable and effective
bureaucracy, which provided current and dependable information; the alliance
between the state and industry; and the obsession and single-minded focus on exports
as prime driver of the economic growth. Through the meetings, government
intentions were conveyed to firms and industries, incentives were negotiated, and
information relevant to government decisions was received. Overall, the meetings
served as a vehicle for the national export promotion campaign and to unify the
export sector. Because of an almost fanatical export discipline that linked export
growth to the nation’s survival, exports increased from $60 million in 1961 to $22
billion in 1982. This process was assisted by a ‘reciprocal control mechanism’, which
combined operationalized targets, accountability and sanctions, and performance
standards (Chang 1994; Amsden 1989; Studwell 2013).
Despite the Ethiopian NECC’s important role in expanding export earnings, its
achievements fall far short of the targets. Several factors have contributed to this.
First, NECC has had a narrow composition. For instance, important government
institutions, such as CBE, DBE, EAL, and institutes responsible for key export
industries were not included. Their participation might have improved the decisions,
coordination, and implementation. Industrialists and their industrial associations were
also not represented on NECC. Had they been, the committee would
have (p.101)been better informed about industrial constraints and opportunities and
better positioned to make sound decisions. Moreover, industries could have better
understood government intentions and priorities, and helped operationalize targets
and associated sanctions. This would have contributed to greater trust and a stronger
partnership between government and industry. Industrialist involvement would, of
course, give rise to technical and political questions about which industrialists to
include, and the mechanisms needed to ensure this marriage did not lead to political
hijacking by particular interest groups. In addition, sector-level dialogue platforms
needed strengthening, to allow for deliberation between government and intermediary
institutions. A political choice should also be made to balance ‘autonomy’ and
‘embeddedness’ (Evans 1995).
Second, limiting NECC’s agenda to direct export issues led to the neglect of other
manufacturing industries, thereby undermining the complementarities between export
and import-substitution industries. For most successful late industrializers, import
substitution has preceded exports, and continues to grow along with the export
industry (Amsden 2001, 2007a). Arguably, this neglect, together with emerging
political pressures from vested interests, contributed to the slow growth of Ethiopia’s
import-substitution industries.
Third, many NECC discussions focused on operational issues, often arising from
weak accountability systems (NECC 2012). The sub-committee structure was
intended to promote operational coordination among NECC members, but failed in
this role. NECC could have focused only on major coordination failures had most of
the more routine coordination problems been tackled at sectoral, regional, or local
levels. This underscores the importance of effective accountability, with sanctions for
those who do not deliver.
Fourth, a major obstacle to effective export target setting was the limited supply
response of the Ethiopian economy, determined by its productive capacity. This
structural rigidity has significant implications for export development. For NECC’s
close monitoring of progress in export expansion to work, it has to have a well-
operationalized target. Such targets have to take account of the supply capacity of the
domestic economy (and hence reflect individual firm’s targets) and the support
required to meet the targets (credit, logistics, land, etc.). This was clearly lacking in
Ethiopia. Therefore, simply setting targets (and even imposing sanctions for
undershooting) may be necessary, but is insufficient. Targets also need to (p.102) be
closely tied to analytical capability, a key area in which there has been too little
development in recent years. These issues are explored further in subsequent
chapters.

3.7.2 Institutes: Organizational key to developing industries

Ministry of Industry as Focal Agency

The focal ministry and lead agency in the manufacturing sector is MOI. There are
also specialized institutes that oversee and support specific industries (such as leather
and leather products, textiles and garments, basic metals and engineering), and
agencies with specific mandates (such as privatization, supervision of public
enterprises). These are part and parcel of a strategy to develop and improve the
institutional interactions between the state and specific sectors and industries (Buur et
al. 2012). MOI has been through a string of restructuring exercises almost every five
years, after each election. Factors such as the ministry’s structure at different periods,
the weight of political appointees (in particular, in improving coordination with other
federal and regional government institutions), and access to the prime minister have
influenced the ministry’s effectiveness. MOI has been largely unable to function as
the lead agency for industrialization due to its limited mandates and lack of political
influence over other institutions. This deficiency was raised on different occasions in
NECC, as MOI tended to bring coordination problems and constraints to NECC,
rather than itself effectively coordinating.
Institutes as Archetypes for Developing Specific Industries
Since 2009, autonomous institutes have been adopted as the main institution to lead
and develop key industries. The three institutes established were the Leather Industry
Development Institute (LIDI), Textile Industry Development Institute (TIDI), and the
Metal Industry Development Institute (MIDI). These institutes were initially used for
export-oriented industries, but later also supported import-substitution. They were
introduced after many failed attempts to use directorates within the regular
bureaucracy. Recently, there have been efforts to expand the ‘institute’ approach to
other sectors. Although belated, the institutes (for leather, textiles, and metal
industries) and an agency (for horticulture) have been instrumental in channelling
supports to their respective industries and firms. They have (p.103) also helped by
dealing with bureaucratic inertia, organizing the incentive structure, and
strengthening government–industry information flows and collective learning. These
institutes have been overloaded with administrative and facilitation tasks because of
the major bureaucratic obstacles in many government offices, a major constraint
facing firms. Company owners and managers argue that public service delivery and
facilitating trade and customs are more important than training and technological
support. The bureaucratic nightmare in government offices is evidence of leadership
failure, as the ultimate responsibility lies with the political leadership. Unless this
challenge is addressed satisfactorily, the institutes will not be able to focus on
developing the technological capabilities of firms. Moreover, the institutes
themselves suffer from a shortage of capable staff and insufficient understanding of
industrial policy, the industries, and related issues. They also lack the passion to push
the industry policy. A twinning programme has been instituted between sectoral
institutes and international organizations, apparently with promising results.
However, research collaboration between institutes and universities remains weak.
These issues are further explored in the following three chapters.

3.8 Summary and conclusions

Ethiopia has embarked on industrial development after a history of inadequate
industrialization from the mid-1950s to the early 1990s, based on ISI that failed to
bring rapid growth and structural transformation. In the 1990s, the government
focused on postwar recovery, economic and political reforms (including federalism),
and the creation of a multiparty polity. The Ethiopian government has certainly
engaged in policy experiments and largely maintained its autonomy. This is partly
because of the degree of legitimacy it achieved throughout the war and war-to-peace
transition. Partly also, it is because of the government’s foresight in crafting and
using the country’s geopolitical significance. Economic growth has been impressive,
although structural transformation lags far behind.
This chapter suggests that Ethiopia in recent years has exhibited features of a
‘developmental state’. It has also suggested that there is a strong case, based on
economic history and structuralist development economics, (p.104) for such a
developmental state, in particular, for an activist focus on industrialization. The proof
of the pudding, however, lies in the detailed evidence of policy implementation and
performance in manufacturing overall and in particular sectors. Manufacturing to date
has played an almost insignificant role in the Ethiopian economy, including
employment creation and export earnings. However, there may yet be significant
implications from recent manufacturing policies and performance for future
directions. This book has been motivated precisely by this history and reasoning and
its aim is to observe and analyse in detail the performance of, challenges to, and
industrial policy experiments of the activist developmental state. More broadly, the
book aims to explore the wider feasibility and rationale for an activist state in
applying industrial policy in SSA.

Notes:
(1) This episode is also recalled in Michela Wrong’s (2005) I Didn’t Do It For You: How the world
used and abused a small African nation.

(2) For instance, see CSO and MCIT (1969).

(3) The estimated dollar–ETB exchange rate was about ETB 10 in 2006, ETB 16 in
2011, and ETB 18 in 2012.
(4) See also CSA Statistical abstracts including (CSA 2001, 2002, 2007, 2008, 2010,
and 2011b), CSO (1983), and CSO (1987).
(5) In Private Island: Why Britain Now Belongs to Someone Else, James Meek (2014)
reveals how privatization was carried out assiduously across many sectors in UK
without bringing about economic dynamism. Rodrik (2012, 151) highlights that
‘privatization would have been the conventional route, but it was ruled out by the
Chinese Communist Party’s ideology’, and China’s economic miracle after the 1980s
is hard to deny.
(6) For instance, see BBC (1 September 2003) ‘Ethiopia hits out at IMF’; AFP (31
August 2003) ‘Ethiopia rejects proposal to privatize loss-making state firms’;
<https://brian.carnell.com/articles/2003/ethiopia-and-the-international-monetary-fund-at-
loggerheads-over-privatization/> ‘Ethiopia and the IMF at Loggerheads Over
Privatization’.
(7) See also World Bank (2012b).

Cementing Development? Uneven

Development in an Import-Substitution
Industry
Chapter:
(p.105) 4 Cementing Development? Uneven Development in an Import-
Substitution Industry
Source:
 Made in Africa
Author(s):

Arkebe Oqubay
Publisher:
Oxford University Press

DOI:10.1093/acprof:oso/9780198739890.003.0004

Abstract and Keywords

This chapter provides a sector-specific analysis of the uneven development of cement
production—a strategic import-substitution industry driven by and contributing to the
growth of the domestic market. The industry has acted as a binding agent of
economic development and transformation. It has undergone major changes in the
period under consideration, growing faster than in most developing countries. The
chapter also examines the industrial structure, linkage effects, policy instruments, and
institutions of the sector. In sharp contrast to the overall domination by multinationals
of the African cement industry, domestically owned firms continue to dominate the
industry in Ethiopia. The analysis explores the role that government policy has played
through direct and indirect interventions. It probes the successes and failures in
policymaking and asks what learning experiences there have been in the cement
sector, given the high-risk trade-offs involved in a context of very scarce resources.
Keywords: Ethiopia, cement, import substitution, learning, industrial structure, construction
industry, housing programme, scarce resources, multinationals

4.1 Introduction
There has been a dramatic transformation in the use and especially production of
cement in Ethiopia over the past hundred years or so. Imported Portland cement was
used during the construction of the Ethio-Djibouti railway between 1904 and 1917.
The Italians, during the invasion of 1938, built the Diredawa Cement Factory, with a
yearly capacity of 30,000 tons. In the mid-1960s, the Ethiopian government built two
cement plants at Massawa (in Eritrea) and Addis Ababa, with a combined annual
output of 150,000 tons. Two production lines (with an installed capacity of 600,000
tons of clinker per annum) were built at Mugher between 1984 and 1991. All state-
owned factories were managed by Mugher Cement Enterprise from the mid-1980s.
After a long period of sluggish development prior to the 1990s, the Ethiopian cement
industry recorded impressive growth between 2000 and 2012. Installed capacity in
the industry rose from 800,000 tons in 1999 to 10 million tons in 2012 (MOI 2012b;
MOI 2013). The average annual growth rate for cement production was more than
twice that of Africa or the globe during this period. By the end of 2012, the number of
firms had increased from a single SOE to sixteen firms. New firms upgraded their
technologies and exploited economies of scale. The cement industry has undergone
major changes throughout this period, and it appears that Ethiopia is likely to become
one of the top three (p.106) cement producers in Africa. With the existing growth
forecast and 300 kg per capita consumption per annum, domestic demand could reach
36 million tons in 2020. Average global cement consumption per capita was 390 kg in
2012 (MOI 2012b). In sharp contrast to the overall African cement industry, where
multinationals dominate, domestically owned firms are the leading players in
Ethiopia.
Government industrial policies have shaped the cement industry’s development
through direct and indirect interventions. Direct support including subsidized long-
term investment loans, investment promotion incentives, and access to the mining
resources required as inputs. On the demand side, the government’s large-scale
housing and infrastructure development programmes, combined with the expansion
of private sector construction, has led to a building boom since 2005
(MOFED 2012b). This became the prime driver of growth in the cement industry.
Nevertheless, the process was also fraught with difficulties. The government did not
make timely interventions in the face of new challenges, and lacked a realistic long-
term strategy for the industry. Consequently, it failed to prevent and contain cyclical
crises in the industry. The country faced critical shortages and was compelled to
import cement between 2006 and 2011, while in 2012 firms had excess capacity and
surplus production. The industry also suffered from lack of competitiveness and low
productivity. Moreover, domestic technological capability remained at a relatively
low level.
Different methodological tools—including surveys, interviews, site visits, and
document reviews—have been used to provide the evidence and insights and enable
further exploration of these issues in this chapter. Where appropriate, the experiences
of other countries (such as China, India, and Korea) with established cement
industries have been drawn upon. The quantitative and qualitative survey was a
quasi-census of all sixteen firms in the cement industry. In addition, qualitative and
in-depth interviews with forty individuals in twenty-eight establishments were
undertaken. Site visits to twelve cement factories provided additional insights. A
review of hundreds of documents was conducted, including primary and secondary
sources. In addition, detailed historical and current data from the two major cement
factories (which produced 100 per cent of Ethiopian output until 2007) were
consulted. Where possible, this chapter uses data updated to mid-2013, thereby
enabling a better grasp of ongoing trends within the industry. The International
Standard Industrial Classification (ISIC) and US Geological Survey (USGS)
definitions have (p.107) been adopted to allow cross-country comparisons and
matching against surveys conducted by the Central Statistics Authority (CSA) of
Ethiopia.
The chapter reaches the following conclusions. First, the state provided support to the
cement industry in many ways, including massive policy-induced industrial
expansion through direct support to the sector, and to the construction industry as a
whole. Second, some of these policies were more effective than others and the sector
has absorbed a large share of very scarce resources. Some policies were perhaps
‘over-designed’—they added little to the main investment stimulus, the rapid
expansion of the market. Third, arguably indirect policies, notably government
investment in housing and infrastructure, were more effective than some of the direct
support. There have been trade-offs (foregoing fiscal revenue to create incentives;
facilitating the import of inputs versus supporting the export sector through
devaluation) and learning experiences. The tensions/trade-offs are particularly
noteworthy: foreign exchange constraints and allocation of scarce resources to
promote an important industry, as well as the mixed results of policy.
The following sections analyse the sector’s performance, followed by discussion of
the emerging industrial structure of Ethiopia’s cement industry. Then, linkages and
industrial development are addressed. The last section discusses policies and related
instruments, institutions, and the policymaking process.

4.2 Sector performance

4.2.1 Output growth patterns

Between 2006 and 2010, the Ethiopian cement industry grew on average by 12.4 per
cent per annum, which was twice the world’s and Africa’s average growth of 4.5 per
cent (Figure 4.1).There was a more than tenfold increase in annual cement output
from the 237,000 tons in 1991–92 to more than 3 million tons in 2010. Growth in
output was steady throughout the period, with the exception of a surge in 2000 due to
the commissioning of a second factory, which doubled output. Cement output also
experienced another surge in 2012, after the commissioning of new factories and
expansion projects. Due to spurts in investment and long gestation periods, the
cement industry often experiences uneven annual rates of growth.
(p.108)

Figure 4.1. Production (in tons), 1978–2013

Source: CSA, Report on Large and Medium Scale Manufacturing and Electricity
Industries Survey, 1978/79-2010/11; MUDC July 2013
Table 4.1. Comparison of growth rate and rank
Description Average annual growth 2013 Installed (million 2012 production
rate tons)

Growth rate Years Actual Ranking

(%) (tons)

International
comparison

World 4.5 2006– 3,700

Description Average annual growth 2013 Installed (million 2012 production
rate tons)

Growth rate Years Actual Ranking

(%) (tons)

2010

China 10.7 1998– 2,150 1st

2010 Globally

African comparison

Africa: total 4.5 2006– 140

2010

South Africa 1.8 2006– 13 1st SSA

2010

Nigeria 14.2 2006– 35 5.4 2nd SSA

2010

Ethiopia 12.4 2006– 12.2 4.2 3rd SSA

2010

Kenya 10.8 2006– 4.1 4th SSA

2011

Source: Own computation from USGS (2010a, 2010b, 2011a, 2011b) and CSA (2012)
As of 2013, installed production capacity in Ethiopia exceeded 10 million tons per
year. This improved the country’s ranking from the fifteenth largest cement producer
in SSA before 2000 to the third behind South Africa and Nigeria. East Africa’s
cement industry has shown faster growth after 2005 than other African regions.
Nigeria and Kenya have similarly witnessed fast growth (Table 4.1). However,
Ethiopia’s per capita cement consumption (69 kg in 2012) still lags the average per
capita consumption of developing countries. In 2012, the per capita consumption of
China, Vietnam, India, and Africa was 1,500 kg, 700 kg, 200 kg, and 150 kg
respectively. This estimate is based on United Nations Department of Economic and
Social Affairs’s (UN-DESA) population figures; and exports have not been deducted
as they are considered insignificant.

(p.109) 4.2.2 Employment growth patterns

As the cement industry is capital intensive, its direct contribution to employment is
limited. Personnel in cement factories increased from 1,648 in 1992 to 7,233 in 2012
(Figure 4.2). The average wage in the cement industry is higher by 122 per cent than
in the leather and leather goods sector. In 2011, the average monthly wage in the
latter industry was ETB 965 for male workers and ETB 864.43 for female. The
cement industry’s average wage was ETB 1,835. Despite these higher wages, labour
costs in cement factories account for 3.5 per cent of total production costs.
On the other hand, the cement sector has generated significant employment through
forward linkages to downstream cement product manufacturers (concrete products
and ready-mix cement). As part of its drive to create jobs and build capacity in the
construction industry, government has promoted small enterprises in the cement
products subsector since 2003. The cement and cement products industry together
accounted for 10 per cent of total recorded manufacturing employment in 2011, thus
highlighting the indirect employment-generation effects of cement production. There
are also small unrecorded enterprises producing cement-based products for
households, which employ many people. This puts the cement industry in second
place after the food and beverage industry and ahead of such labour-intensive sectors
as textiles and leather (Table 4.2). It is worth recalling that the employment created in
the construction industry has not been included here, as the direction of linkage is
primarily backwards from the construction industry to cement, not the reverse.
Nonetheless, to the extent that the expansion of a domestic cement

Figure 4.2. Total employees in the cement sector, 1993–2011

Source: CSA, Reports on Surveys on Manufacturing Sector from 1993 to 2011
(p.110)
Table 4.2. Annual growth rate of cement industry, 2006–11 (in per cent)
2006–07 2007–08 2009–10 2010–11 Average annual growth rate
(%)

Total manufacturing 5.2 13 24.7 (1) 4.2

Cement industry 26.6 7 21.1 4.3 14.7

Cement products 70 20 (10) (17) 15.7

industry

Source: CSA Survey on Manufacturing and Electricity (2012)

Table 4.3. Personnel composition and profile in cement and leather/leather goods
sectors
Variables or ratios Cement Leather Variance in %

Female: total employees 13.4 46.3 28.9

Professional staff: total employees 8.2 4.3 190.7

Variables or ratios Cement Leather Variance in %

Technologists-engineering: professional 42 22.2 189.2

staff

Technicians: total employees 16.3 6.5 250.8

Expatriates: total employees 7.9 4.5 175.6

Production: support staff 89:11 77:23 115.6

Source: Oqubay (2012)

industry has been fundamental to the continued expansion of construction in the
country, cement may be argued to play an employment-generation role significantly
greater than what is captured in direct employment data for the industry and in
statistics covering forward-linked enterprises.
The scaling down of regional housing development programmes is a major reason for
the slower growth in the cement products industry in 2009–10 and 2010–11. Cement
enterprises rely less on female employees (below 15 per cent), and use twice as many
higher professional and technologically skilled personnel than more labour-intensive
industries such as leather and related products (Table 4.3).

4.2.3 Capacity utilization and cost efficiency

Capacity utilization and cost efficiency in the form of the cost per ton of cement offer
clear insights into the industry’s performance. The former is the achievable maximum
output without increasing the unit cost. In the cement industry, capacity utilization is
measured either in terms of clinker production, which is directly related to kiln
capacity, or of cement production. Cement plant and equipment manufacturers
usually specify the maximum attainable capacity of cement factories for
benchmarking. (p.111)
Table 4.4. Capacity utilization in Messebo and Mugher Cement
Year Mugher (in %) Messebo (in %)

Cement Clinker Cement Clinker

2000–2002 105.26 101.18 33.61 34.80

2003–2005 106.25 101.41 101.74 83.73

2006–2008 92.19 96.09 139.03 109.53

2009–2011 69.01 67.97 101.14 80.85

Source: Own computation based on Messebo and Mugher data (2012)

Moreover, in terms of cost efficiency and productivity, the key determinants are
capital productivity and capacity. Due to the capital-intensive nature of the industry,
and high fixed costs, labour plays a marginal role in total productivity.
Table 4.4 shows the pattern of capacity utilization. Although Mugher Cement
Enterprise (MCE) enjoys a proximity advantage over Messebo Cement Plc (MCP) to
the major of Addis Ababa market (90 km vs. 870 km), Messebo exhibited better
performance in terms of cost efficiency and capacity utilization. This was because of
advantages in plant technology and a more responsive management. Capacity
utilization in the first few years following commissioning are usually low. Between
2006 and 2010, capacity utilization increased as market demand rose (Table 4.4).
Power shortages are the major factor that reducing achievable capacity utilization.
After 2011, with the entry of new firms into the industry and improvement in the
electricity supply, excess production capacity in cement was created. The domestic
market failed to expand to match increased productive capacity. Six firms were
operating below 50 per cent of capacity, four firms at between 60 and 80 per cent,
and only four reached 80 per cent capacity. Messebo and Mugher have operated at
higher capacity in 2011–12, and between 50 and 60 per cent in 2012–13, and
produced/sold 2.15 million tons. This variance in capacity utilization is associated
with the excess production capacity of factories. In the survey, the firms identified
fierce competition for a limited market as the main reasons for low capacity
utilization. The firms also indicated uncertainty and seasonality of demand as their
prime concerns.
The second related performance parameter is productive efficiency (cost efficiency),
which shows the ability to produce a ton of clinker or cement with the least possible
inputs and resources. Energy efficiency can be an (p.112)
Table 4.5. Energy and cost efficiency (Mugher and Messebo)
Year Share of energy of the total cost (%) Energy cost per ton cement (ETB)

Mugher Messebo Mugher Messebo

2000–2002 55.80 18.14 171.43 223.22

2003–2005 49.46 33.56 197.50 216.54

2006–2008 60.47 48.61 333.97 322.52

2009–2011 67.95 44.86 692.47 299.81

Source: Mugher and Messebo Data (2012)

important proxy for measuring productive efficiency, since energy is the major cost
component in cement production. Energy has increasingly become the most
important cost component, given global fuel scarcity and related price increases. The
cost per ton of cement should be contrasted with best industrial practices and
industrial benchmarks. Energy efficiency in Ethiopia during 2000–11 was much
lower than the accepted industrial norm (Table 4.5). Mugher’s cost of energy per ton
exceeded $50 in 2009–11. Messebo has eventually reduced its cost by a third by
shifting to coal. There has been no national energy savings programme, unlike in
Asian countries such as Japan, China, and India. According to studies on energy
efficiency, energy savings benchmarks can help reduce consumption by up to 15–25
per cent (World Bank 2009a; UNCTAD-UNIDO 2011). Energy accounts for 40 per
cent of manufacturing costs in India, and coal accounts for up to half of the energy
costs (CSTEP 2012). Until recently, the competitive pressure in Ethiopia was low
(low rivalry among domestic firms, weak pressure from imports, and less pressure to
export), which has weakened the motivation to improve performance and
productivity. The cost efficiency of small firms that use Vertical Shift Kiln (VSK)
technology was higher than those using rotary kilns, and unit costs decreased with
bigger kiln capacity. In general, factories have cut their energy costs from ETB 240 in
2011 to ETB 180 in 2012.
These characteristics and the trends in the Ethiopian cement industry have policy
implications for managing demand, developing technological capability, increasing
capacity utilization, and raising energy efficiency. Without investment in energy-
efficient technology and raising productivity, the competitiveness of the Ethiopian
cement industry may be eroded further.

(p.113) 4.3 Industrial structure of the cement industry

4.3.1 Global trends

This section discusses the technical and economic characteristics of the cement
industry, its shifting structure, the roles of cement manufacturers, and the
implications for the political economy. The industry has undergone structural shifts,
particularly since 2008. Because of improvements in process technology, economies
of scale have increased: for instance, kiln size doubled in terms of clinker output.

Global Cement Industry

Global cement production has shown a constant annual growth rate of about 5 per
cent from the 1970s. After that date, growth has shifted to emerging economies in
Asia, including Japan, Korea, China, and India. In 2011, total global production
reached 3.7 billion tons of cement (USGS 2013). There were 2,360 integrated cement
factories and 750 independent grinding plants globally. China’s share is 1,000
integrated plants and 350 grinding plants (excluding obsolete factories). Currently,
Asia is the biggest player, contributing more than 70 per cent of production and
consumption (see Table 4.6). The share of emerging economies has increased to about
three-quarters of total production. China is the world’s biggest cement producer and
consumer, with about 60 per cent of world production (USGS 2013).
Given the product’s low value–high volume characteristics and the global abundance
of the necessary raw materials, cement is usually characterized as a non-tradable
commodity. Although shipping and mass transport advances have enhanced its
portability and tradability, less than 10 per cent of output was traded globally during
the last decade
Table 4.6. Global cement production outlook (2012), in million tons
Country Amount Share (%) Country Amount Share (%)

World 3,700 100

China 2,150 58.1 Iran 65 1.8

India 250 6.8 Turkey 60 1.6

USA 74 2 Russia 60 1.6

Brazil 70 1.9 Japan 52 1.4

Vietnam 65 1.8 Korea (Rep) 49 1.3

Source: USGS (2013)

(p.114) (Selim and Salem 2010; COMTRADE 2012). In 2010, the international trade
in cement was 5 per cent, a volume of 151 million tons (67 per cent through seaborne
trade). This was 8 per cent less than the 2007 figure. The top five exporters were
Turkey (19 million tons), China (17 million tons), Thailand (14 million tons), Japan,
and Pakistan (10 million each). The dominant producers and consumers in the global
market were multinational corporations such as La Farge (French), Holcim (Swiss),
Heidelberg Cement (German), Italcemento (Italian), CEMEX (Mexican), and CNBM
(Chinese), which between them produced about one billion tons of cement in 2012. In
terms of technological advancement and equipment manufacturing, West European
manufacturers dominated the industry during the twentieth century. Currently, China
is the dominant player, accounting for the manufacture of more than 40 per cent of
global cement equipment (China Cement Association 2013). This global shift has
shaped the technology used in Ethiopian industry. While almost all large cement
factories sourced their technology from Europe until 2000, China has over the last
decade become the source of such technology in Ethiopia. For instance, Sinoma
International and CNBC supplied large-scale cement manufacturers in Ethiopia
(Mugher and Dangote, and Messebo and Derba respectively), thereby lowering
investment layout and making production relatively cheaper.
Africa’s share of global cement production was less than 5 per cent in 2011, and the
cement industry was dominated by multinationals (USGS 2010a, 2010b, 2011a, 2011b,
World Bank 2009a, 2009b; Wu 2004). In 2011, annual cement production in Africa
reached 150 million tons, which were produced in 190 plants in more than thirty
countries. Only a quarter of these plants were integrated plants, producing clinker and
cement. The biggest producers were the North African countries Egypt, Algeria,
Morocco, Tunisia, and Libya, which accounted for more than half of total African
production (USGS 2010b). This was not exclusively driven by growing domestic
demand, but also partly by stricter environmental regulations in Europe that forced
firms to relocate to North Africa (Selim and Salem 2010). Recently, the regional share
of East and West Africa has increased, while North Africa’s share declined relatively.
This stemmed from the political situation in North Africa since 2011, and increased
investment in SSA countries such as Nigeria and Ethiopia. Four multinationals, La
Farge (fifteen factories), Holcim (seven factories), Heidelberg Cement (eleven
factories) and Italcemento dominate the African cement (p.115) industry, with the
bulk of their production concentrated in North Africa (Imara 2011). These
multinationals produced close to half of African output in 2011. Currently, a new
player, Nigerian-owned Dangote Industries, has increased its presence in many
African countries, including Ethiopia. The efficiency and productivity of the African
cement industry is low compared to the performance of the industry elsewhere
(World Bank 2009a, 2009b).

4.3.2 Performance latitude, economic, and technological characteristics

Economic and Technological Characteristics

Given its technological and economic characteristics, three features of cement
production are particularly significant in low-income countries like Ethiopia. First,
the capital intensity and economies of scale, as well as the strategic significance of
cement in a period of rapid growth and structural change in a very low-income
economy, created a build-up of pressure that called for active involvement by the
state. In particular, the size of the investment required has put considerable pressure
on DBE, the country’s only investment bank. Second, the nature of production and
economies of scale and scope have necessitated the rapid acquisition of effective
organizational capabilities. Unlike the family firms that can thrive in leather or
floriculture, success in larger cement firms has required salaried managers and
professionals with technical skills. Evidence from elsewhere (Chandler 2004;
Amsden 1989) suggests that large national enterprises may play a greater role in
creating capabilities transferable to other industries. Third, the cement industry has a
narrow latitude for failure. Demanding performance standards play a positive role in
the implementation of industrial policy. As the evidence reviewed in this chapter
shows, however, a narrow latitude for failure does not prevent inefficiencies and
mistakes in industrial policy.

Economies of Scale and Scope

The economies of scale and scope differ from industry to industry and are related to
the specific production process used by firms (Chandler 2004). According to Drucker
(1999), the commonest production processes are the ‘unique product production
process, mass production (flexible and rigid), and continuous production’.
Continuous process (flow) production (p.116) systems are operated in the cement
industry, among others. Technological progress shapes the nature of the production
process. For instance, early cement plants were small and used VSK, in contrast to
modern cement production using bigger rotary kilns. The cement industry’s
continuous process depends on the uninterrupted flow of inputs such as energy and
raw materials. This places serious demands on electricity providers, fuel logistics,
and the organization of inputs and throughput. It also determines the nature of
distribution and marketing of goods.
The large-scale nature of the industry acts as a major barrier to entry and exit. With
increased scale, unit costs of production diminish. Hence, certainty of market demand
is both a prerequisite and an advantage. This is why, particularly in developing
countries, the government is better placed to invest in such large-scale undertakings.
This is in line with what Hamilton highlights, that ‘to produce the desirable changes
as early as may be expedient may therefore require the incitement and patronage of
government’ (Hamilton 1934: 204).
During the different development phases of Ethiopia’s cement industry, kiln size has
increased and technology (automation, process) has improved, bringing significant
increases in factory capacity and capital investment (see Table 4.7).
In the Ethiopian case, the Mugher SOE and a factory in Tigray owned by EFFORT,
the regional endowment fund, were the pioneers and first movers in the cement
industry. They played the lead role, taking risks,
Table 4.7. Economies of scale and capital intensity in the Ethiopian cement industry
Factories Period Technology Capital
investment

1st generation (Massawa, 1960s Small scale; Rotary kilns: 100 tons per day Below ETB 10
Diredawa, Addis Ababa (TPD); Annual capacity: 30,000–150,000 million
cement plants) tons

2nd generation (Mugher- 1980s Large scale; Rotary kilns: 2 X 1,000 TPD Below ETB 300
Line 1 and Line 2) Annual capacity: 850,000 tons million

3rd generation (Messebo) 2000 Large scale; Rotary kilns: 2000 TPD Annual ETB 1,200
capacity: 850,000 tons million

4th generation (Mugher, 2010 Large scale; Rotary kilns: 3,000 TPD; 5,000 $120–351
Messebo, National, Derba TPD Annual capacity: 1.4 million tons each; million
and Dangote) 2.5 million tons each

Source: Own summary (based on firms’ documents and interviews)

(p.117) and absorbing or facilitating externalities. After seeing the growing demand
and generous returns, many private businesses began to invest in the industry. Of
course, there were private sector investments in mini-cement plants between 2005
and 2011. Rather than being based on a long-term business development strategy,
these investments were made to exploit the excessive rents created by shortages in the
domestic market. Despite skyrocketing cement prices, the huge capital requirements
of such projects is another factor preventing the private sector from making
immediate investment decisions. Thus, the private sector was (understandably)
slower to comprehend the full significance of investing in the cement industry in
Ethiopia. Consequently, the government and the endowment fund, although
themselves rather late starters, played a pioneering role in terms of investing at the
appropriate scale.
In terms of economies of scope, large cement factories were forced to integrate their
activities vertically. For instance, in distribution and sales, they relied on their own
marketing and sales force (except the mini-cement plants) rather than using external
wholesale or retail outlets. Likewise, they relied on their own or closely affiliated
transport facilities (for instance Messebo and Derba). One of the most critical inputs
is packaging, and most enterprises have established their own packaging factories
(for instance, Mugher, Messebo, and Derba). In terms of energy inputs, some
factories have engaged in coal mining (Messebo, Derba, National Cement,
Abyssinia), although this was still at an early stage at the time of this research.
Almost all cement manufacturers have their own quarries for mining limestone and
related inputs. This shows the significant dynamics in terms of economies of scale
and scope.

Narrow Latitude for Poor Performance

The latitude for failure in production depends primarily on the intrinsic technological
characteristics of the production process of an industry. Hirschman (1967: 87)
emphasizes that the lack of latitude for failure or, positively, ‘the presence of
discipline’, imposes ‘propensities and pressures to which the decision makers
themselves are subject’. His popular example is the airline industry, where
developing countries have the potential to develop successful airlines due to the
narrow latitude involved. In this industry, due to safety requirements, the space for
operating at a sub-standard level is limited. Safety and other features of cement
production have indeed imposed pressures on decision-makers. The (p.118) product’s
perishability exerts additional pressure on production and distribution. The shelf-life
of cement is 120–180 days. Moreover, seasonality and uncertainty of demand adds to
the pressure exerted by the perishability effect, as the rainy season (June to
September) is the slack period for construction in Ethiopia. Demand uncertainty is the
basic feature of cement and other similar industries, creating cyclical swings due to
the long gestation period. This implies that caution and long-term focus are essential
in decision-making.
Consistency of quality is an unconditional requirement for cement manufacturers,
because of the safety considerations for buildings and construction works. Ensuring
consistent product requires laboratory facilities in large cement factories, and also in
mini-cement plants. As cement is used for various purposes and concrete
specifications depend on geology, weather, topography, and building codes, constant
research into product development is critical to the industry’s growth. And
environmental protection is also becoming important, as the cement industry is a
major polluter.
4.3.3 Changing structure of the industry and political economy dimensions

Shifts in Ownership Structure

A major structural shift has occurred in the cement industry in the past two decades.
In terms of ownership, the industry has shifted from a state-owned monopoly
(Mugher) until 2000 towards private sector domination. Mugher accounted for two-
thirds of cement production in 2000, decreasing to a quarter in 2011–12. Another
ownership dimension of the Ethiopian cement industry concerns the role of domestic
firms. Two local cement manufacturers (Mugher and Messebo) dominated the
industry until 2005, accounting for about 95 per cent of the country’s cement
production. This ownership structure appears to be the outcome of deliberate
industrial policy. The fact that Mugher survived the privatization programme was due
to the government’s decision to maintain a presence in the industry. Likewise,
Messebo was one of the first of EFFORT’s investments.
Two interesting developments in the domestic private sector were the emergence of
National Cement from the old, small, privatized firm, Diredawa Cement, which was
acquired by East African Business Group (p.119) (EABG); and the establishment of
Habesha Cement, promoted by former managers of the Mugher SOE. In 2002, EABG
negotiated for and bought a small, abandoned state-owned cement mill in Diredawa
(built by the Italians in 1938) and upgraded the plant into a 150,000-ton capacity
facility. EABG then commissioned a study on a large-scale cement factory with a
capacity of 1.4 million tons. The factory was in the commissioning stage in mid-
2013. The owner of EABG is Ethiopian, a third generation industrialist who has built
an industrial zone and a food processing complex. Neither Habesha’s promoters nor
EABG had political ties to the state or ruling party. Habesha initiated the project,
mobilized equity from more than 16,000 shareholders (the highest number of
shareholders in large-scale manufacturing), attracted two South African partners, and
secured loans from DBE. This is a striking example of entrepreneurial dynamism in
the private sector.
Likewise, other domestic businesses entered into joint venture arrangements. Ture
Business group (which had imported Chinese equipment for more than thirty years)
entered into a joint venture with two Chinese private manufacturers. The joint venture
established a cement-grinding factory in Addis Ababa and two small cement plants in
Diredawa. Currently, domestically owned firms continue to account for about half of
total installed capacity in the industry. This contrasts starkly with the situation
elsewhere in Africa, where multinational corporations (MNCs) dominate (Global
Cement 2012). The foreign firms operating in Ethiopia are not traditional
multinationals. Among foreign investors in the cement industry, the Chinese are
dominant, with seven firms, followed by Indians. A Saudi tycoon owns the biggest
plant, Derba Cement, which is part of the largest conglomerate in Ethiopia, Midroc
Group, which is owned by Sheik Mohamed Al Amoudi, an Ethiopian by birth.
Another major investor is Dangote, whose plant was still under construction when
this research was carried out. Foreign-owned firms together account for the other half
of installed capacity in Ethiopia’s cement industry.
Changing Competitive Structure
Competition was virtually nonexistent when the two state-owned cement
manufacturers dominated the cement industry. Mugher Cement had a full monopoly
until 2000, when Messebo Cement began operations in northern Ethiopia. The
Messebo plant was built between 1998 and 2000, and has a capacity of 850,000 tons
of Pozolana Portland Cement (PPC). (p.120) Mugher had an advantage in terms of
proximity to the main market around Addis Ababa, and the first mover advantage in
acquiring skills, but also a steeper learning curve. After 2005, mini-cement plants
sprung up, but their market share has never exceeded 10 per cent, and they have had
an insignificant influence in shaping the cement industry.
After 2011, there was a shift to an oligopolistic market structure. A key milestone was
the increased production due to new entrants and expansion at two factories. Derba
Cement (the largest firm so far in the industry, with installed production capacity of
2.5 million tons), East Cement (with installed capacity of 700,000 tons), and
expansion projects of Mugher and Messebo (additional capacity of 2.8 million tons)
became operational more or less simultaneously in 2011. The market will become
more competitive as projects under construction come on stream in 2015. In addition
to sales promotions, transport and credit facilities are increasingly being offered to
customers by cement manufacturers (Nazret 2010) to win market share.

Interplay of Industrial Structure and Government Policy

Against this backdrop, it is important to examine the political dimension of this
changing competitive structure. Porter’s generic analytical framework is moderately
useful in addressing features of industrial structure, policy, and performance as they
change over time. However, a word of caution is needed: his framework tends to
undermine the pivotal role of the state, resulting in a deceptive separation of market
and state and a restricted focus (Aktouf 2004, 2005).
According to Porter (2008: 80), ‘industry structure grows out of a set of economic and
technical characteristics that determine the strength of each competitive force’. The
five underlying forces that determine the evolving industrial structure are the: threat
of new entrants; bargaining power of buyers; threat of substitute products; bargaining
power of suppliers; and domestic rivalry among existing competitors (see Table 4.8).
Porter adds that the relative pressure of each factor and their relationship are context-
specific: they vary by industry and across time. Likewise, an industry’s development
may involve a shift driven from inside the industry or from outside. More
importantly, government policies affect each of these five forces. In view of this
framework, it is noteworthy to assess the underlying changes in Ethiopia’s cement
industry and the effects of government policy.
(p.121)
Table 4.8. Interplay of shifting industrial structure and government policy
Five Policy intervention Shifts and outcomes
competitive
forces

Threat of new Cement industry has high a) High entry barrier was overcome; Number of
entrants: entry and exit barriers due to players increased to sixteen, primarily FDI and
economies of scale and high domestic private sector;
capital requirements. The
government policy included: b) Significant increase in total production capacity;

a) Long-term loans made c) Technology and equipment modernized, with

available for capital increased kiln capacity;
investment;

b) Providing investment d) The gate for newcomers has now been closed as
incentives to mitigate risks FDI participation is restricted and finance is
and induce new players; temporarily unavailable.

c) Mining resources close to e) Mugher continued to be an active industrial player

the market made available to with 20 per cent market share (currently has an
firms; installed capacity of 2.3 million tons);

d) Foreign currency
allocated on preferential
basis;

e) Government involvement f) The privatization of Diredawa cement factory

directly through SOE and induced new private investment of 1.4 million tons.
quasi-public enterprises;

f) FDI barred since the end

of 2013; opportunity for
loans for new entrants
narrowed.

Bargaining a) Government was the a) Construction boom acted as key driver of the
power of biggest buyer (above 50 per cement industry until 2011;
buyers cent or two-thirds share
between 2005 and 2013); b) Demand failed to cope with the supply and excess
capacity;

b) Shortage (2006–2011) and c) Buyer’s bargaining position increased; cement

buyers bargaining position price decreased and availability improved;
was weak. Hence, high price
and delay;

c) Import of cement during d) Seasonality, market uncertainty, product

shortages; perishability push the industry to improve
Five Policy intervention Shifts and outcomes
competitive
forces

d) Regulation of market e) Government’s action to stimulate demand

during shortages (allocation, (domestic and export) required;
price, etc.)

Threat of Alternative inputs (steel etc.) a) Insignificant impact in reducing demand;

substitute not extensively available;
products b) Limited scope for substituting concrete in the
a) Government introduced long-term.
the use of agro stone
(substitute) in government
projects, with modest
success.

Bargaining Transport and energy are key a) Supply of inputs at subsidized rate;1
power of supplies. Government:
suppliers
a) supplies heavy-fuel oil, b) Affected by the firms’ technological choices (kiln
coal, pet coke, and electricity types and capacity, distance from sources);2
to all firms;

b) Transport cost increased c) Transportation cost affected by distance from main

and shortage increased firms’ market and fleet availability;
margins. Government
imported 1,200 trucks for d) Backward and forward integration opportunities
cement transport, and are observed.
allowed firms to own
transport fleet;

c) Packaging materials:
major factories own
packaging factory.3

Rivalry among a) Competitive edge of firms Derba followed aggressive price war; Mugher was
existing differed. No scope for still in ‘sleep mode’; Messebo aggressively worked
competitors specialization; on efficiency improvement (production process and
energy use) and transport optimization. National
b) Price competition driven cement has focused on markets in neighbouring
primarily by price as product countries, reducing its investment cost, and selling
is homogenous. clinker to mini-cement plants. Mini-cement plants
were surviving by focusing on market niches.

Notes: (1) Mugher, Messebo, and Derba have their own packing factory and sell packaging material
to others. Messebo uses the permanent transport fleet of a sister company. Derba has imported 700
heavy trucks. National Cement was importing trucks. Currently, there is a shortage of freight
transporters and often reliability is a problem. This, and the existence of incentives (duty-free
import of vehicles for own transport service), have influenced the decision by cement factories to
operate their own transport fleets.

(2) HFO and coal are imported by government, and the administrative costs are
absorbed. Private petroleum distributors barely add to their profit margins. Although
official data are not available, the subsidy is estimated to exceed 10 per cent.
(3) In terms of kiln type, rotary kilns consume less energy than VSKs; bigger capacity
kilns consume less energy than smaller kilns; factories with coal conversion
technology benefited by using coal (cheaper than HFO) to power the kiln. Thus, the
supply has a varied effect on firms, depending on their technology choices and scale
of production.
Source: Own summary (based on collected data and analysis)
(p.122) It is possible to observe from this matrix that these patterns have significant
implications for industry, cement manufacturers, and government policy. First, in an
increasingly diversified sector, government leverage and ability to guide the industry
has not decreased substantially. Such means as guiding Mugher and influencing other
firms (such as Messebo, National Cement, and Derba), procuring cement from
foreign markets, supplying inputs, and becoming the single major buyer have been
used. Although new institutional mechanisms (such as the role of the industrial
association) will evolve, the presence of domestic firms (which have good working
relations with government) will allow government to lead the sector.
Second, the new competitive environment could be used by government to improve
the sector’s long-term competitiveness regionally and internationally, by directing it
towards exports through incentives, restrictions, and enforcement, and through
support schemes that focus on improving productivity and quality. Clearly, also,
stimulation of the domestic market requires definite government intervention. Current
low (p.123) capacity utilization and the resultant risk of bankruptcy have significant
implications for financial health of DBE, an indispensable component of industrial
policy in Ethiopia.
Third, the cement industry is a classic oligopoly, where a few firms dominate
(Chandler 2004). This can be measured by weighing the ‘four-firm’ concentration
ratio by measuring the market share of the four largest firms. In 2012, the
concentration ratio was 75 per cent, meaning that the ‘big four’ (Derba 2.5 million
tons, Messebo 2.3 million tons, Mugher 2.3 million tons, East Cement 700,000 tons)
had installed capacity to supply three-quarters of the market. Total installed capacity
of the overall cement industry was 10 million tons (excluding National Cement and
Ethio-Cement, which were not operational). This domination could lead to restrictive
trade practices through collusion, market-sharing arrangements, raising prices, or
curtailing production. Experiences elsewhere provide extensive examples of such
practices. The US cement industry (1930 and beyond), the Indian cement industry,
Pakistan, and South Africa are some examples. Conditions that lead to anti-
competitive behaviour and, to an extent, cartel formation, are high market
concentration, excess capacity, high entry and exit barriers, a collusive history in the
industry, and the role of the industrial association. Most of these factors appear to fit
the Ethiopian cement industry well. Thus, new industrial policies are now required to
prevent undesirable oligopolistic outcomes in the industry. With the ‘right
competition policies’ and government guidance, the oligopolistic market could be
turned to more efficient ends, including lower prices, better quality, and higher
productivity. For instance, in 2011, the per-ton cement price in China and Vietnam,
both low-cost producers, was $53 and $62 respectively. Brazil and the US were
medium-cost producers ($91–$92), while SSA countries were high-cost producers
(Nigeria $223, Angola $250, Ethiopia $175 for Ordinary Portland Cement [OPC]).
Mugher’s price per ton was ETB 2,900 (dollar exchange rate was ETB 16.5).

4.4 Cement-construction infrastructure linkages

4.4.1 Construction–cement linkage

This sub-section highlights the powerful linkages and feedback loops between
construction and the cement industry in Ethiopia. The Ethiopian (p.124) construction
industry grew by about 13 per cent per annum between 2004 and 2011, exceeding the
annual GDP growth rate of nearly 11 per cent. Its share of the economy expanded
from 4.2 per cent in 2000 to 5.8 per cent in 2011. The building sub-sector has
multiple socioeconomic impacts. First, it employs hundreds of thousands of unskilled
and skilled workers, thereby helping reduce urban unemployment (World
Bank 2009c). According to a CSA survey on construction in 2008–09, there were
1,384 construction firms (Grade 1–6), which employed 252,977 people at peak times
and 171,965 during the slack season. In addition, the producers of construction
materials employed 86,279 persons. This figure may be much higher since not all
smaller firms are registered. According to MOFED (2012a), the Integrated Housing
Development Programme (IHDP) and urban road construction programme employed
193,000 and 373,800 workers respectively in 2011. Second, it supports and
accelerates industrialization through cheaper and timely construction of factories.
Third, it contributes to the economical expansion of social infrastructure. Fourth, it
stimulates the manufacture of building materials. Fifth, it contributes to wealth
creation, housing and property development, and promotes savings and the
development of the financial sector. The construction industry can also play a
significant role in generating foreign exchange.

4.4.2 Policy learning through housing development programme

The government-sponsored IHDP demonstrates construction sector/cement industry
dynamics rather well. This programme is part of broader urban development and is
designed to redress the housing shortage, which is also a hot political issue for most
residents, particularly in Addis Ababa. In the capital, there was an estimated backlog
of 450,000 houses in 2005 (AACG 2003), which has increased since because of
population growth and rural-to-urban migration (UN HABITAT 2010). The mounting
demand for houses necessitated the massive construction of public housing through
IHDP.
The United Nations Human Settlement Program (UN HABITAT 2010) concludes that
the programme is unique in being ‘large-scale and pro-poor; an integrated approach
to housing, slum prevention and economic development’; as well as in creating
‘access to home ownership’. The report added that ‘the IHDP is not just a housing
programme, but a wealth generation programme through low-cost housing…In light
of Ethiopia’s (p.125) uncoordinated and inefficient housing sector, the Integrated
Housing Development Programme has provided a highly successful tool for
affordable housing delivery at large scale.’ The IHDP has gone through three phases:
pioneering the programme in Addis Ababa, scaling up, and a subsequent scaling
down. After successful completion of the pilot project, IHDP went on to construct
more than 30,000 houses in Addis Ababa in the first year. An autonomous agency
staffed with a committed and qualified management team was established. The
technical assistance of an international agency (German Technical Cooperation, GTZ-
IS) was sought for the pilot project, and the city government paid for its services.
Discussions with the public were conducted and 453,000 applicants registered for the
programme. The programme’s popularity and experience gained during the first
phase in Addis Ababa prompted the government to replicate it in other regional
towns. IHDP’s cement consumption exceeded half a million tons per year, amounting
to more than a quarter of production capacity in the country.
The second phase included the scaling up of the housing programme at national level.
The programme was one of the key priorities in the PASDEP five-year national
development plan, which called for the building of up to 900,000 houses through both
IHDP and private sector initiatives. Regional institutions were established in four
states and in selected towns. The Ministry of Works and Urban Development
(MWUD) spearheaded the programme.
Other government-sponsored building projects included the construction of thirty-two
public universities, more than 1,000 health centres, and housing projects for new
sugar projects. Financial resources and foreign exchange were earmarked for IHDP.
Through it, more than 150,000 houses were built in sixty-five towns. The cement
requirement reached one million tons per annum, accounting for about half the
country’s production capacity. IHDP, together with other public infrastructure
programmes and private real estate investments, created excess domestic demand for
cement. Cement imports thus became necessary (see Table 4.9). In 2011–12, the
national programme was scaled back. Meanwhile, high housing demand in Addis
Ababa continues. Potential homeowners interested in public housing have reached
one million (947,376 registered in 2013 out of which 42 per cent are female). The
housing programme involves several contribution schemes, whereby beneficiaries
can contribute 10, 20, or 40 per cent to reflect and encourage saving. (p.126)
Table 4.9. Cement imports 2006–11 (in tons)
Importer 2006 2007 2008 2009 2010/11 Total

Private 306,829 694,362 839,242 – – 1,840,433

Importer 2006 2007 2008 2009 2010/11 Total

Government – 11,000 335,147 821,547 650,000 1,817,694

Total 306,829 705,362 1,174,389 821,547 650,000 3,658,127

Source: MOI (2012b) and MWUD (2010, 2013) (Unpublished)

Linkage Effects to Building Materials Manufacture

The construction boom has boosted the local manufacture of building materials.
Consequently, the first glass factory, ten steel factories (mainly rebar manufacturers),
and fifteen cement factories were built between 2004 and 2012. In the meantime, the
construction industry suffered until domestic cement manufacturers could catch up
with growing demand. Capacity constraints delayed projects by more than a year and
construction costs increased by about 15 per cent (MWUD 2013). The government
used various instruments to tackle cement shortages and regulate the market (price
controls, cement allocations, import permits, and direct imports by government). The
government also intensified quality controls to curb the sale of sub-standard cement,
which had become a serious concern. It also increased transportation capacity.
Although the effects were limited until local cement production matched demand in
2012, these measures did enable partial containment of price hikes created by the
cement shortage.

4.4.3 Capacity building and modernization of the construction industry

A deliberate decision was taken to use public construction programmes to enhance
domestic construction capacity. The housing development programme is typical of
how government develops the capacity of domestic contractors, consultants, and
small enterprises. The shortage of building contractors was a critical bottleneck in the
housing programme. Modular, standardized, and economical housing designs, and the
innovative application of a sub-contracting model fuelled participation by small and
medium contractors. Government eased the entry barrier by revising the contractors’
licensing requirements to encourage participation by new and young professionals. It
also supported new construction firms with (p.127) government-financed training
packages, collateral-free loans, and direct access to government building contracts.
As a result, the number of contractors increased fivefold to more than 2,500. The
number of contracting firms with owners with engineering backgrounds also
increased, improving the technical capacities of the firms (MWUD 2009).
Likewise, the housing programme promoted the development of domestic
architectural and engineering firms. New architects and engineers were registered as
consultants, thereby allowing consulting firms to undertake design, contract
administration, and supervision. The increase in contractors and consultants has
intensified competition, changing the landscape of the construction industry. About
3,000 items of construction and transport machinery were also imported under a
government-sponsored and subsidized programme to develop contractor and
transport capacity. This included some 1,200 heavy-duty vehicles, 1,000 tippers, 500
wheel-loaders, and more than 150 other pieces of machinery. The investment required
a loan facility of ETB 3 billion. Government arranged collateral-free loans, with a
debt ratio of up to 70 per cent. It took a lead in carefully preparing specifications and
making bulk purchases, thereby saving half the investment cost (MWUD 2010). As
this package was designed and implemented with the participation of construction
firms, it helped to boost industry capacity. According to one firm, government
intervention in procurement has increased its annual savings by up to ETB 30
million. Another advantage associated with these programmes was the University
Capacity Building Programme (UCBP), which was fully financed by the government
with technical support from GTZ. Local contractors, consultants, project managers,
and small enterprises were supported through the construction of thirteen universities.
The development of standardized, modular, and economical housing technology also
improved cost efficiency in the construction sector. According to UN HABITAT
(2010: vii), ‘the programme has also built the capacity of the construction sector,
addressed…existing slums and…been a significant generator of employment.’
From a policymaking perspective, a different kind of ‘narrow political margin for
failure’, especially after the 2005 elections, shaped policy design and influenced the
effectiveness of implementation. The ruling party won the 2005 national election but
lost Addis Ababa. In 2007, it was able to win the local election in Addis. These
elections revealed the fragility of the government’s political base and policies in
urban areas (p.128) (Simon 2011; EPRDF 2013c). It was a milestone for the ruling
party, prompting it to come up with comprehensive urban economic, social, and
governance packages (MWUD 2007). Youth frustration at the lack of employment
was critical, making the housing programme and the development of micro/small
enterprises central to urban development programmes. Government interventions to
develop the construction industry and improve cement transportation were both
policy- and politics-induced. These linkages were not fostered sooner, since the direct
effect of the transportation bottleneck was not felt. Transport shortages were a critical
obstacle to the politically important housing programme. This underscored the
linkage dynamics and compelled federal and regional government agencies to act
collectively to improve coordination. Private sector collaboration and participation
was also key to the success of the programme (in construction, building material
manufacturing, and transportation). In a way, this demonstrates a propensity for ‘anti-
fragility’ (Taleb 2012) in Ethiopian policymaking, that is, the ability to resolve
conflicts and policy contradictions, sometimes pre-empting them before a crisis
emerges. This anti-fragility had significant effects in developing the cement industry.

4.4.4 Infrastructure development: energy and transport

Infrastructure development is a key determinant of manufacturing sector growth. In
particular, the large-scale cement industry depends on the growth of infrastructure.
Cement manufacturers unanimously agree on this. The cement industry in turn
contributes to infrastructure growth through added demand for improved services (to
address shortages of power, disruptions, and quality issues); and by manufacturing
industrial goods used in construction. Energy supply and bulk transport services are
the most critical inputs for the cement industry. In fact, the cement industry and
power sector interlock.
The Ethiopian government has invested massively in the development of roads. For
instance, in 2013 the federal government allocated 27 per cent of its budget to roads.
In addition, in 2010 Ethiopia began construction of an extensive electrified railway
network covering more than 5,000 km. This network has eight corridors, including
corridors to Djibouti, Kenya, Sudan, and South Sudan. The project costs more than
$15 billion, with China, India, Turkey, and Brazil (in process) providing part of the
financing.
(p.129)

Figure 4.3. Growth of industrial energy consumption (in Ethiopian Calendar)

Source: EEPCO 2012
In addition to inputs such as coal, heavy fuel oil (HFO), and pet coke, the
uninterrupted supply of electricity is a necessary condition for cement manufacture.
In Ethiopia, electricity generation increased from 370 MW in 1992 to more than
2,179 MW in 2011 (Figure 4.3). With the completion of fourteen ongoing projects,
generation capacity will increase nearly fivefold to 10,000 MW (EEPCO 2012).
Ethiopia’s electricity tariffs are cheaper than those in many African countries,
including Mauritius (by 35 per cent), Kenya (37 per cent), and Uganda (66 per cent).
Ethiopia also has huge hydropower capacity, one of the cheapest and cleanest sources
of energy. This power could be exported to neighbouring countries, contributing to
regional economic integration.

4.4.5 Summary
The observations above on the development of linkages underscore a number of
points on the role of policy. First, incentives seem to have had an effect on
investment, particularly duty-free importation, tax holidays, repatriation of generated
profits in foreign exchange, and provision of land and quarries, etc. It is highly
unlikely investments would have been forthcoming without these incentives. Yet they
were not sufficient on their own. Market demand had also to expand to trigger
investment in (p.130) the cement industry. Government incentives—or ‘intermediate
assets’—were also costly. For instance, significant long-term industrial financing had
to be provided because of the gigantic scale of the new projects in the cement sector.
This drained the government’s limited capital (and foreign exchange) resources,
which were also needed to support other industries. It should also be emphasized that
the use of a public development bank to support industrial policy is rare in SSA,
though such banks are more common in Asian and Latin American countries.
Second, there is an art to adjusting policies in the face of changing circumstances.
Amsden and Chu (2003) emphasize that the role of the state changes as industrial
policies are upgraded to meet changing requirements during each stage of
industrialization. In light of the prevailing domestic oversupply of cement,
government has abandoned investment incentives and even prohibited new FDI in the
industry. The incentives may have outlasted their usefulness through time. However,
prohibiting investment in the industry altogether has been overhasty. Such decisions
might be justifiable if all the relevant information was available. The potential effect
of this rash decision is compounded when we consider that cement exports were not
given serious consideration. Structuralist perspectives highlight the strategic
importance of exports to growth and economic transformation: exporting is the only
‘true component’ that comes from outside the economic system, where demand is
therefore autonomous and can cover required imports, such as capital goods crucial to
growth (Thirlwall 2002). Moreover, the current state of the industry seems to bring
larger issues to the forefront. On one hand, a strong interest group emerged as a result
of the promotion of the cement industry, and the industry’s strategic importance in the
growth of the construction sector and indirect employment creation increased. On the
other, there is a question about the strategic primacy of the sector, given its limited
contribution to foreign exchange earnings and dependence on imported machinery.
Will the power of the interest group or simple policy inertia prevent flexibility or
adjustments in the selection of strategic priorities? It is clear that powerful foreign
industrial interests have moved into Ethiopian cement, and it may be difficult to resist
their control, given international cartels, economies of scale, and technological
advantages. The critical question is whether this matters at this stage. What is the
thinking on this? What are the institutional mechanisms? The government still
has (p.131) Mugher and Messebo to guide the industry, and its role as a major
consumer of cement gives it some advantages in influencing the industry. Yet, the
industry clearly requires new policies and institutions.
This sector shows that in an age of privatization, the Ethiopian government has not
simply held on to some SOEs but actively sought to use them as key tools in
industrial strategy. Governments play a central role by developing innovative
institutions that overcome market barriers and mobilize and concentrate capital
investment. The main role of institutions in latecomer states is as ‘functional
substitutes’ to mobilize resources and to deploy them to capital-intensive projects.
Reflecting this developmental approach is the government’s decision not to privatize
but expand Mugher Cement between 2006 and 2011 with an investment outlay of
$150 million, thereby tripling the company’s capacity to 2.3 million tons and
ensuring it a 20 per cent market share in Ethiopia. Moreover, the Council of Ministers
has agreed to restructure Mugher into a much bigger conglomerate, the Ethiopian
Chemical Corporation, which would produce building materials, fertilizers, and
chemical products. Messebo Cement was EFFORT’s first industrial project and
EFFORT has also expanded Messebo’s capacity to 2.3 million tons. This strategy is
fully in line with government policy, and its investment focused on the northern
region of Tigray. Again, this affirms government’s interest in maintaining its presence
and playing a direct role in the industry.
Moreover, fundamental to the growth of the cement industry was the growth of the
construction industry. The latter, in turn, was shaped by government policy. Current
demand constraints could put the industry in a precarious position. Broader policies
that stimulate domestic demand through the expansion of construction and (more
modest) promotion of exports, are necessary for the sustained growth of the cement
industry. Policies that foster productivity improvements in line with global
benchmarks are also critical. In addition, we are witnessing the growing importance
of large corporations, which are playing a role as national champions in the
development of the industry. The Ethiopian manufacturing sector is dominated by
small and medium enterprises, and historically large corporations have played an
insignificant role. In view of increased economic globalization, economies of scale,
and the ‘global business revolution’ (Nolan 2003, 2012), developing countries
(including Ethiopia) also need larger corporate entities.

(p.132) 4.5 Industrial policy in the cement industry

This section reviews government policies and policy instruments in the industry more
broadly, and focuses on direct instruments. There is no evidence to suggest selective
targeting prior to 2002. Nevertheless, the cement industry was positioned as a priority
in the IDSE of 2002 (FDRE 2002) and the PASDEP and GTP five-year plans.
According to the latter: ‘This result [the gap between actual production of 1.7 million
tons and the PASDEP target of 4.7 million tons] suggests the need to increase the
production and supply capacity of cement in order to meet the needs of the fast
growing construction industry’ (MOFED 2012b).

4.5.1 Investment promotion and incentives

The investment promotion policy in the cement industry involves FDI and domestic
actors, incentives, and access to mining resources. Since the first investment
proclamation of 1992, the sector has been open to foreign investors. However, it was
only after domestic consumption expanded to create excess demand that FDI entered
the sector. The incentives to investors included a tax holiday on profits of up to three
years, and a loss carry-forward benefit. Investors could import equipment and
machinery and up to 15 per cent of spare parts duty-free. Moreover, expatriates were
exempted from income tax for two years. In September 2012, a new law was passed
reserving investment in cement production to domestic investors. It also reduced the
investment incentives, in particular the profit-tax holiday. In general, this move
reflects the diminishing relevance of investment incentives to the cement industry.
Third, government facilitated access to factory land and quarries for limestone,
gypsum, clay, and pumice, among others. Factory land was made available at
giveaway prices (between ETB 10 and 25 per square metre, depending on location)
on long-term lease arrangements. In most cases, the price of land barely exceeded the
cost of compensation to the relocated farmers. The quarries were also available
nearby (mostly within a radius of 25 km), in particular for clay and limestone, which
account for 80 per cent of inputs. The federal Ministry of Mines (MOM) provides
exploration and mining licences to foreign investors, while domestic investors get
them from regional mining agencies. A total of twenty-seven mining licences and
thirty-eight exploration licences were (p.133) given (at a very low rate) during this
period (MOM 2012). Government royalties per cubic metre of limestone, clay,
gypsum, and pumice are ETB 4.29, ETB 4.01, ETB 7.50, and ETB 12.48 respectively
(Mugher 2013). The major input into OPC is limestone and pumice.
Fourth, the government is the sole provider of electricity, and this energy-intensive
industry has enjoyed tariffs that are among the lowest in SSA and other developing
countries. The tariff was $0.043 per kilowatt hour (KWH) in 2008, compared to
$0.02–0.46 in SSA, and $0.05–0.1 in other developing countries (UNCTAD-
UNIDO 2011; Mugher 2013; EEPCO 2012). The recent tariff is $0.039 per KWH, and
Ethiopia entirely relies on renewable energy, in particular hydropower.
Investment promotion has attracted more than 100 investors into the sector, especially
after 2003 (Table 4.10). Only 2 per cent of the investment was registered before 2002,
while the investment flow became significant after the construction boom and cement
shortage became evident. By July 2012, the number of investors involved in actual
implementation and operational phases exceeded twenty. Twelve of these were mini-
cement plants (with less than half a million tons annual production capacity and using
VSK), while the remainder were medium and large cement factories.
The incentives have eased the financial burden on investors and improved their
profitability by reducing investment costs. This was a significant advantage, since
cement factories have to operate at reduced capacity for the first few years of
production. Nonetheless, expansion of the domestic market was the prime reason for
investing in the sector. In addition to political stability, the growth of the domestic
cement market was the most important factor in the increased investment after 2002.
Prior to that date, the domestic market was weak and investment in the industry was
low. By 2012, according to the CEO of Pretoria Portland,
Table 4.10. Summary of investment certificates for cement industry
Period Cement manufacture Cement products manufacture Total

Firms % Firms % Firms %

1992–2002 (ten years) 2 2 40 22 42 11

2003–2012 (ten years) 101 98 239 78 340 89

1992–2012 103 100 279 100 382 100

Source: Own computation from unpublished FIA data, August 2012

(p.134) South Africa’s major cement manufacturer, ‘the country’s current investment
plans, combined with one of the fastest growing cement demands in Africa, makes us
extremely confident about the sustainability and growth of this investment.’1
It is difficult to weigh the value of the substantial resources committed by
government, in a context of extreme macroeconomic constraint and the relatively
high opportunity cost of subsidizing this industry, vis-à-vis the effect of the stimulus.
Cement production would probably have expanded, perhaps at a slower rate, without
such intervention, and with the foreign-dominated ownership common to low-income
countries. The government’s policy represents a classic, centralized use of rent to
encourage the strategic development of a sector, but at the risk of rent dissipation
among mini-plants, eventual oversupply, drain on foreign exchange, etc.
Another issue relates to the GTP, which targeted tenfold growth in five years. This
proved to be grossly unrealistic, and was not founded on a careful market study by
government. In 2012, new and existing cement factories faced market problems, and
they had to operate below one-third of capacity. Government decided to halt new
investment and to suspend or slow projects that were at an early stage. Questions
arise as to whether this was necessary in the absence of full knowledge of the cement
market, and such a decision should be based on a longer term perspective. It is also
noteworthy that implementation of the investment policies was relatively easy, and
required no complex administrative capacity.

4.5.2 Industrial financing of cement industry

Cement requires large investment outlays and long-term financing. The government
has used three, apparently successful, financing instruments to support the industry.
The major mechanism was long-term DBE loans, and when necessary, co-financing
through CBE. In addition, government used industrial development financing for
Mugher Cement, and foreign equity and debt financing for foreign investors.

(p.135) Dbe Investment Financing

The main source of long-term financing was DBE. It provided fifteen-year loans at a
subsidized interest rate (about 5 per cent), and financed up to 70 per cent of the
investment requirement without collateral. In terms of allocation, cement was a top
priority (DBE 2009, 2012a). A quarter of loans for manufacturing were directed to the
cement industry (Table 4.11). Moreover, half of the six largest borrowers and a third
of loans exceeding the single-borrower limit were in cement. DBE has so far financed
Messebo, Derba, National Cement, and Habesha, which have a total capacity of 7.5
million tons. CBE has participated in co-financing, providing working capital and an
international banking service (opening letter of credit). It has also acted as a channel
for Exim Bank of China loans to Mugher Cement. The major constraints were the
difficulty borrowers had in mobilizing their equity satisfactorily, the single-borrower
limit, and the banks’ capacity to mobilize sufficient resources.

Industrial Development Fund And Foreign Equity Financing

The Industrial Development Fund (IDF) was designed to finance SOE expansion, and
to be replenished from the profits of public enterprises. All public enterprises can
retain a maximum of 15 per cent of corporate incomes. Mugher Cement Enterprise
has benefited from this scheme, receiving ETB 700 million. Some foreign-owned
investments have
Table 4.11. Industrial sub-sector loans above 1 per cent of total portfolio (in ETB)
Manufacturing sub- No. of Total portfolio Total portfolio % share from total
sectors loans including comm. excluding comm. loan portfolio with
bal. bal. comm. bal.

Textile industry 32 4,994,845,112 3,618,933,985 29.56

Non-metallic 16 3,816,033,665 3,261,478,706 22.59

manufacture: Cement

Chemicals and 22 668,583,821 530,628,318 3.96

chemical product
industry

Leather and leather 21 465,425,456 412,883,257 2.75

products industry

Food industry 214 458,803,339 431,980,303 2.72

Beverage industry 7 410,370,887 381,208,559 2.43

Total 312 10,814,062,280 8,637,113,128 64.01

Note: Commitment balance means loans approved but not disbursed fully or partly. Such balance is
always there for projects under implementation.
Source: DBE (2012b), loan portfolio concentration report, 31 December 2011
(p.136) accessed foreign loans that occasionally need government clearance or
consent. The International Finance Corporation (IFC), African Development Bank
(AfDB), and European Investment Bank (EIB) participated in financing Derba
Cement. The China-African Development Fund (CADF), Exim Bank of China, and
the South African Industrial Development Cooperation (IDC) were additional
participants.

Foreign Exchange
The major devaluation in June 2010 of the ETB had a negative effect on cement
production. This was due to its coincidence with project implementation, and the
heavy dependence on imported heavy machinery. The cost increase amounted to
some 20–25 per cent. Firms state that it disrupted their cash flows, increased
investment costs, and forced renegotiation of loans. This is a typical dilemma in
industrial policy, where governments choose a specific policy critical to the survival
and growth of one industry, but which has the opposite effect on another.
4.5.3 Mixed results in skills development
Clearly, government has played a central role in the development of this sector as a
major buyer and direct industrial player. It helped kick-start the industry. It helped
stabilize the market-stabilizing role. What has been less remarked upon is the fact
that the cement SOEs helped accelerate growth by generating positive external
economies in terms of expertise, experience, knowledge, and skill. New cement
projects used the expertise of ex-SOE staff during project implementation and to
operate. As the CEO of Pretoria Portland remarked of Habesha: ‘I have been
impressed with the professionalism of Habesha management and their advisors. They
have significant experience in the cement industry and we have already built great
relationships.’ SOEs also assisted by way of training, laboratory testing, etc. for
newly established factories.
However, there have been serious missed opportunities in technological and skills
development. In many developing countries, the cement industry is considered an
intermediate or heavy industry, with large economies of scale and capital intensity,
and requiring professional management. Project design, implementation, and
operation are complex, requiring investment capabilities. A number of countries have
used this industry as a springboard for late industrialization and to develop their
technological (p.137) capabilities. Korea is one such, with Hyundai in particular
reckoned a success. Hyundai not only successfully pioneered the local cement
industry, but also used exports for developing technological capabilities in the 1960s.
Chung Ju-yung, founder and, until his death in 2001, chairman, established the first
cement plants in the early 1960s, before diversifying into the motor industry (late
1960s), heavy industry (including the biggest shipbuilding yard, at Ulsan),
electronics, aerospace, defence, steel, etc. The investment capability developed
during the construction of the cement factories was transferred to construction works
and to building cement projects on a turnkey basis (Amsden 1989; Studwell 2013).
In China, after the late 1980s the growth of the cement industry was integrated into a
strategy to increase asset concentration and economies of scale. More importantly,
China developed the capacity to design cement plants and manufacture equipment,
and enhanced its capacity to build cement factories on a turnkey basis and to
undertake innovative research in the cement industry. The two largest Chinese
technology and machinery suppliers are CNBC and Sinoma International. They have
increased economical kiln capacity to 10,000 tons per day (the biggest size currently).
All the major new cement factories in Ethiopia are being built by Chinese
manufacturers including Derba Cement, Messebo Cement, Mugher Cement, National
Cement, and Dangote. The growing domestic market, coupled with appropriate
policy, enabled China to become the powerhouse of the global cement industry. In
Ethiopia, the government arguably missed an opportunity to use the growing
domestic market to develop the country’s technological capability, although there
were pioneering examples from which government and industry could have drawn
positive lessons (see below).
Development Of Investment Capabilities
One case that demonstrates the role of the government in the development of
technological capabilities is investment capability. According to Amsden (2001),
technological capability requires production capability, investment capability, and
innovation capability. Investment capability involves personal training, pre-
investment feasibility studies, project management, project engineering, procurement,
embodiment in physical capital, and operational start-up. Investment capability is
critical to reducing project costs and time and to meeting project aims. It can also
improve a firm’s future competitiveness and productivity. Such
capacity (p.138) enables future projects to be effectively managed, and enhances the
country’s domestic capacity to implement major projects.
Both Korea and Japan used various policy instruments to develop technological
capabilities, such as local content requirements, guidance in selection of technology,
prohibiting turnkey projects, giving preference to the breaking down of technologies,
equipment, and activities. In Ethiopia, development of technological capabilities was
not pursued. First, there was no local content requirement to encourage local
manufacturing. As the major financier of the cement projects, the government had the
influence to insist on this requirement.
Second, almost all major projects (Derba, Messebo, and Mugher) adopted a turnkey
approach, and were unable to develop their investment capabilities. The incentives
and instruments were not designed to promote learning (technological capabilities)
and learning rents were not made available. National Cement adopted a non-turnkey
approach. The owner established his own project office and hired a foreign consulting
firm. He sub-contracted the civil and electro-mechanical work to a Chinese contractor
and procured equipment from China. He claims to have saved a quarter of the
investment cost by this means. He is also involved in developing a coal mine with
other partners: this is at an early stage. In a different context, Messebo built its first
factory without adopting a turnkey model, and completed construction within twenty-
four months (1998–2000). It also used a local industrial equipment manufacturer to
manufacture some machinery locally, which also reduced costs (for instance,
transportation). A local consulting firm (in a joint venture with an Indian firm) also
supervised the project. During the second-phase expansion of Messebo (2008–11),
which used a turnkey model, project execution took about thirty months, thereby
showing that the turnkey mode does not necessarily guarantee success. Project
promoters of many mini-cement plants also built the factories themselves without
adopting the turnkey approach. Where they did use that approach, they did so
primarily to avoid delays and minimize risks (see Table 4.12). This vividly illustrates
how the lack of appropriate policy and institutions to promote technological
capabilities leads to failure in developing a critical industry. Knowledge acquired and
experiences gained even through isolated initiatives were lost (for example, a local
equipment manufacturer, MIE, which developed some experience in manufacturing
cement-making equipment in 1998–99 did not sustain its first initiative).
(p.139)
Table 4.12. Turnkey model choice
Response Frequency Per cent

To minimize risk 3 23.1

To optimized project investment 2 15.4

To avoid delay 4 30.8

Lack of experience in cement industry 3 23.1

To make easy project management 1 7.7

Total 13 100

Note: Some responses are tallied more than once.

Source: Oqubay (2012)
Third, government failed to guide the selection of technology, and firms did not
benefit from collective action, which would have been more favourable to
government. Because of this, project execution took longer, with increased costs; and
domestic project-execution capacity did not develop. One example of the failure to
guide technology selection is the missed opportunity to introduce coal conversion
technology, which is related to the introduction of coal as a source of fuel. Even the
state-owned Mugher firm failed to incorporate this technology during its expansion.
Contrary to the neoclassical assertion that technology is a freely available shelf item,
in developing or late-industrializing countries, technological supply is plagued by
profound constraints. It typically takes state interventions to circumvent these
constraints. Such interventions include developing technological infrastructure, skill
formation, and promoting in-firm technological capabilities (Lall 2003; Amsden 2001;
Rodrik 2011). Choice of kiln capacity could have been improved with collective
action and state guidance. Similarly, government guidance would have helped mini-
cement plants improve technology imports. Instead, they imported inefficient, poor-
quality equipment, which frequently broke down. Downtime was high and plants
operated at less than a third of capacity. Joint support in technology selection could
also have promoted an association of mini-cement plant investors.
Government’s failure in this respect lies primarily in the lack of long-term
perspective on the cement industry and of a clear institutional and policy framework
for building technological capability. There was no comprehensive plan focused on
the long-range development of the industry, other than the general five-year plan. The
latter was mainly concerned with increasing output, not technological development or
improving the industry’s productivity and competiveness. This also relates to the lack
of (p.140) emphasis on technological guidance. The presence of a strong agency to
lead the sector and of an industrial association might have helped.
4.6 Industrial policymaking: Policy instruments and institutions
It is hoped that the wide-ranging aspects of policymaking raised in preceding sections
provide a comprehensive perspective on the political economy of the Ethiopian
cement industry. Additional issues, in particular institutions and industrial policy,
policy responses to energy use, and industry regulations are examined in this section.
Discussion of these issues may provide additional insights into policymaking.

4.6.1 Institutions and industrial policy in cement industry

In several ways, bureaucratic constraints and coordination difficulties have weakened
policy effectiveness.

Intermediary Institutions
There was and is no association of cement manufacturers. Dialogue between
government and manufacturers was not institutionally supported, and infrequent and
informal consultations were the norm. This is puzzling, given the role of industrial
associations in many countries. Seemingly, however, this did not significantly
constrain the growth of the sector for many reasons. First, the SOE had a monopoly
until 2000, and until 2005 there were only two firms, both closely affiliated with
government. Therefore, they enjoyed direct and indirect access to policymakers and
government agencies. It also seems government tolerated the firms, and collected rent
on their monopolistic position. This policy was driven by the government’s interest in
expanding the sector and minimizing DBE’s risk (cement manufacturers had to repay
loans acquired before 2000 and new loans for expansion). The companies’ profit
margins exceeded 20 per cent throughout the period. The large size and influence of
the major players (private and public) seems to have facilitated their direct access to
policymakers and government agencies after 2005.

(p.141) Coordinating and Support Institutions

Multiple institutions are involved in implementing industrial policy in this sector,
such as DBE, MOI (through FIA and the Chemical Directorate), MWUD, MOM, and
MOFED. In addition, service providers, in particular the Ethiopian Electricity Power
Coroporation (EEPCO), played a key role in the sector’s development. The Ethiopian
Petroleum Corporation (EPC) also played a role as the sole provider of HFO
throughout this period, and in 2012, of coal imports. Moreover, influenced by
experience in the export sector, in 2011 a high-level National Committee for Import
Substitution, Construction, and Employment Creation (NCISEC) was established.
Chaired by the late prime minister, its members included MOI and Ministry of
Construction and Urban Development (MUDC, MWUD’s successor), among others.
The committee took important decisions in relation to the cement industry, such as
making the use of coal compulsory. In contrast to the NECC, it was, however, far
from successful. Apparently, there is insufficient clarity on the body’s role and
agenda. Some of the topics related to export activity fall under NECC, while others
fall under infrastructure development (another committee). The decision was made in
late 2012 to incorporate import-substitution industries (cement and steel,
pharmaceuticals, food processing, and beverages) into the NECC agenda. This will
allow for better coordination and a clearer focus on the manufacturing sector.
The Chemical Directorate of the Ministry of Industry serves this sector as well as
pharmaceutical and other chemical projects, and has a staff of sixteen junior
professionals. This team lacked the skills and knowledge to support the cement
industry. This has mattered in particular in market/demand analysis and forecasts, as
well as in project implementation, including the introduction of coal conversion.
Recently, government, recognizing the institutional gap, has established a new
institute for the building materials, manufacturing, and chemical industries.
Firms indicate that dealing with government offices was a major hindrance, although
with fewer implications than for sectors where smaller firms dominate (Table 4.13).
Despite this, most firms maintain that horizontal coordination among federal agencies
was better than vertical coordination among federal, regional, and local
administrations. The small number of players in the industry, the size of the firms,
and the established contacts with government were probably important. In addition,
cement plants require less service from the public administration (p.142)
Table 4.13. Management time spent in handling government-related issues
Management time (%) Number of firms Per cent

≤15 5 31.25

16–30 6 37.50

>30 5 31.25

Total 16 100.00

Source: Oqubay (2012)

than export-oriented sectors, which require customs clearance, import and export
permits, tax refunds, etc.

4.6.2 Policy response in energy provision and energy efficiency

For the energy-intense cement industry, energy is a strategic issue. A key issue for the
industry is energy utilization and the efficient burning of raw materials. This is the
single biggest cost to cement factories. It also has a significant effect on foreign
currency savings. Thus, the cement industry is dependent on the continuous flow of
energy, which affects the quality and cost of the products. The sector cannot expand
where the energy infrastructure is inadequate or the power supply unreliable. Despite
the government’s commitment to invest in energy infrastructure, the energy supply
continues to be a binding constraint on the cement industry. Globally, cement
factories use coal as the major fuel source. For instance, in the US, coal’s share is 67
per cent, followed by 14 per cent for pet coke and 1 per cent for HFO. Energy costs
amount to only 25–30 per cent in developed countries, but are as high as 50 per cent
in developing countries. Ethiopia’s industrial energy intensity (measured by ton of oil
equivalent per $1,000 of MVA) rose by 65 per cent between 1990 and 2008 (from
1.989 in 1990 to 3.275 in 2008). Vietnam, China, and Pakistan reduced their energy
costs by 17, 65, and 28 per cent respectively in 2008 (Vietnam 0.928, China 0.791,
and Pakistan 0.953).
Energy costs include the cost of electricity and of burning materials. In terms of
electricity, 94 per cent of firms surveyed confirmed they were adversely affected by
power interruptions. About three-quarters of firms stated that power shortages had
significantly affected them, particularly during plant construction. Moreover, 90 per
cent of firms emphasized that the electricity supply had a delaying effect on the
commissioning of cement plants. Lost production days due to power shortages
and (p.143)
Table 4.14. Lost revenue due to power shortages (Mugher Cement Factory)
Year Lost production days Lost production and revenue

In tons In ETB

1998 9 20,160 9,160,704

1999 7 15,680 7,124,992

2000 27 60,480 30,149,280

2001 1 1,859 986,943

2002 8 17,920 10,571,008

2003 13 29,120 17,177,888

2004 1 2,016 1,383,379

2005 0.3 717 492,005

2006 0.3 650 663,910

2007 0.3 605 617,947

2008 45 100,800 156,240,000

2009 119 266,963 517,240,812

2010 42 93,542 190,591,825

2011 97 217,280 638,803,200

Year Lost production days Lost production and revenue

In tons In ETB

2012 5 11,021 23,695,150

Total 375 838,813 1,604,899,043

Note: Lost revenue is based on ex-factory. 2012 involves only nine months.
Source: Mugher Cement Enterprise records (2012). Figures are rounded.
disruptions range from 60 to 400 working days. Table 4.14 shows that revenues lost
by a cement SOE to power shortages were significant.
The National Committee on Import Substitution, Construction, and Employment
Creation decided that cement factories must shift to pet coke and gradually to coal.
The decision focused on the use of imported coal and an ultimate shift to domestic
coal. This policy was driven by foreign currency savings rather than full appreciation
of linkage dynamics. A comprehensive policy package could have led to the
development of a strong coal mining industry. Before this policy decision was made,
a study was conducted by experts from EPC, which was to import coal and pet coke
(replacing the HFO imports), MOM, MOI, and cement firms. This effort by
government to make a study and engage stakeholders was positive, but it also shows
that policymaking was ad hoc. But which agency is responsible for energy provision
and efficiency and can champion them? Most survey respondents agree that energy is
the biggest component of their costs and support coal conversion. However, they are
concerned the change was initiated without sufficient preparation.
Factories that had bought new plants prior to this decision (including the SOE), were
not advised to procure coal conversion technology. This (p.144) lack of guidance
suggests either government shortsightedness or complacency in directing
technological capabilities during procurement and project implementation, especially
as thermal energy is the norm in most cement-producing countries.
This lack of planning and agency could be one reason the coal conversion policy has
faced impediments during implementation. Despite favourable attitudes, only a few
firms readily adopted the technology. Thus, although government imported coal for
cement firms, half of them needed extra finance and time (12–18 months) to equip
their factories with the conversion technology. Mugher’s higher energy costs are
related to older plants and the use of HFO rather than coal. When this survey was
conducted, anthracite was not being imported for mini-cement plants.
In sum, the government’s commitment and plan to develop the electricity sector was
an important strategic response that facilitated growth and expansion in the cement
industry. It is impossible to imagine this growth without development of the energy
sector. In addition, with political commitment and the right policies, it should have
led to successful backward linkages to the local coal industry. It should be noted that
in this case industrial policymaking failed by not inducing new activities and
industries. The initiative to introduce coal use was not taken by the lead agency or
MOI, but by some of the individuals involved. Even then, government did not pick up
the issue, or persistently pursue its implementation.

4.6.3 Coping with market challenges: Shortages and excess capacity

The cement industry has faced recurrent vicissitudes over the last two decades. The
market was more or less stable until 2004. However, both market and government
were caught by surprise when critical shortages of cement occurred in 2005, which
lasted until 2010. Again, both market and government appeared ill-prepared when the
industry suddenly experienced excess supply and productive capacity in 2011. The
policy responses are documented below. Four instruments were used: price control,
product allocation, import licences, and direct imports by government.

(p.145) Price Control

Price control was enforced to stabilize the price of cement. For instance, the SOE had
to get government approval for any price increases during the period 1992–2005.
After 2005, MoTI introduced a price control mechanism for Messebo and Mugher
factories during the period of shortage. More than twenty price increments were
approved between 2005 and 2010. The price control did not focus on the small
cement plants, whose market share was only 10 per cent. Price controls helped to
contain costs on government projects, but did nurture rent-seeking in various forms.
Cement traders exploited the shortages and benefited through speculation. Various
administrative measures were taken, but had limited results. The two cement factories
used every loophole to gain an unfair advantage from these practices. One cement
factory sold in retail outlets at a higher price through its sister company. Cement was
also resold at higher prices, resulting in two parallel markets. Various forms of
leakage were widely reported (MWUD 2009). Price control was lifted in 2011, when
production exceeded demand.

Product Allocation
Allocation of cement products was prioritized: the first priority was the housing
development programme; second were the infrastructure programmes; and third
private sector firms, in particular major manufacturers. Allocation was also necessary
to optimize transport costs: allocations were based on proximity. MWUD, which
coordinated all housing and building projects (2006–10), made allocation decisions.
Allocations gave rise to contention and controversy, as supply and demand never
matched. While this policy helped direct critical input to priority public and private
construction projects, it prompted rent-seeking activities too. For example, some
contractors and businesses that received priority allocations resold the cement at
higher prices (MWUD 2009). This is one of the commonest rent seeking practices,
whereby people collect rent from their permits without paying a penny,2 and is a good
example of the economic behaviour identified in Kornai’s (1980) work on shortage
economies (see also Lindbeck 2007).

(p.146) Import Permits

When demand outstripped domestic production, the government decided to import
cement for the first time. The cement industry had always been protected by an
import ban rather than an import tariff. Government first gave an import licence to a
single corporation, which seemed to have the institutional capacity, finance, and
foreign exchange and had close ties with the ruling party. To the embarrassment of
the government, this initiative ended in fiasco when the corporation failed to import
on time. Then, under a new directive, import licences were issued to those engaged in
trade and construction on a franco-valuta basis.3 Franco-valuta can have an indirect
impact on the foreign currency market. In the face of increased black market activity,
a normal process of opening a letter of credit replaced the franco-valuta arrangement.
Even this stratagem proved ineffective, given the difficulty of controlling the
importers’ source of foreign exchange, and was abandoned.

Direct Imports
The government, persuaded by increased requests for import licences, finally made
the political decision to directly import cement through MWUD. Sufficient foreign
currency was allocated on a priority basis. MWUD set up a logistics project office to
implement this scheme. Later, the project office became an important mechanism for
capacity building among transporters and construction firms. All the government
agencies involved, including MOFED, MoTI, the Ethiopian Revenue and Customs
Authority (ERCA), Ethiopian Standards and Quality Authority (ESQA), Ethiopian
Shipping Lines (ESL), and CBE enhanced their coordination efforts under the
ministry. Contact with the international cement trade following the decision to import
was an eye-opener for government and the domestic construction industry. Sourcing
cement from Pakistan proved the best procurement strategy. For instance, the Karachi
free on board (FOB) price of Pakistani cement was $53 per ton, and landed cost at
Djibouti $80 in 2010. The cement price in Ethiopia had surpassed ETB 3,000 ($200
based on 2010 exchange rates), although this figure had dropped to $100–120 in
2013. This price disparity between imported and domestic cement does raise
questions about the decision to back (p.147) this sector, given the large volume of
scarce resources allocated to it. But the dynamics are more important and complex,
and the price is gradually dropping. For the first time, government was also able to
gain experience in handling and shipping a perishable commodity. The coal-use norm
in Pakistan appears to add to the pressure on government to adopt this technique.
Imports were allocated primarily to government projects, and were sold through the
government wholesale agency, MEWIT. The agency was not known for its high
performance, but played its designated role thanks to close supervision by MWUD.
The transport shortage was solved by importing more than 1,200 heavy trucks, which
were allocated to transport operators under an open bid system.
In 2011, government and cement manufacturers were again taken by surprise by the
excess capacity, accentuated by the fact that all major factories commenced
production at the same time. Government was not prepared to give guidance through
export promotion or by stimulating the domestic market. Its immediate response was
to ban new investments, suspend investment loans, and slow ongoing projects in the
early stages of construction. The decision was an immediate reaction based on
nervousness, rather than on comprehensive understanding of market conditions. This
experience shows that industrial policy in Ethiopia is a work in progress. What has
been unique in Ethiopia is the willingness of government to change course when
things go wrong without having to incur additional transaction costs.

4.7 Conclusions
This chapter has shown, first, that the cement industry has responded to rapid growth
in the construction industry. Despite some mismatches with demand, the Ethiopian
cement industry has now grown such that Ethiopia has become the third largest
cement producer in SSA. This is mainly attributable to public policy (incentives,
credit, housing and infrastructure development, etc.). The tensions/trade-offs are
noteworthy, in particular between foreign exchange constraints and the allocation of
scarce resources to promote an important industry, as well as the mixed results of
policy.
Second, shifts in the industrial structure are marked by an ownership pattern that is
dominated by domestic public and quasi-public enterprises, (p.148) and is different
from the patterns in other SSA countries. Domestic firms still play an important part,
although the role of FDI has increased. This pattern is partly the result of deliberate
policy choice.
Third, the incentives yielded different outcomes, some more effective than others.
Incentives alone are not sufficient, but when the key factor—demand—reached or
exceeded a certain level, the incentives triggered investment in cement to such an
extent that the government had to abandon them. Indirect policies in housing and
infrastructure development played crucial roles in the expansion of domestic demand
and, consequently, the cement industry.
Fourth, there have been no deliberate, comprehensive policies to develop the
technological capabilities of the industry and individual firms. Such efforts as there
were, were fragmented, short-lived, and ineffective. Moreover, the failure of policies
to enhance domestic competition and expose firms to exporting appears to have
contributed to the low competitiveness and productivity of the Ethiopian cement
industry.
Nonetheless, the cement industry has performed as a binding agent, albeit flawed, of
economic development and transformation in multiple ways. The evidence suggests
that the industry has been a partial transformational player in economic development.
This industry has the potential to positively influence the development of heavy
industry. The organizational capabilities of emerging large corporations have the
potential to make them national champions. Government policies were the key
drivers of the transformation of the cement industry, and its expansion was not based
on factor endowments.
The research has clarified how very mixed the cement industry’s experience has
been: it has been hugely important, experienced some notable successes, and
signalled the importance of the state. It has demonstrated synergies among industrial
and other public policies, the interplay between economic and political factors, the
dynamics of policymaking, and the significance of the narrow latitude for poor
performance. While the success of government policies is evident, there have also
been failures, and hence fragility, which have slowed sustained growth. Adaptability
and learning were inherent, once again showing the anti-fragility of Ethiopian
policymaking.

Notes:
(1) <http://www.cemnet.com/News/story/150263/south-africa-s-ppc-invests-in-ethiopian-cement-
firm.html>.

(2) In Amharic, የሲሚንቶ የአየር በአየር ንግድ

(3) Franco-valuta means using one’s own foreign exchange to import cement, as
opposed to the ordinary practice of using official foreign currency.

Beyond Bloom and Bust? Development and

Challenges in Floriculture
Chapter:
(p.149) 5 Beyond Bloom and Bust? Development and Challenges in Floriculture
Source:
Made in Africa
Author(s):

Arkebe Oqubay
Publisher:
Oxford University Press

DOI:10.1093/acprof:oso/9780198739890.003.0005
This chapter explores the striking performance of the floriculture sector, and looks at
output, the sector’s structure, forward and backward linkages, and the content of
policymaking that is credited with the success of the sector. Success in flower
production has attracted international attention and some debate about the main
drivers of the success. This chapter contributes to this debate and, while it celebrates
success, also highlights the new constraints uncovered by that success and the
problems these pose for policymakers. As with other chapters, this one highlights the
role of government intervention, but also the interplay between policymaking and
institutions, and the dynamics of industrial structure and interest groups. The
floriculture sector is a classic demonstration of how unemployed, underutilized local
entrepreneurial potential and natural endowments can be mobilized for economic
development, but do not just bloom automatically under the warm sun of comparative
advantage.
Keywords: Ethiopia, floriculture, government intervention, entrepreneurial
potential, endowments, comparative advantage

5.1 Introduction and overview

Floriculture tells a story similar to that of the cement industry, but adds to it in ways
that help reveal both the achievements and reasons for the unevenness of industrial
policy outcomes in Ethiopia. The economic success of this sector has attracted
international interest and debate (The Economist, 7 February 2008 and 8 April 2009;
Reuters, 2 March 2009; Mano and Aya 2011; Gebreeyesus and Iizuka 2010;
Rodrik 2008b). Floriculture resembles manufacturing more than traditional agriculture
(UNCTAD-UNIDO 2011). It emerged in 2004 and has since shown sustained growth.
Virtually the entire output of the sector in Ethiopia is destined for export markets,
markets characterized by intense international competition. Europe is the major
outlet, with the Netherlands the main hub and destination. Flowers imported into the
European Union (EU) must meet technical requirements, including MPS
(environmental sustainability certification for floriculture) and Global GAP, which
focuses on good agricultural practices. Product variety, quality, consistency, and vase
time are important considerations. Floriculture production is mainly clustered around
Addis Ababa because of better infrastructure and logistics. FDI has played a key role
in this sector, facilitating the spread of technological know-how and penetration into
international markets.
Since 2004, floriculture has generated close to $1billion in export earnings, making
Ethiopia a major global player in the sector (EHDA 2012a; UNCTAD-UNIDO 2011).
Moreover, it has created direct employment for (p.150) about 40,000 people, higher
than the combined employment in the cement and leather sectors. It has also been a
springboard into new exportable goods such as vegetables, fruits, and herbs, thereby
creating more jobs and export revenues (EHDA 2011a, 2011c, 2012a). The government
is working to make more land available for expansion beyond the existing 1,500 ha.
Many agree that Ethiopian floriculture has indeed been successful, although the
explanations for this success diverge. One widely shared explanation has been
‘comparative advantage’, which emphasizes natural endowments as the key
determinant, rather than policies or the role of the state (Lin 2009; Dinh et al. 2012;
Singh 2011). Although endowments are important, this line of reasoning fails to
explain why a strong sector did not emerge earlier in Ethiopia, or in neighbouring
African countries with similar endowments. It also fails to explain why the
Netherlands, a high-income industrial country, continues to be the industry’s leader
despite its lack of cheap labour and natural endowments.
Another explanation associates the success with the ‘discovery process’ and private
sector activism (Rodrik 2004). Although collective learning may have played a major
role, there is no convincing reason to assume that this was the sole or prime factor.
Moreover, the collective learning and activism was mutual, involving state and
private sector agents, rather than unilateral. For instance, the establishment of the
industrial association, which has played an active role in the development of this
sector, was partly initiated and facilitated by government. The government’s
readiness to make bold policy decisions and political commitments to develop this
sector was equally important. Thus, the discovery process theory does not fully
capture the actual development of Ethiopia’s floriculture industry.
Patrimonialism has also been advanced to explain the sector’s growth. In essence,
growth flows from clientelism nurtured by the ruling party/government. This study
shows, however, that not a single firm was owned by government or party officials,
and investment opportunities were open to all interested parties regardless of politics
or networks. Another less prevalent explanation emphasizes the role of the ‘Dutch
trio actors’. This narrative suggests, tacitly, that the success of the sector in Ethiopia
is entirely due to the role played by three external actors, namely Dutch FDI; Dutch
market structures, in particular the flower auction centres; and Dutch development
cooperation (Melese and Helmsing 2010). This may partly be true, but the argument
fails to recognize that these factors are not (p.151) unique to Ethiopia. Implicitly, this
view disregards internal dynamics as the key drivers of industrialization and
exaggerates the role of external forces in African economic development.
Contrary to these arguments, an alternative explanation credits much of the success to
state activism and industrial policy. For instance, UNCTAD and UNIDO, in their
2011 report on African industrial development, state that ‘there are also cases in
Africa where industrial policy has led to success in either developing new export
products or adding value to existing products. For instance, in Ethiopia, state activism
played a critical role in the successful development of the cut flower industry’
(UNCTAD-UNIDO 2011: 63). This view rejects the absolute determinism of factor
endowment and is associated with the structuralist tradition and the political economy
perspective, which emphasize the state’s role in economic development. It also
recognizes that political processes, interests, and constraints determine the choice and
outcome of economic policies.
To examine these arguments and provide a more nuanced explanation, this chapter
presents a detailed analysis of the floriculture industry in Ethiopia and offers a more
comprehensive explanation of the drivers of its growth. The chapter argues that
government policy was critical to nurturing and expanding floriculture. The rise of
this new industry can be better understood by exploring the interplay between
policymaking and institutions, and the dynamics of industrial structure and interest
groups. Analytically, the chapter draws especially on Hirschman’s concept of linkage
effects. As a subsidiary approach, Rodrik’s view of industrial policy as a discovery
process is adopted. The booming floriculture sector is a classic demonstration of how
unemployed labour, underutilized local entrepreneurial potential, and natural
endowments can be mobilized for economic development.
Given natural endowments such as favourable altitude, water supplies, and
temperatures; and fertile soils, cheap labour, and proximity to Europe (relative to
Kenya, Tanzania, and Zambia), comparative advantage has certainly been important.
But static comparative advantage is only one element in the rapid expansion of the
sector. The evidence presented in this chapter suggests it may not even have been the
dominant factor. Rather other factors, notably an activist state applying an industrial
policy to the sector, have been at least equally important.
A number of insights can be derived. First, the government’s commitment to
developing this sector was evident in the concerted use of policy (p.152) instruments
that were exceptionally clear and coherent. This phenomenon is sometimes loosely
referred to as ‘political will’, but the chapter concludes with a better understanding of
this commitment. It also demonstrates how a politically well-designed and
implemented industrial policy can lead to the development of a new industry that
would otherwise have been unlikely to emerge under a laissez-faire regime. Second,
the collective learning of the state and industry was also impressive, showing that the
industry ‘picked the state’ while the state ‘picked the firms’. Learning was not
without problems, but in the floriculture sector, the benefits outweighed the costs.
Third, this process was aided by investors with the necessary technology and
international market connections, thereby easing the learning process. Fourth, this
sector benefited from the narrow latitude for poor performance as defined by the
technological nature of the production process. Export discipline was crucial to
survival as there is very little domestic demand for flowers. This situation served to
catalyse the evolution of performance standards and policy interventions. Fifth, the
policies that helped initiate a successful take-off became inadequate to the challenges
of the industry as it grew, thereby necessitating appropriate new policies.
This chapter’s main theme is the causes and drivers of this growth (the ‘why’), which
also includes the growth and realization of linkage potential, and what this reveals
about policy effectiveness. The study was chiefly based on qualitative research,
whose design was linked to a preliminary quantitative survey of sixty-two of the
sixty-nine firms in the sector. Semi-structured and in-depth qualitative interviews
were then conducted with thirty participants, including firm owners and managers
and heads of government agencies. Site visits were made to flower farms, packaging
factories, and the airfreight terminal. Document reviews were carried out, covering
the Ethiopian Horticulture Development Agency (EHDA), the lead agency of the
sector, and the NECC, the national spearhead in the export sector. Content analysis
and decision tracing were used in these reviews.

5.2 High growth in floriculture

The most important indicators of growth in the floriculture sector are its export
performance, employment patterns, and productivity trends. In (p.153) particular, the
sector’s significance for Ethiopia lies in its easing of balance of payments constraints,
and its potential to create job opportunities for a large, unskilled and semi-skilled
labour force in conditions of widespread unemployment in rural and small towns.
Where data are available, performance is compared to performance in other countries
and other sectors.

5.2.1 Rapid export growth

The floriculture industry continues to be almost exclusively export oriented, and the
export trade determines the survival and growth of the sector and is the key indicator
of the sector’s international competiveness. Export performance is presented in terms
of growth, diversification, and volatility.

Export Growth
Floriculture was almost nonexistent in Ethiopia before 2004. Immediately after its
establishment, exports grew rapidly (Figure 5.1). Cut flower exports increased from
three tons in 2003–04 to more than 50,000 tons in 2011–12, with export earnings
rising from $0.32 million to about $200 million. The average annual growth rate was
an astronomical 400 per cent, which was unmatched in the country’s history. For
instance, the annual average growth of the overall exports sector in the same period
was 22.6 per cent, and earnings from manufactured exports were less than 16 per
cent. The horticultural sector is broadly divided into floriculture

Figure 5.1. Export value of floriculture and horticulture, 1999–2012 (in $ million)
Source: Unpublished data from ERCA Planning Department, October 2012
(p.154)
Table 5.1. Floriculture exports, 2004–12 (value, volume, growth rate)
Year In value ($ millions) Volume (millions kg) Comparators: growth rate of value
(%)

Amount Growth Amount Growth rate All Manufactured Vegetable/fr

rate (%) (%) Exports uit

2004 0.32 700.00 0.03 169.76 21.16 (8.95) 12.14

2005 7.72 2,312.50 2.73 10,265.28 10.41 4.56 93.17

2006 21.97 184.59 6.23 128.94 40.09 41.26 (12.68)

2007 63.61 189.53 13.60 117.72 14.64 (0.60) 26.98

2008 111.76 75.70 22.40 64.71 28.52 23.82 (15.48)

Year In value ($ millions) Volume (millions kg) Comparators: growth rate of value
(%)

Amount Growth Amount Growth rate All Manufactured Vegetable/fr

rate (%) (%) Exports uit

2009 130.70 16.95 29.17 30.20 (1.18) (27.85) (12.81)

2010 170.20 30.22 35.96 23.28 38.03 (8.59) 166.38

2011 175.28 2.98 41.56 15.58 37.40 97.16 (0.72)

2012 196.97 12.37 46.79 12.60 14.77 20.35 (41.91)

Average annual 392 1,203 22.65 15.68 33.21

growth rate (%)

Source: ERCA Planning Department, October 2012

and food horticulture, comprising vegetables, herbs, and fruit. During this period, the
growth of non-floriculture horticultural exports was less than 33 per cent. Within the
horticultural sector, floriculture dominates, with an 83 per cent share of total
horticultural export earnings.
Within a decade, cut flowers became one of the top five export products, accounting
for more than 6 per cent of merchandise export earnings. Between 2004 and 2012, the
industry has earned close to $1 billion (Table 5.1). Sometimes it is argued that its
contribution to net foreign exchange earnings was minimal, as the industry consumes
huge amounts of foreign exchange. This is a valid statement, but account needs to be
taken of the contingent role played by the sector in increased import substitution
through the local production of inputs, and in air transport. The latter exceeds half the
total cost, as flowers are high-value and low-volume products that rely heavily on
efficient airfreight services. With the increased dominance of the state-owned EAL,
and increased import substitution of inputs, the concerns about net foreign exchange
earnings appear to be unfounded. In other words, floriculture has become an
increasingly important contributor to Ethiopia’s tight balance of payments. It is,
therefore, macroeconomically significant.

Export Diversification
Export diversification is an important feature of sustainable growth and structural
change, especially if it is dominated by high-value goods, (p.155) such as
manufactured goods and modern agricultural products. In terms of export
diversification, the share of floriculture in export earnings increased from mere 0.05
per cent in 2004 to more than 6.2 per cent in 2011, making it the fifth largest source
of foreign exchange. The share of vegetables, fruit, and herbs was constant during
this period. Ethiopia’s mono-crop dependence on coffee exports has decreased as a
result, declining from 60 per cent in 1998 to 26 per cent in 2011.
Destination and Price Patterns
Europe absorbs more than 94 per cent of Ethiopia’s floriculture exports, distantly
followed by the Middle East (2.5 per cent of exports), and Asia/US (2 per cent). The
main export destinations are the Netherlands, which accounts for 85 per cent of
floriculture export earnings, and Germany, 5 per cent. The price patterns were robust
compared to the volatile export prices for other primary commodities. For instance,
the price of coffee has always been volatile and decreased by 31 per cent in 2012–13.
In terms of floriculture’s export earnings in 2013, the biggest six players were Sher-
Ethiopia, AQ Roses, Red Fox Ethiopia, Herburg Roses, Ziway Roses, and Linssen
Roses, which combined had a 58 per cent share of export earnings.
In sum, floriculture has shown strong growth and had a significant impact on export
diversification. It should also be noted that the rate of growth slowed in 2011 and
2012. The cause was neither low demand nor competition from other countries. On
the contrary, the global market is favourable, most existing farms are seeking to
expand, and new growers are looking to invest. The key challenges will be examined
later.

5.2.2 Substantial employment generation

Employment generation is another important indicator of policy outcomes. First, the
floriculture industry is labour-intensive and dependent on specific skills. Second,
employment creation is one of the central aims of the industrial development strategy
as well as the floriculture sector. In view of high youth unemployment in urban
centres and high population growth, the job creation agenda has increased in
prominence. Ethiopia is the second most populous country in Africa (85 million in
2012, and population growth of 2.6 per cent), and more than 2 million people are
added annually (CSA 2011).
(p.156)

Figure 5.2. Number of employees in floriculture, 2007–12

Source: DLV, Quantitative Unified Information, and EHDA 2012a
In 2007, employment in floriculture stood at about 25,000, rising to 40,000 in 2011,
an increase of more than 55 per cent (Figure 5.2). According to EHDA, employment
in the flower industry grew to 50,484 in 2012 (EHDA, 2012b). This trend becomes
even more significant when both the direct and indirect employment created is
considered. Floriculture also generated indirect jobs through the associated expansion
of horticulture. Overall, the horticulture sector employed 183,804 people in 2012
(EHDA 2012a), mostly (133,320) in non-floriculture horticulture. In general, the
employment contribution of this sector, which deploys twenty persons per ha. is
impressive compared to other agricultural activities (traditional smallholder farms or
mechanized commercial farms) and the overall manufacturing sector, which currently
employs less than 175,000 people. The leather and leather industry and the cement
industry together employ fewer than 30,000 workers.

Social Dimension
The above employment creation shows the sector’s strong contribution to poverty-
reduction. This is consistent with findings by Cramer, Sender, Johnston, and Oya
(forthcoming) that floriculture has a positive effect on poverty reduction and direct
and indirect employment. It also contributes to social equity by employing
predominantly female workers, who account for more than 75 per cent of the
workforce. More than 95 per cent of employees were between the ages of eighteen
and twenty-nine, and according to forty-one of the responding firms (89 per cent of
respondents), more than 90 per cent of employees were recruited from the
region (p.157) in which the farms are located. Both these factors add to the sector’s
social and political significance.
Most workers (76 per cent) were permanent, only 14 per cent were on contract, and 9
per cent were temporary. On most farms workers are unionized. Above 93 per cent
are production workers with only 7 per cent engaged in administrative and support
activities, indicating the pressure for production efficiency. Most of the workers are
semi-skilled and unskilled, with only 1.5 per cent professional staff (of whom only
one-tenth were technologists) and below 2 per cent technicians. This highlights the
lesser technological depth compared to manufacturing, such as in the cement
industry.
The labour issue has become a serious challenge to the industry, and for government
intervention. Recently, many of the farms (especially those around Addis Ababa)
have faced major problems in retaining experienced workers. Indeed, this is an issue
of wider concern, which has been given frequent attention in the media, including a
series of TV programmes. Meetings on the issue during 2013 were addressed by the
prime minister. Many garment factories around Addis Ababa and employment
agencies are targeting employees in the floriculture industry. The farms have been
easy and attractive targets for recruitment, since the farm workers are judged to have
better skills and a better work ethic.
This continues to be a major problem. For instance, between June 2012 and May
2013, more than 1,350 workers (mostly with three years of experience) left five
firms, mostly for the Middle East. Of these, 74 per cent were from Sher-Ethiopia (in
Ziway and Koka), 8.3 per cent from Redfox Ethiopia, 7.4 per cent from Ethiopia
Cutting Plc., and 10.9 per cent from Lafto Rose, ET Highland, and Dugda. Sher-
Ethiopia lost 1,000 workers in five months (EHDA 2013). On average, the five cut-
flower farms worst affected lost 272 workers per month. Consequently, many farms
were working below capacity, and it usually takes five to six months to train staff to a
minimum level of output. This challenge required a comprehensive response by
government and the industry association, although by mid-2013 they had been unable
to reverse this trend. Thus on 17 April of that year, the former EHPEA chair reported
‘the shortage of women workers continued to be a serious constraint on ET Highland
Flora and all other flower farms’ (Oqubay 2012), a finding confirmed by other
observers.1
(p.158) While it may be true that employees in floriculture are targeted for their skill
and discipline, it also may be that young women are using such employment as a
stepping stone. They seek work on the farms to save some money (to pay
employment agents, transport, and associated costs), get to know Addis and its
periphery, network with experienced migrants, reach out to employment agents, and
finally leave for factories and the Gulf. This is natural in any labour market, and
floriculture employers have to adapt. This leads to a second point. Given high urban
unemployment and huge rural underemployment, this turnover may not be entirely
undesirable, in line with Lewis’s theory of the unlimited supply of excess labour.
However, even though basic employment in the greenhouses is not highly skilled,
there are frictions and costs associated with this turnover, and employers in some
areas (for example, near Bisheftu, east of Addis Ababa) claim that the labour market
is tightening. From one perspective, this illustrates the strong positive externalities of
the floriculture industry. If other employers have a preference for women with flower
sector experience, floriculture employers are effectively subsidizing the development
of a capitalist labour force that is available to other employers. From another, it raises
a policy challenge as the industry matures and faces stiffer competition and higher
labour costs. While government can help to create the right conditions, the onus is on
the industry itself to develop a dynamic comparative advantage if it wishes to remain
competitive. Such a predicament is not unique to floriculture: all industrial sectors go
through a similar process of change, adaptation, and in the worst case, decline. Firms
that innovate survive, those that don’t die out unceremoniously—such is the nature of
capitalist development.

5.2.3 Productivity: Key challenge and work in progress

Coping with international productivity levels is critical for developing countries in
order to exploit their comparative advantage in having a cheap and trainable labour
force (Schwartz 2010). Productivity growth depends upon many factors, such as the
pace and level of learning, as well as on government support in socializing risks. The
major productivity indicators in this sector include labour productivity (stems cut per
labour-hour), yield (stems) per ha., investment per ha., cost per stem, and profitability
per worker.
(p.159) The closest comparator for measuring productivity in Ethiopia is Kenya.
According to earlier data from Global Development Solutions (GDS) in 2006,
marketable stems per ha. were higher and total cost per stem was lower in Kenya.
The selling price of an Ethiopian flower was higher due to the variety (type) and
ecological factors. The discrepancy in productivity was caused by the gap in learning
in the latecomer Ethiopian floriculture industry, and a slower pace of productivity
improvement. Nonetheless, between 2006 and 2012, Ethiopian floriculture showed
increases in productivity (Table 5.2). The marketable stems per ha. increased by more
than 9 per cent, and the farming cost per stem decreased by the same per cent.
Quantitative information from 2012 (DLV 2012) gives the yield per ha. as 2.03
million stems. Many Ethiopian firms visit Kenyan farms, indicating proactive
learning by copying from more experienced establishments. Many firm owners stated
that the gap has been narrowing lately due to learning by doing and the resultant
accumulation of industry experience, which is essential given the intensity of
international competition.
It was also observed that learning by doing in floriculture was relatively quicker than
in other industries. Workers got used to the discipline of work and agricultural
professionals acquired skills rapidly. Tens and probably hundreds of agricultural
graduates (from Hawasa, Jima, and Haremaya universities) have replaced expatriates
from India and Kenya as production managers on many farms. For instance, the
biggest firm (Sher-Ethiopia) and second biggest (Linssen Rose) had no foreign
employees in 2012.
Table 5.2. Benchmarking rose production in Ethiopia and Kenya
Unit of Ethiopia Kenya Ethiopia Kenya
measurement 2006 2006 2012 2012

Production (farm) cost $ 63,334 81,134 63,131 85,000

ha.

Plants/ha. Number 65,000 80,000 75,000 75,000

Yield per ha. (Stems/ha.) Number 1,685,000 2,300,000 1,850,000 2,000,000

Loss rates Per cent 2 5 2.5 2

Marketable stems/ha. Number 1,651,000 2,180,000 1,803,750 1,960,000

Farming cost/stem ¢ 3.84 3.50 3.5 4

Post-harvest, transport, ¢ 11.66 8.60 12.35 13

marketing

Total cost/stem ¢ 15.50 12.10 15.85 17

Sales price/stem ¢ 18.3 14.40 19.2 21

Source: The 2006 data are adopted from Melese and Helmsing (as adopted from GDS
2006). Data represent running costs and exclude set-up costs. The 2012 data are
based on EHPEA’s data on average performing firms in Kenya and Ethiopia.

(p.160) 5.3 Industrial structure of floriculture

The industrial structure and economic and technological characteristics of floriculture
differ significantly from traditional agricultural activities and share many of the
characteristics of the manufacturing sector. Industrial structure refers to technological
and economic characteristics, market structure, and international positioning
(Hirschman 1967; Chandler 2004; Evans 1995). Labour-intensive industries are
typically characterized by fewer and smaller economies of scale and scope. Industrial
structure is an important determinant of sectoral performance in particular contexts
and is a constraint on (but also sometimes facilitator of) government policy. Hence,
understanding industrial structure is a key factor in the design and outcome of
industrial policy.

5.3.1 Firm ownership and corporate structure

There were sixty-nine active firms in this sector in 2012, all privately owned. Foreign
firms accounted for 63 per cent (thirty-nine firms), domestically owned firms for 26
per cent (sixteen firms), and the remaining seven firms were jointly owned. Almost
all foreign-owned firms had prior experience in the industry before coming to
Ethiopia. Of these firms, 32 per cent were from the Netherlands, 17 per cent from
India, and 12 per cent from Israel. The biggest such firm, Sher-Ethiopia, was among
the first to invest in Ethiopia.
Among domestic firms, 71 per cent (fifteen firms) were owned by local investors,
while the remainder were owned by members of the Ethiopian diaspora. This
diaspora is estimated to be more than one million, with the highest concentration in
the US (up to half a million) and the Middle East. Most domestically owned firms
were new entrants, and had different backgrounds (Table 5.3). This finding differs
from the observations in the Enterprise Map of Ethiopia, which generalizes that most
successful business people in different industries had prior trading experience (Sutton
and Kellow 2010). The current study proves that this was not true of most of them.
For example, one successful firm is owned by a member of the Ethiopian diaspora, a
professional educated in the UK, who had worked for several organizations,
including the Bank of England. When he wanted to enter the business, he studied the
industry, secured a soft loan from the state-owned bank (in less than six months),
and (p.161)
Table 5.3. Owner employment background/work experience
Background Number of firms Share in %

Same industry (floriculture) 32 56.1

New entrant 11 19.3

Trade 8 14.0

Manufacturing 5 8.8

Trade and manufacturing 1 1.7

Total 57 100
Background Number of firms Share in %

Source: Oqubay (2012)

received land around Bisheftu. Sometimes the distinction between diaspora and
domestic is not clear-cut: he is also from a four-generation Merkato family (involved
in trading). Another self-made Ethiopian entrepreneur started as a shoe shiner, later
became a shop owner, and currently owns a general trading house. He worked hard,
became literate, and he earned his BA in management through evening classes.
Before he decided to invest in this sector (along with his partner), he too studied the
industry, including visits to foreign-owned firms and Dutch auction centres. He
secured loans in less than half a year and acquired land. He employed a highly paid
farm manager (Indian), and has a regular weekly management meeting at the farm.
Both these firms have performed well. Among those firms that failed, some were
owned by people who were also engaged in the import trade and had shown little
inclination to learn about and manage their flower farms. They regarded flowers as a
supplementary or side business.
Ownership most often took the form of family businesses, and family experience
played a critical role in the industry. Among foreign firms, 42 per cent have
businesses in other countries, with 20 per cent in Holland and 12 per cent in Kenya.
Unlike the cement industry, medium-sized firms dominate in floriculture. Owner-
managers manage close to 73 per cent of farms. More than 55 per cent of managers
have prior experience in the sector. The key decision-makers in 83 per cent of firms
(out of sixty firms) were owner/family members, and only seventeen were managed
by a corporate manager and board of directors (Oqubay 2012). Other studies show
that flower farms in many countries (including the Netherlands) are family owned,
and predominantly small and medium in size (CBI 2002, 2013; Melese and
Helmsing 2010; Nico 1998). Their size allows them the advantage of fast and flexible
decisions, reduced overheads, and a hands-on management style. Although
conceptions of size vary from country to (p.162) country, farms of less than 25 ha. are
considered medium-sized in this book.
The total area in Ethiopia under flower farms increased from 922 ha. in 2008 to 1,500
ha. in 2012. During this period, foreign-owned firms increased their land holdings
from 615 ha. to 1,101 ha., and their share of flower-producing land from 67 to 76 per
cent. Land developed by domestically owned firms increased from 56 ha. to 104 ha.,
accounting for barely 7 per cent. Land held by jointly owned firms has been limited
to 237 ha., with their share declining from 27 per cent to 16.5 per cent in the same
period. As regards farm size, there is no variation between domestic and foreign-
owned firms. Between 2005 and 2011, the domestically owned firms’ share of
exports was very limited and decreased from 25 per cent in 2008 to 13.3 per cent in
2012. In terms of volume, it decreased from 20 per cent to less than 10 per cent in the
same period. Raising domestic firms’ share in production and export, and transfers of
technology, should be an important priority of industrial policy for this sector.
5.3.2 Market and competitive intensity
Understanding market structure and competitiveness is important to comprehending
the underlying pattern in the floriculture sector in Ethiopia. The global floriculture
market in 2009 stood at €26.2 billion (Table 5.4). Europe had the highest share (€11
billion), followed by Asia (€7.6 billion), and North America (about €5.5 billion)
(FloraHolland 2010, 2012, 2014). During the financial crises of 2007–09, market
growth in Europe and the Americas was stable, while Asia showed a modest growth
of 10 per cent (FloraHolland 2010, 2014). According to International Trade Centre
(ITC) and UN Commodity Trade Statistics (COMTRADE), in 2011 the world export
leader in cut flowers was the Netherlands, with 53.7 per cent of the total, worth more
than €3.2 billion. Following far behind were Colombia (€858 million, 13.5 per cent),
Ecuador (€393 million, 7.35 per cent), Kenya (6.5 per cent), and Ethiopia (1.83 per
cent). In 2010, Kenya exported 117,000 tons, worth €0.5 billion (MilcoRikken 2011),
while Ethiopia exported 50,000 tons, generating €146 million. Although the scope for
growth in the flower market seems modest, these figures show that Ethiopia has
considerable opportunity to increase its share of the world market. (p.163)
Table 5.4. Global production in floriculture (in million euros), 2007–09
Year Total Africa South North Asia Middle Value in
production value America America value East value Europe
value value value

2007 24,356 504 1,450 4,059 6,891 220 11,232

2008 24,395 594 1,382 3,998 6,865 220 11,337

2009 26,196 634 1,441 5,450 7,608 220 10,843

Source: FloraHolland, 2010

Major Markets and Destinations

Markets are classified broadly as auction markets and direct markets. There are
auction markets in Amsterdam, Dubai, and Germany. The biggest is FloraHolland,
with 4,000 employees, a €4.35 billion turnover, and 12.4 billion products traded in
2013 (FloraHolland 2014). Close to 80 per cent of Ethiopian cut flowers are sold on
Dutch auction markets. Direct sales are through supermarkets and niche markets.
Other major destinations are Germany, the UK, France, Italy, Belgium, and recently
Russia (EHDA 2012c). The major flower products are roses, cuttings,
carnations, gypsophillia, hypericum, and eryngium. Prices are usually higher during
winter, when European farms are less productive, and around events such as
Valentine’s Day or the New Year. Buyers look for bigger volume transactions and
competitive prices, indicating their increasing bargaining power (CBI 2013). This also
underscores the competitive pressures to increase productivity and operate at lower
costs with thin profit margins.
Air Transport
Most cut flowers have a high value–weight ratio, and are very perishable. Flowers are
consumed throughout the year, and must respond to varying consumer requirements
through time. Thus, the industry depends on air transport and cold-chain logistics
throughout the value chain. Air transport fees account for more than half the total cost
of the product (EHDA 2012b, 2011b), and combined with marketing account for up to
75 per cent of total costs. Frequency and timeliness of flights and other logistical
arrangements are of utmost importance in meeting orders on time and ensuring that
flowers arrive fresh to maximize vase time. The strategic importance and role of EAL
becomes clear from this perspective (see Section 5.4 below).

(p.164) Price-Driven Competition

International trade patterns for flowers are significantly affected by their perishability
and by air transport services. Income levels and the state of the global economy affect
consumption and price patterns. The 2008 recession slowed the growth of the
international trade, and prices declined by 5–10 per cent from 2002 to 2012. Big-
headed roses, especially from Ecuador and the Netherlands, are the priciest, and
small-headed roses the cheapest. Big-headed roses from Kenya and Ethiopia are in
the medium price range, with those from Uganda, Tanzania, and Zimbabwe at the
lower end. Hybridization and breed development are significant factors in improving
prices, productivity, and market share. Dutch breeders are the leaders in the field, and
usually earn up to 2 per cent in royalty fees for a limited number of years.

Competitive Intensity
The major competitors in the European and American rose markets were Colombia,
Ecuador, Kenya, Ethiopia, Israel, Uganda, Tanzania, and Zimbabwe. Major flower-
growing countries in the developed world were downsizing under the pressure of
high labour costs, increased energy costs, and land scarcity. Floriculture has also been
downsized in Zimbabwe due to political factors and lack of price competitiveness.
Uganda and Tanzania are facing major competitive challenges because of higher
production costs. Kenyan floriculture is vibrant and has nearly fifty years of
experience behind it. Kenya’s flower farms cover 3,000 ha., while Ecuador’s covered
6,000 ha. in 2012. By contrast, Ethiopian floriculture is only a decade old and has
grown rapidly to make the country the fifth largest exporter in the world and the
second largest exporter in Africa (Table 5.5). This only came about through an active
industrial policy. However, the
Table 5.5. Exporter rankings in world floriculture (2010)
Country Export volume in ‘000 tons Export value in million €

Netherlands 639 3,151

Columbia 220 858

Country Export volume in ‘000 tons Export value in million €

Kenya 117 500

Ecuador 102 393

Ethiopia 50 146

Source: Rikken (2011)

(p.165) growth of the industry has begun to slow in 2012 and 2013, suggesting the
need for new policies.

5.3.3 Technological characteristics of floriculture

Technological factors shape ownership type, production management, and economies
of scale and scope.

Value Chain in Floriculture

Effective and intensive management of the value chain requires an integrated
approach from farm to final point of purchase. The perishability and limited vase
time of the product, and production uncertainties, create a greater need for such
management by owners. Almost all owners have to be at their farms most of the week
and in constant communication with agents and customers. For instance, the (Dutch)
owner of the biggest firm spends two weeks at his Ziway farm in Ethiopia, and two
weeks in Holland to manage his business there. Moreover, his son is permanently
based at Ziway. The owner of the second biggest firm and his son are always at their
farm near Addis Alem. This reflects the narrow latitude for poor performance in
floriculture.

Economy of Scale and Scope

Floriculture is even more labour-intensive than traditional farms in Ethiopia. Also, in
comparison with smallholder farming or other commercial farms, it requires more
capital, for greenhouses, irrigation, and cool chain infrastructure. An investment of
$300,000 to $500,000 is required per ha. According to the survey undertaken for this
book, forty-five of the sixty-two firms had invested more than ETB 3.5 billion in
total, of which close to 58 per cent was spent on land development, greenhouses, and
buildings, while 17 per cent was for the purchase of machinery, equipment, and
vehicles. In a way, floriculture is also technology-intensive, as it is dependent on
developing new varietals (which requires extensive research and development),
sustained propagation, and a skill-intensive production process. Many farms in
Holland are used for breeding and hybridization, and use sophisticated technology for
maximizing productivity.
Most of the farms in Ethiopia average less than 20 ha. Indeed, 47 per cent of firms
have plots of 11–20 ha., and one-quarter held allotments of less than under 10 ha.
Only 20 per cent of the firms held more than 21 ha. (p.166) Sher-Ethiopia is unique in
owning or leasing 350 ha. of developed land (with greenhouses and irrigation
facilities).

Narrow Latitude for Poor Performance

The product is perishable: according to one grower, sixty days of effort can be lost in
minutes. It requires maximum care throughout the year. The implications of
perishability are clear throughout the value chain from farm to market. Vase time is
affected by the time span of the value chain. Cold-truck transport and coldroom
storage at airports and during flights are crucial. Hygienic and phytosanitary
requirements are strict and may affect trade, placing significant pressure on firms.
Phytosanitary procedures involve the inspection, testing, surveys, and treatment
related to plant quarantine
Timely delivery, high quality, competitive pricing, and product variety are critical to
success. These factors make for a narrow latitude for poor performance, dramatically
narrower than, for example, Ethiopian textile, garment, and leather/leather products.
Firms are under constant pressure, and their narrow latitude for poor performance
also affects the support industry (such as packaging producers), air transporters,
logistics and transit companies, EHDA, and other regulatory bodies.
These features are compounded by the technological uncertainties found in
floriculture. It is these, rather than process technology, capacity utilization, or demand
uncertainty that are the key drivers of a flower firm’s competitiveness. Flower farms,
despite their greenhouse and irrigation technology, are dependent on water supply,
wind patterns, weather changes, etc. The production process is permeated by
significant uncertainty. Intensive management, knowledge, experience, and often
subtle adjustments are called for, meaning that flower farms can seldom be run as a
side business, as some persons seem to have imagined.

5.3.4 Summary
It is clear that the particular features and structure of the floriculture sector have
analytical and policy significance. First, floriculture is highly competitive and
exclusively export-oriented, which has compelled firms to make maximum efforts to
survive and thrive. This has served as a positive pressure to complement government
policies. Second, the management-intensiveness of the business, product
perishability, production uncertainties, (p.167) and intense international competition
have powerful implications for policy. Effectively, they translate a narrow margin for
error in production and distribution into a narrow margin for policy failure. Because
of this, there may be particular scope in this sector for policy learning, with potential
(but not automatic) linkages to policymaking in other sectors. Just as employers in
other sectors appear to prefer workers with floriculture experience, so too is there an
emerging demand in government for officials with knowledge and experience of
developing and adjusting policy interventions in the flower sector.
Third, the industry has been dominated by FDI. The primary role of FDI is
technology, market access, and a ‘demonstration effect’, rather than as a source of
capital. Local entrants agree on the importance of FDI, and working relations are
friendly. A significant number of local firms have so far survived the competition.
This issue will be further discussed later in the chapter.
Fourth, medium-size, family-owned firms dominate the industry. This means that
knowledge and technology is less codified and ‘learning by doing’ plays a critical
role.
Fifth, the nature of the product and markets is such that air transport is a critical
component of competitive advantage in terms of cost, quality, and timely delivery.
This has significant implications for the services provided by air carriers. Finally, the
backward and forward linkage effects to activities and other mechanisms provide
wide opportunities for the development of new industries, products, and processes.
Nevertheless, realizing this potential calls for appropriate industrial policies and
related instruments.

5.4 Linkages and industrial development: Value-chain spin-offs from floriculture

This section discusses the new industries and spin-off enterprises aligned with the
floriculture sector: packaging, air cargo, and new growth corridors. Linkage effects,
even apparently obvious ones, do not always manifest themselves. It therefore makes
sense to think of linkage potential and to pay attention to factors that advance or even
compel such investment decisions. In other words, linkages are not always automatic.
Unlike the cement sector, where the major cost is energy, floriculture’s major costs
are air transport, fertilizer and chemicals, packaging, and labour (Table 5.6). (p.168)
Table 5.6. Cost components of floriculture
Cost in millions (ETB) Share in per cent Source

Airfreight 961 55.2 Mainly domestic (EAL)

Fertilizer & chemicals 311 17.8 Imported

Labour 182.6 10.4 Local

Packaging materials 171 9.8 Domestic manufacture

Total cost 1,748

Source: Oqubay (2012) based on forty firms

These components can be categorized as locally manufactured inputs, imported
inputs, utilities and distribution, and labour and related services. For instance, in 2011
packaging accounted for 98 per cent of locally manufactured inputs and fertilizer and
chemicals for 90 per cent of imported inputs. Airfreight accounted for 92 per cent of
service and utilities costs. This input and output mapping helps us to understand
linkages.

5.4.1 Backward linkages and value-chain spin-off: The packaging industry

This section explores linkages by relying on Hirschman’s concept of linkage effects
(1958, 1981, and 1992). The floriculture sector has been constrained by
underdeveloped domestic inputs and a weak support industry. According to recent
studies, up to 80 per cent of inputs were imported, sharply undermining corporate
competitiveness because of longer delivery times, increased costs, and higher
working capital needs. Moreover, importation involved the loss of hard-earned
foreign currency and of the opportunity to generate employment and build the
country’s productive capacity. Input–output analysis reveals that packaging materials
are one of the three major input costs (35.5 per cent) in the industry. Packaging is a
basic input as it is non-substitutable, and has a major value addition and marketing
effect. It also affects the quality of flowers, space utilization, and airfreight charges.
Packaging materials cost ETB 171 million (out of a total of ETB 482 million for the
three inputs), accounting for up to 10 per cent of overall production costs. Packaging
materials were imported in huge quantity (more than ETB 100 million or $7.5
million) in 2011, and doubling in 2012 (EHDA 2012d, 2012e).
With the right industrial policy, imports can serve to develop new industries in
accordance with ‘the gradual swallowing of manufactured (p.169) imports’
hypothesis, a process ‘in which the growth of imports induces domestic production’
(Hirschman 1958: 110, emphasis added). Hirschman (1958) stresses the creative
function of imports in stimulating new comparative advantage. He argues that
‘traditional theory could hardly be expected to see a connection that could also be
formulated as follows: countries tend to develop a comparative advantage in the
articles they import…We have stressed here the “creative” role imports can play in
the development process, a role that has been almost entirely overlooked’
(Hirschman 1958: 113). This involves value-chain creation, with further expansion of
employment, and strengthening backward and forward linkages. The packaging
industry in Ethiopia demonstrates this notion. With the development of the flower
industry, consumption of packaging articles reached a certain threshold, making their
local manufacture economically feasible. This in turn induced investment in the
packaging industry.
By the end of 2012, there were sixteen firms, with an installed capacity of about
75,000 tons, supplying corrugated boxes for flowers. This process was accelerated by
strong public support. EHDA, jointly with the Ethiopian Conformity Assessment
Enterprise (ECAE) and Ethiopian Standards Agency (ESA), established standards for
these articles. EHDA played an active and leading role in the development of the
packaging industry, further strengthening floriculture’s support industries
(EHDA 2012d, 2012e).
To induce the packaging industry, many policy measures were taken. In short, the
evolution of this direct linkage was policy-dependent. These measures consisted of
both incentives and sanctions to enhance the competitiveness of the flower industry.
First, the investment incentives relevant to floriculture were applied to the packaging
industry to make it competitive with imported items. Factory gate prices were also
negotiated and agreed. Second, standards were set and the producers were given
technical support to meet them. To enforce compliance with quality standards,
factories were shortlisted, thereby accelerating the process. Gradually, the quality of
the domestic manufactured products improved, which encouraged floriculture firms
and other horticulture exporters to use them. This has saved significant foreign
currency, and built up local manufacturing capacity.
Unlike in the floriculture industry, Ethiopians dominated the packaging industry,
accounting for 60 per cent of factories and 75 per cent of installed capacity. The main
challenge is lack of locally available raw (p.170) materials, due to the
underdeveloped state of the pulp and paper industry. It is often argued (for instance
Khan, 2011) that ex ante incentives and intermediate assets are inefficient because
they can be swallowed up and still not perform, but that ex post incentives are much
more effective. Although this may often be true, the packaging experience suggests
that ex ante incentives can be effective. What is important is putting in place
appropriate and easily implementable control mechanisms. Administering ex
ante incentives can, however, be tricky and may fail to induce the necessary action by
economic actors. One example is the duty drawback scheme in the flower and leather
sectors, which became impossible to implement and consequently failed to motivate
firms.

5.4.2 Air cargo and the state-owned airline’s developmental role

Floriculture cannot flourish without a reliable and competitive airline industry. Until
recently in Ethiopia, no such partner has been able to meet floriculture’s requirements
for frequency, cost competitiveness, and service quality. This is because of the losses
airlines can incur as a result of the occasional variations in the volume of flowers.
Many airlines participated occasionally, but were not sufficiently reliable in terms of
frequency and cost. EAL, the state-owned national carrier founded in 1945, has been
the flagship passenger carrier of Ethiopia. Despite intensified international
competition that bankrupted other airlines, EAL has expanded, modernized its fleet,
and upgraded its infrastructure. It trains pilots, technicians, and other staff, and
provides overhaul services for itself and other airlines at its maintenance centre. As
part of its initiative to modernize its cargo fleet, it has acquired Boeing 777s and is
the first African airline to own the Boeing 787 Dreamliner. EAL’s success mirrors
Hirschman’s hypothesis of how the narrow latitude for poor performance can help
developing countries to build a successful airline. The experience of EAL also refutes
conventional neoliberal criticisms of SOEs.

Air Cargo Challenges

Cold chain management on farms, in the form of refrigerated vehicles, and at airport
cold storage facilities was a binding constraint, particularly in the early days of
floriculture. Irregular flights were another major problem. EAL has played a critical
role in reversing this situation, to the extent of operating at a loss. EAL, EHDA, the
Ethiopian Horticulture (p.171)
Table 5.7. Perishable lift capacity on freighter flights, 2009–12 (in kg)
2009 2010 2011 2012 2009–12(average)

Monthly average 2,130,411 2,863,439 2,984,180 4,467,010 2,804,332

Annual total 25,570,411 34,361,265 35,810,164 38,866,096 33,651,982

Annual growth rate +34% +4.2% +8.5% +52%

Source: EAL (2012)

Producers and Exporters Association (EHPEA), and other regulating bodies jointly
paved the way for EAL’s crucial intervention. The state’s motives for playing an
additional role through the EAL correlate with Hirschman’s (1981: 80) finding that ‘it
is also possible, that the state, as a result of having intervened successfully in one
sector of the economy, will acquire the capability and the appetite to tackle advances
for other sectors or for the economy in general’.
The development of the floriculture industry put significant pressure on EAL in many
ways. Historically, cargo transport was a secondary EAL operation. With pressure
mounting to support the floriculture industry, government and EAL resolutely seized
the opportunity to develop an air cargo business. This required long-term,
multibillion dollar investments in aircraft and infrastructure. Thirty-five new aircraft,
including Boeing 777s for flower transport, were ordered in 2009, and additional
aircraft were leased as a temporary solution. Lift capacity for perishables grew by 52
per cent between 2009 and 2012, reaching 40,000 tons a year in 2012 (Table 5.7).

Strategic Response
EAL’s experience in this regard is a classic example of how a government can use a
public enterprise to support industrial policy and industrial development
(Amsden 1989; Chang and Singh 1997). The strategic importance of perishable cargo
has been incorporated into EAL’s strategy as set out in its Vision 2025. EAL’s freight
capacity increased fivefold from 37,000 tons in 2003 to 181,000 tons in 2012, and the
airline plans to increase its annual cargo capacity to 710,000 tons by 2025
(EAL 2012). Fleet size will increase to ten Boeing 777 and seven 757 freighters at a
cost of about $2 billion. A new provisional cold storage facility was built at Bole
international airport’s cargo terminal in 2013 to meet the demands of floriculture
firms and address the cold storage bottleneck. Similar cold storage facilities were
built by government in Mekelle, Bahirdar, and (p.172)
Table 5.8. Government fuel subsidy for floriculture (2008–09)
Period Original price New price Variance 30% variance

April 2008 5.16 8.64 +3.48 1.04

May 2008 5.16 9.66 +4.50 1.35

June 2008 5.16 11.01 +5.85 1.75

July 2008 5.16 11.13 +5.97 1.79

August 2008 5.16 11.05 +5.89 1.77

September 2008 5.16 9.41 +4.25 1.28

October 2008 5.16 8.48 +3.32 1.00

November 2008 5.16 7.84 +2.68 0.80

December 2008 5.16 6.99 +1.83 0.55

January 2009 – 6.27 +1.11 0.33

Note: EAL’s 30 per cent subsidy is not included

Source: MoTI (14 April 2008)
Diredawa to serve newly developing horticulture clusters. By 2020, the capacity of
the Bole cargo terminal will have increased to 1.2 million tons.
A challenging moment for EAL and the floriculture industry was the more than
doubling of fuel prices in 2008–09. Such increases have obviously to be passed on by
airlines to the end user. The challenge for the Ethiopian government and EAL was
how to ensure the survival of the floriculture sector, given that airfreight accounts for
more than 50 per cent of its production costs. At its meeting in January 2008, the
Council of Ministers took the difficult and bold decision to subsidize 30 per cent of
the cost increase (Table 5.8). Another 30 per cent of the increased cost was subsidized
by EAL.
This decision to subsidize a third of the fuel cost increases was difficult politically, in
view of the government’s earlier decision to end subsidies on fuel for transport,
factories, or homes. It also raises concerns that such subsidies could be a bad
precedent. But, if many floriculture firms had gone bankrupt, large lay-offs would
have resulted and the new industry would have been doomed. The experience is clear
evidence of the importance of an industrial policy in Ethiopia, where state activism
and the role of SOEs have been critical in a situation of systemic market
imperfection. It also shows concretely how governments can play an effective
coordination role.
5.4.3 Propagating the model: horticulture and new growth corridors
Linkages can involve not only new spin-off businesses but also new geographic
areas. Hirschman (1981: 76–7) maintains that ‘with the (p.173)
Table 5.9. Performance of non-floriculture, 2008–12
Year Developed land (ha.) Employees Export volume (in 000s ton) Export (in million $)

2008 1,124 – 41,120 18.53

2009 1,665 33,300 39,830 17.41

2010 1,841 36,820 66,410 31.86

2011 5,214 62,570 93,010 40.00

2012 11,110 133,320 123,600 53.15

Source: EHDA Statistical Bulletin Issue 01, October 2012

broader linkage concept, a new activity could also be defined as one that yields the
same product as before but is carried on in a new place [emphasis added]’. The
evidence suggests that while new activities have been induced in the horticulture
sector, regional diversification was limited. Recently, non-floriculture (vegetables,
fruits, herbs) has rapidly expanded, engaging thirty-two firms. In 2012, it generated
more than $53 million in export earnings, and employed 133,000 people on 12,552
ha. of land (Table 5.9).
Close to 90 per cent of non-floriculture farms are concentrated around Addis Ababa
and in central Ethiopia, where the flower farms are clustered. Three factors have
contributed to the growth of this sector. First, floriculture firms have diversified into
other horticulture activities, as is evident from the fact that half of the thirty-two
horticultural firms were originally engaged in flower farming. Second, the linkage
has mainly been generated by the flower sector due to its externalities and spill-over
effects (technological diffusion, management skills, technical staff), as both sectors
require similar expertise. Despite additional incentives provided by regional
governments (for instance, by providing land freely, which has attracted fifteen
firms), only four firms are operating in Amhara and Tigray. This starkly demonstrates
the fact that policy instruments can be less effective in countering economic
agglomeration and is in line with the observations on Ethiopia’s cut flower industry
by Mano and Suzuki (2011).
Third, institutional arrangements have positively induced these linkage effects. For
instance, the lead government agency EHDA, the sectoral association EHPEA, and
the NECC have focused on the whole horticulture sector (not just floriculture). The
presence of linkage dynamics from the floriculture industry seems real. It is also
noteworthy that the non-floriculture sector has a greater employment-creation
potential than the (p.174) flower sector. It also has much larger potential in the
domestic market. Moreover, it appears to have much greater potential linkages with
smallholder farming (for instance, through outgrower schemes). Due to wider latitude
for poor performance, the participation of domestically owned firms is much higher
(43 per cent in contrast to 25 per cent in flowers).
In conclusion, although floriculture may be weaker than leather and leather products
in terms of backward linkages, its linkage dynamics have been exploited and
promoted through more effective policies. In the next chapter, a direct comparative
analysis between floriculture and leather is developed.

5.5 Discovering ‘new’ sources of growth: Rise of floriculture

ADLI (Agricultural development-led industrialization), adopted as the country’s
development strategy in 1995, acknowledged agriculture as the engine of economic
growth. It focused on smallholder farms, labour-intensive activities, and export
promotion. The importance of high-value agricultural products and labour-intensive
industries was also stressed. Nonetheless, it did not become evident that floriculture
was one of the priority sectors until 2002.

5.5.1 Genesis of a new industry

During the Derg regime (1975–91), the Upper Awash State Farm and ET Fruit (both
state-owned commercial farms) were producing flowers on 160 ha. of land, and a few
flowers were exported to Europe, amounting to a few tens of thousands of dollars.
These were summer flowers, and were not based on greenhouse production. In the
late 1990s, a new summer-flower farm was started by Ethio-Flora (owned by an
Ethiopian investor) on a 5 ha. plot in Ziway. In 2000, Meskel Flower (owned by a
member of the diaspora) was the first to inaugurate a rose farm in a greenhouse on 5
ha. at Ziway rented from farmers. Ethio-Dream, ET-Highland, and Golden Rose
followed between 2001 and 2003. These pioneering firms secured no government
support in terms of financing or land. There were no comprehensive guidelines and
the sector encountered many challenges.
In 2004, these five small domestically owned flower farms established EHPEA to
lobby the government to address these issues. This far-sighted (p.175) action, which
drew on Kenyan experience, was the exception in the Ethiopian private sector at the
time. EHPEA managed to persuade government to establish a lead agency. Following
these small local firms, large and foreign firms entered the industry. Thus, it is not
always FDI and large firms that play the pioneering role. Encouraging firms to take
such a role is an important characteristic of industrial policy. To what extent, then, did
government specially reward pioneer firms?
From 2003 until EHDA was established as the lead agency in 2008, MOI (through the
Ethiopian Export Promotion Agency) provided support to this sector. In 2004, NECC
was established to coordinate and lead the overall export sector. It played a vital role
in addressing many of the constraints the firms faced. This demonstrates the state’s
political commitment and the ability of the top political leadership to pick and make
winners. Government opted to ‘pick’ the whole sector, and NECC made it a priority.
Land on state-owned farms in Oromia region, within 200 kms of Addis Ababa, was
made available to the industry.

From Latent to Policy-Led Competitive Advantage

Most flower growers have a background in floriculture in Holland, Kenya, Ecuador,
India, or Israel. According to most of them, their decision to invest in Ethiopia was
based on natural endowments (land, altitude, water, and soil), cheap labour, and
government investment incentives. Affordable land was a key factor in attracting
investors from other flower-growing countries such as Kenya, and was a consequence
of government policy. Land leases in Ethiopia were cheaper than in Kenya. Annual
lease prices for Grade 1 land (in Sebeta, Bisheftu etc.) ranged between ETB 1.23 and
4.01 per square metre, and for Grade 2 land between ETB 1.01 and 3.01. Second,
geographic location or distance to the main European market compared to other
competitors (such as Kenya, Zimbabwe, and Ecuador) was another advantage, in
view of the effect of air transport as a key cost. Third, the availability of cheap and
trainable labour ($1.0 per day in Ethiopia vis-à-vis $2.50 in Kenya) was an important
advantage in a labour-intensive industry. However, these endowments did not
translate into competitive advantage until the right policy and accumulation of
capability were in place. Investment interest materialized when all the necessary
supports (such as investment finance) were provided at the right time and at the
required scale. Natural endowments can be more than compensated for by developing
technological capability, as witness the Dutch (p.176) floriculture industry. The Dutch
are global leaders in floriculture production and international trade, despite the high
cost of labour and energy, harsh winters, and being at sea level. They built the
industry by making it more technology-intensive, supported by the required
clustering, infrastructure, logistics, and trade facilitation.

5.5.2 Inducing change in a new industry: Investment, export promotion, and other
instruments

Investment Promotion
The key investment promotion policies involved attracting FDI and domestic firms to
the sector and related incentives. The latter included a profit tax holiday of up to five
years, and provisions for loss rescheduling. In addition, duty-free privileges on all
capital goods and related spare parts (up to 15 per cent of the value of capital goods)
and construction materials were provided. According to FIA data (2012), 314 projects
in the industry were given investment certificates until 2012. In terms of execution,
only 32 per cent of licensed projects were in operation, 14 per cent were in the
implementation phase, and the remaining 46 per cent were in a pre-implementation
phase. Only twenty-seven investment projects (a mere 9 per cent) were registered
between 1992 and 2001. Investment interest increased after 2002, with peak of
seventy-five projects in 2008 (Table 5.10). This number gradually decreased after
2008, falling below a third of the growth in 2003–08. This highlights the difficulty
government faces in coping with the growth of the sector, in particular in providing
serviced land and finance.
FDI increased due to targeted promotion of Dutch firms, which was supported by a
Dutch government initiative as part of the Dutch development cooperation
programme. Following a successful Dutch trade
Table 5.10. Investment certificates in floriculture, 1992–2011
Period Number Share from total (%) Average firms per year

1992–2011 315 100 15.7

1992–2002 27 8.6 2.45

2003–2011 288 91.4 32

2003–2008 249 79.1 41.5

2009–2011 39 12.4 13

Source: FIA (2012a)

(p.177) mission in 2004, the Dutch government encouraged Dutch firms to invest in
Ethiopia. These firms benefited from Dutch grants covering up to 60 per cent of
initial investment. Stricter environmental and spatial planning regulations, the narrow
scope for domestic expansion, and rising production costs acted as push factors in
Holland (Melese and Helmsing 2010). The investment incentives offered by the
Ethiopian government were also very attractive. In addition to soft loans, serviced
land was available on favourable lease terms, and land stock (under government
control) was readily available.
In sum, the investment incentives were relatively straightforward and did not require
elaborate administration. For instance, three-quarters of the firms surveyed confirmed
that there were no implementation problems with the profit-tax holiday. The
floriculture sector clearly represents a sizeable net benefit to the Ethiopian economy,
in terms of foreign exchange and especially employment, as well as in less
quantifiable ‘soft technology’ transfers and the development of industrial enterprise
capabilities. But rather than unfolding as an inherent comparative advantage, these
effects are the result of sustained and significant government intervention; and, in the
case of Dutch foreign investment, of a double subsidy from both Dutch and Ethiopian
governments. This deviates from Justin Lin’s propositions on comparative advantage
and Rodrik’s industrial policy as a discovery process. Nonetheless, as the following
sub-section shows, not all government support was as straightforward.

Export Promotion and Devaluation

Export promotion policy included several instruments (see Chapter 3). In August
2010, government used exchange rate policy to promote export competitiveness by
sharply devaluing the ETB. Survey results indicate this policy has helped floriculture
firms, which are 100 per cent exporters. These actors appreciate its effectiveness, and
the fact that the measure did not require the cumbersome government procedures
associated with other incentives. Many firms utilized their profits for additional
expansion, and the ratio of imported inputs in the total cost structure is relatively
modest and can be covered from their retention accounts.
Sixty-two per cent of firms surveyed indicated that the voucher system had
significant implementation problems, while an almost identical percentage asserted
that the VAT and duty-drawback have had similar difficulties. More than 55 per cent
of firms believed that export retention (p.178) directives were equally problematic.
However, 68 per cent of firms indicated that the customs branch at Bole airport and
EAL provided satisfactory service. In general, as observed in leather and leather
products, export promotion instruments were not fully effective. The main reason is
that they were ineffectively coordinated and inefficiently executed by the
bureaucracy.

5.5.3 Industrial financing of floriculture

The prime source of long-term investment financing was the state-owned DBE. This
bank provided financing at a subsidized interest rate without any collateral
requirement, and the loan covered 70 per cent of the investment project (DBE 2012a).
The DBE president remarked that ‘in the absence of collateral, the loans to this sector
held significant risk to the bank, considering the perishability of the product’
( Oqubay 2012). Yet, nearly ETB 1.2 billion was given out as loans to almost two-
thirds of floriculture firms. Private banks provided limited loans to some firms,
following in the tracks of DBE. In 2012, more than 84 per cent (some ETB 1 billion)
of DBE’s total loans were good-performing (Table 5.11). The average loan was ETB
29 million, while the maximum was ETB 149 million and the minimum ETB 6
million. Outstanding loans in 2012 were below ETB 200 million, fairly modest
arrears in relation to total loans provided. Only ten floriculture firms faced
foreclosure, their arrears amounting to ETB 120 million. Seven of these firms were
domestically owned, two were foreign-owned, and one was jointly owned. In June
2013, all the bankrupt firms were transferred to third parties. Studies by the bank,
confirmed by firms, show that the causes were poor management, lack of knowledge
of the floriculture sector and skills (selection of
Table 5.11. DBE loan performance in floriculture (in million ETB), 2007–11
Amount

Number of firms 40

Total approved loans 1,167

Average loan size 29

 Amount

Maximum loan size 149

Minimum loan size 6

Good performing loans 981

Outstanding loans 199

Arrears 243

Source: DBE (2012c)

(p.179) varieties, farming methods, marketing), and the failure to use expertise. For
instance, some of the firms overinvested in non-essential facilities, focused on low-
value flowers, and depended solely on auction markets. Additional factors included
disagreements among shareholders or promoters, and poor screening by banks.
In summary, lack of management and experience in the industry, and lack of
decision-making capability and policy execution on the side of DBE were the major
factors behind arrears and foreclosure (DBE 2012c). Nevertheless, the government
used its policy bank to promote the industry and it is unlikely that the industry would
have received such huge loans in the absence of DBE. Empirical evidence from
elsewhere (for instance, Korea and Taiwan) confirms the role development banks
play during catch up (Amsden 2001; Ocampo et al. 2009; Wade 1990, among others).
In this author’s survey (Oqubay 2012), about 83 per cent of firms confirmed that
DBE’s support was satisfactory, the highest rating for any federal institution. There
were also major problems. The first shortcoming was DBE’s lack of knowledge about
the floriculture industry. This affected screening, appraisal, loan decisions, and loan
monitoring. Second, DBE’s policies, standards, and terms were rigidly designed and
inflexibly executed. For instance, one loan proposal was rejected, even though the
project was the most economical (about $250,000 per ha. investment). DBE refused it
because the firm proposed to use wood to build greenhouses and
traditional chicka (mud) to construct storage buildings. In other words, this firm was
‘too cheap’ in comparison with higher priced investment proposals relying on
technologies and materials regarded by industry experts as unnecessarily costly. Such
comparability became a major challenge for the bank, which had to ensure
transparency and consistency of decisions.
A survey commissioned by the Dutch embassy confirms that commercial banks not
only charged higher interest rates but also that loan amounts were inadequate (Royal
Netherlands Embassy 2012). It added that the ‘process for obtaining a loan is long and
very thorough’ and expressed concern about abuse of attractive loans. In May 2012,
one of the successful flower firms complained that DBE would only accept 30:70
equity:loan ratios (even when proposals contained ratios more favourable to it). The
firm’s owner also complained that DBE refused to accept early loan repayment. Some
investors, who had used their own capital in order to speed-up the project, were
rejected when they requested (p.180) loans from DBE, as the bank’s rigid procedures
would not allow this. One of the firms whose proposal was dismissed had a
successful track record in the industry in Ecuador. There have been renewed efforts
by DBE to address these shortcomings.

Application of Reciprocal Control Mechanism

A key dimension of industrial policies is the principle of reciprocity, though this is
really a euphemism for the state’s disciplining of capitalist enterprises that it supports
with incentives. This issue has important implications for improving standards and
the use of rents for productive ends (Amsden 1989; Khan and Jomo 2000). For
instance, in land provision, if a firm did not start operations on time or if it misused
the land, the latter was taken back and given to other investors. The risks and
leakages associated with an inadequate ‘reciprocal control mechanism’ appear to be
not too significant. This is because of a combination of positive factors such as the
high competitiveness of the export sector. Moreover, given the perishability of the
product, established export channels, and limited domestic demand, leakage was not
as serious a concern as it is for export commodities such as coffee and chat.
Nonetheless, although DBE had loan access criteria and a monitoring scheme focused
on timely loan repayment, these were not enough to shape the behaviour of actors.
The incentives were not sufficiently linked to performance or the reciprocity
principle, although the government coped over time. In the early stages, there were
instances of over-invoicing for equipment purchases, which DBE, new to the
industry, was in no position to control. NBE has been exercising strict control over
the repatriation of export earnings, and recently the DBE has been nominated as the
focal agency to administer the firms’ export transactions. Due to the sensitivity and
high-risk nature of these loans, DBE took swift measures. It commissioned a study of
all farms, classified the firms into three groups, and took exemplary steps by taking
over two foreign and two domestic farms. The categories included firms with
management problems, ethical problems, and firms which have a prospect of
recovery. Their purpose was to influence the firms’ behaviour.2 It also rescheduled
the loans for most firms affected by the economic recession in 2007–08.
Table 5.12. Land development and utilization
Year FDI Domestic Joint Total

2007/08 Ha. 615.6 56.5 249.9 922.0

Share (%) 66.8 6.1 27.1 100.0

2008/09 Ha. 840.2 62.2 337.5 1,240.0

Share (%) 67.8 5.0 27.2 100.0

Ha. 886.0 72.6 347.4 1,306.0

Year FDI Domestic Joint Total

Share (%) 67.8 5.6 26.6 100.0

2010/11 Ha. 963.0 86.0 251.0 1,300.0

Share (%) 74.1 6.6 19.3 100.0

2011/12 Ha. 1,100.9 103.9 237.6 1,442.40

Share (%) 76.3 7.2 16.5 100.0

Source: EHDA Report to NECC (2012b)

(p.181) 5.5.4 Provision of affordable land, infrastructure, and logistics

Land is publically owned in Ethiopia, and the government provided land to investors
on long-term leases and at low prices within 160 km of Bole airport terminal. Close
to 1,500 ha. was being used for floriculture by 2012 (see Table 5.12). The cost of land
was below $10 per square metre in 2008 in Ethiopia, compared to $30–40 in Kenya.
According to one company’s annual report, land comprised about 1 per cent of total
costs in 2011. Initially, firms were given land that was formerly part of state farms.
Later, allocations became more difficult as land had to be provided from farmed land
or by regional administrations, often by means of complicated procedures. Two-thirds
of the firms surveyed reported problems in acquiring land. Delays in handing over
land and escalating lease prices became major obstacles to the expansion of the
sector. Moreover, land provision was in some regions hampered by weak political
commitment and rent seeking at different levels of the administration. Surveys by the
Dutch embassy in 2010, 2011, and 2012 show that corruption has increased,
particularly in lower level government offices and local administrations. More
recently, efforts were being made to establish a land bank under the EHDA. Lease
terms are favourable as firms can pay on an annual basis over twenty-five to thirty
years (EHDA 2012b; EHPEA 2007).
Most of the farms are close to urban centres, so the infrastructure constraint has not
been binding. However, disruptions to electricity supplies were a major complaint,
especially as greenhouse temperatures have to be controlled. Most firms also
complain that the provision of other infrastructure had implementation problems.
Almost all of them (85 per cent) stated that this was especially true of logistics
services. Moreover, (p.182) two-thirds of firms believed that the services provided by
ESL and the Ethiopian Electric Power Corporation (EEPCO) were poor. Clearly,
there is significant room for improvement.
5.5.5 Summary
Above all, a growth perspective, appropriate policies, the political commitment to
execute, and learning capability appear to be the key factors in exploiting
comparative advantage to develop a competitive new industry. However, the key
findings are threefold. First, the outcomes were a product of endowments,
institutions, enterprise, experiment, and learning: they were produced by policy, but
also by politics. Industrial policy involved multiple instruments, which were coherent
and compatible with the country’s broader development strategy. This policy and
related instruments were constantly improved, and created a conducive and
supportive environment for the industry. These policy instruments were also assisted
by ‘push factors’ (for instance, Dutch government policy to encourage Dutch
companies to invest in Ethiopia and support them).
Second, implementation of policy was often inadequate as a result of lack of capacity
and coordination gaps among government agencies. In addition, a lack of long-term
vision and its translation into plans has become an impediment to sustained growth of
the industry. Actors in the sector appeared complacent about the achievements to date
and future prospects.
Third, adequate reciprocal control mechanisms were not put in place. This deficiency
did not lead to unintended results in terms of performance or rent seeking, as it was
offset by the narrow latitude for poor performance in a dynamic market. The role of
credit policy and public development banks in industrial development needs to be
underlined.
Fourth, mistakes were made (for instance, leading to some bankruptcies) in policy
execution. However, the evidence suggests there have been net positive gains and that
any sector-wide cost–benefit analysis needs to take account of dynamic and often
unpredictable features rather than applying a rigid, narrow, and static perspective. It
should be emphasized that the benefits of the development of a competitive flower
industry outweigh the costs, since overall the policy helped generate new industries
with multiple long-term benefits.
(p.183) Finally, this story shows that even with such costs, active industrial policies
are significantly better than laissez-faire policies, which had hitherto failed to induce
the development of the industry. Government has gradually learned lessons and
shown the ability to cope (as in the management of bankrupt firms) in managing rents
and disciplining industrialists.

5.6 Policymaking and policy organizations

The research shows that the outcomes of industrial policies depend not only on policy
content, but also on the complementarity and coherence of industrial policy
instruments. More importantly, the outcome depends on effective institutions,
coordination, and collective learning by government and economic actors.
5.6.1 Specialized institutional support to a new industry
Chapter 3 sketched the overall coordinating institutions, lead agencies, and
intermediary institutions. The establishment of EHDA as lead agency to develop the
horticulture sector was an important innovation. The agency’s aims are ‘to ensure the
fast and sustainable growth of horticulture production and productivity; to facilitate
the export of diversified horticulture products that meet international food safety
standards; and to coordinate the development of supporting services’ (FDRE 2008b:
2–4). In discharging these multiple duties and responsibilities, the agency is expected
to collaborate and coordinate its activities with EHPEA, the industrial association for
floriculture. The agency became fully functional within six months, and a director
general was appointed. The organizational structure was designed around the core
activities of sector development, technological capabilities, and market development.
EHDA was also made a member of NECC.
Working visits and study tours by government officials and industry actors were
made to Kenya, Ecuador, and the Netherlands. Benchmarks, codes of practice, and
standards have been frequently referred to and applied with the close collaboration of
the Dutch embassy. This suggests policy learning and evident improvements in policy
capabilities. Such learning opportunities, including learning by copying or from
models, are known to be important in catch up by developing countries
(Amsden 1989, (p.184) 2001). Second, open dialogue between government and the
industrial association was also effective, although it was not supported by a
systematic institutional structure, such as Japan’s deliberation forums (see
Johnson 1982).
More than 70 per cent of the firms surveyed are convinced that EHDA works closely
with economic actors in the sector. Nevertheless, only 53 per cent believe that EHDA
has played a significant role in developing the sector, hinting at higher expectations
of EHDA, or at least highlighting the difficulty for those involved of neatly
distinguishing the factors that influence outcomes. Moreover, 57 per cent of firms
believe EHDA focuses more on regulating than on supporting firms. In addition,
about a quarter of firms state that EHDA lacks appropriate knowledge of the industry.
In sum, EHDA needs to improve its institutional capacity and shift towards providing
effective support to floriculture, especially in the industry’s next stage.

5.6.2 Effective institutional coordination and policy coherence

The key coordinating institution in the overall export sector is the NECC. Among
banks, DBE is the key institution for industrial financing, with EAL its counterpart in
providing air transportation and air cargo facilities. NBE and Ethiopian Revenue and
Customs Authority (ERCA) are the regulatory bodies for export activities and the
repatriation of foreign currency. A critical intermediary is the EHPEA.

NECC
NECC made floriculture one of its top priorities from 2004 to 2011. Its coordinating
role was impressive, particularly before EHDA’s establishment. Once EHDA was in
place, it became part of NECC, providing a platform for identifying and discussing
constraints and securing government agency support. Since 2004, NECC has
discussed 365 issues during eighty meetings. Among the four categories of issues, the
least discussed are linkage effects (18 per cent), followed by logistics and market
issues (23 per cent), and firms and capacity building (27 per cent), while the incentive
package was the most discussed issue (32 per cent). Among the single issues, the
most discussed are investment and export incentives, land and infrastructure,
industrial financing, capacity building, and the cold chain logistics system. Although
this breakdown hints at priorities (p.185) and the comprehensiveness of the issues, it
does not give relative weighting to the topics.

Persistent Coordination Drawbacks

Implementation of industrial policies is usually dependent on the active support and
facilitation of multiple agencies, and the development of an appropriate institutional
web for the coordination of efforts was an expensive process. Organizations involved
include government agencies and intermediary institutions, and there is also public–
private dialogue. Coordination remains a major failure, along with (and because of)
bureaucratic inertia and low government agency capacity. The floriculture industry is
affected by federal, regional, and local levels of government. Forty-seven per cent of
floriculture firms reported loose coordination among federal institutions, while 54 per
cent noted that coordination among federal, regional, and local administrations was
even worse. However, 60 per cent rated the support of regional states favourably.
These firms strongly recommended that coordination obstacles among government
institutions be resolved. Close to two-thirds of the firms were satisfied there is
dialogue between government and economic actors, but recommended a more regular
and institutionalized dialogue forum and emphasized the need for greater consultation
and communication. Moreover, they complained of the frequent revisions to
government directives without sufficient consultation with industry. They also
stressed the need for one-stop public service and enhancing government officers’
knowledge of the industry. This points to the need for a stronger ‘business–
government’ dialogue.
Most firms complained that bureaucratic obstacles were so great that management
teams and owners spent significant time in handling government-related affairs: a
quarter spend more than 30 per cent of their time on this, and an additional 41 per
cent spent 16–30 per cent. In short, coordination remains incomplete and stands out
as a binding constraint. The strength of these views reflects the size and family
ownership of the firms and the nature of the industry: the firms cannot afford to spend
time on bureaucratic issues.

Intermediary Institutions and Industry–Government Dialogue

Intermediary institutions play a key role in all industries, not least this one. The
Netherlands, Ecuador, and Kenya have well-developed (p.186) associations that
actively lobby for the industry and market and technological development. Kenya, for
instance, has a developed horticulture industry of about 300 firms. The Horticulture
Crops Development Authority, established in the 1970s, has a headquarters and
twenty-four branches. The Ethiopian counterpart is EHDA, which has no branch
offices. Kenya’s Product Health Inspection Service (KPHIS) provides a high standard
of phytosanitary service with laboratories at Nairobi airport. Kenya also has an
industrial association (Fresh Produce Exporters Association of Kenya [FPEAK]) for
the sector, established in the 1970s and comprising 121 members, of whom forty-nine
are in floriculture. In the Netherlands, floriculture firms have a strong association as
well as world-renowned auction facilities. The Ecuadorian industry association also
plays an active role in market development.
Survey results indicate that nearly 91 per cent of the firms in the sector belong to
EHPEA, a much higher percentage than for other Ethiopian industrial associations.
As regards EHPEA, 90 per cent of firms think it plays an important role in
knowledge transfer and training, and almost three-quarters agree that it lobbies
government and promotes investment. This is positive, especially in relation to the
other industrial associations in Ethiopia. About 80 per cent of firms agree that the
Dutch embassy plays a positive role in sector development.
As regards unions, almost all workers are unionized. Unions negotiate with
employers about salary and working conditions, although they are weakened by the
constant influx of labour. Labour–employer relationships differ from firm to firm.

5.6.3 Policymaking
Three government directives in 2011–12 are selected to demonstrate the
policymaking process and divergent political views and interests. These are NBE’s
new export directive, the new cargo directive, and new packaging standards.
Together, these examples highlight the complexity and tensions underlying simplistic
ideas regarding ‘political will’. Unless policymaking is transparent and participatory,
mutual suspicion between government and private sector will be reinforced. The
‘political will’ to act weakens where transparency and participation in decision-
making threatens the interests of powerful groups benefiting from the status-quo.
Lack of ‘political will’ and state capture are two sides of the same coin.

(p.187) New Export Directive

NBE and EHDA prepared a new NBE export directive in 2012. Firms were consulted
and the draft discussed, and the directive was implemented within a short time of
preparation. The unit of measurement in floriculture was changed from stem to
weight. Moreover, the amount to be repatriated was to be based on average auction
prices. During the survey, 69 per cent of firms believed the new directive was
unnecessary and the consultation was unsatisfactory. In addition, 71 per cent insisted
that the preparation time was too short. On the effect of the policy on firms, 59 per
cent indicated it was significantly negative. Among firms listed as delinquent, more
than half stated they were wrongly included on the list due to errors and lack of
coordination among government offices (DBE, NBE, ERCA). About 16 per cent
blamed their inclusion on delays caused between buyers’ and corresponding banks.
Differences in the interpretation of the directive (whether related to free on board
[FOB] or cost, insurance, and freight [CIF]) persisted until mid-2013. This problem
was compounded by the over-restrictive implementation of export directives for
many years. The current NBE export directive applies rigid controls to foreign
currency repatriation rather than a more flexible, balanced approach that promotes
exports and repatriation (NBE 2012a). Firms suggest that the current monthly export
permits and the acceptable margins ought to be reviewed. NBE claims that exporters
do not repatriate foreign currency earnings on time or indeed in full. For instance, on
1 June 2012, NBE instructed banks not to issue export permits for the month to
ninety-nine flower growers. However, non-compliance with the existing export
directive does not necessarily arise from deceitful practices by exporters. As
experience elsewhere shows, intense balance of payments constraints can on occasion
cause government to be overly strict with exporters, to the extent that ‘the goose that
lays the golden egg’ is killed.

New Packaging Standards

EHDA and related government agencies worked with the industry to develop a local
packaging capacity and to improve quality. Almost all firms agreed this project was
necessary, but insisted that consultation was inadequate and preparation time too
short. Due to instructions to buy from the limited number of factories that complied
with the specified standard, 77 per cent of firms mentioned they incurred higher
costs. The (p.188) majority (54 per cent) acknowledge this requirement helped reduce
damage to perishable goods. The firms also recommended that more manufacturers
become involved in packaging to reduce prices, and insisted that obligatory
requirements be lifted.

New Cargo Directive

A new directive on air transport was introduced in 2012. Intermediary logistics firms
were dropped and a direct contract between EAL and each firm was inaugurated.
Two-thirds of firms supported the new directive, but again complained about the
insufficiency of consultation and preparation time. Firms demanded more transport
options, improved handling at the airport, and reduced freight costs. This book was
written shortly after the introduction of this directive, but additional feedback (after
the new provisional cold storage became operational) showed increased satisfaction
with EAL service.

5.6.4 Emerging challenges

Overreliance on Fdi
FDI’s role in floriculture has been positive not only in terms of export earnings,
employment, and linkages, but also in terms of the sector’s very development. Even
owners of domestic firms acknowledge this. Survey results indicated that more than
two-thirds of domestic firms regard FDI as positive in both know-how transfer (87.5
per cent) and market access (69 per cent) (Table 5.13).
Externalities and spill-over effects have been positive, as domestic firms found the
opportunity to learn from farms in their vicinity, or from Dutch auction centres and
Kenyan farms. Such learning opportunities were in
Table 5.13. Domestically owned firms’ views on FDI
Responses Technology and know-how Market access On FDI
transfer development contribution

No. of firms % No. of Firms % No. of Firms %

Important 11 68.75 7 43.7 12 75

Moderate 3 18.75 4 25 4 25

Little 2 12.5 5 31.2 – –

Total 16 100 16 100 16 100

Source: Oqubay (2012). Response rate is 90 per cent

(p.189) most cases initially facilitated by foreign-owned firms. Such firms facilitated
introductions to Dutch breeders, and to logistics suppliers, and foreign marketing
infrastructure. The clustering of farms eased and encouraged learning by doing and
copying. It also allowed the ample supply of skilled human resources (farm
supervisors, technicians, and semi-skilled workers). Recently, most foreign-owned
firms use only Ethiopian staff, including as farm managers. This FDI dominance
raises eyebrows, however, as it makes the sector reliant on foreign actors, with
associated long-term risks.
Heavy dependence on foreign investment, limited strategic negotiation with foreign
investors, and limited direct linkages between foreign and domestic firms highlight
some of the limitations of industrial policy in this sector. It also shows how
government has been quite timid in its efforts to maximize domestic firm governance,
technological know-how, and other benefits in the sector. First, there have been
repeated complaints that foreign-owned firms are making a lesser financial
contribution to the country by over-invoicing overseas costs and declaring lower
selling prices (through subsidiaries or sister companies). This arguably highlights the
leakage of foreign exchange from the country. Second, Dutch breeders usually sell
their new and high earning breeds primarily to the foreign-owned firms. Third, some
of the foreign firms (due to their dominant size, logistical advantages, market access,
economies of scale) secure excessive rents by exploiting their local monopoly. In
addition, there is ample evidence that foreign-owned firms were less enthusiastic
about government interventions to address market inefficiencies (for instance, in
packaging or airfreight). There are also associated risks of a footloose industry, which
in one case appear to be real. One (anonymous) firm told the author it had decided to
expand its farm in Kenya in 2012 rather than in Ethiopia. Thus, while the role of
foreign-owned firms was largely positive, it poses a fundamental challenge that
requires an effective policy response for the promotion of domestic actors.

Developing Strong Domestically Owned Firms

The issue of developing the domestic private sector was raised at the second meeting
of the NECC in February 2004, but was given no prominence in the Growth and
Transformation Plan, 2010–15. Such domestic firms as have survived, have so far
shown good, if uneven, progress (p.190)
Table 5.14. Benchmarks of International Capacity Building Programme
International benchmark Very good Good Poor

Productivity (stem) 8 kg per m2 0 4 14

Waste in bush management ≤5% 7 7 4

Fertilizer optimality 100% 3 4 11

Cool chain management 2–3° C 9 9 0

Cool chain management 15° C 11 7 0

Human resource utilization 10 per ha 0 2 16

Source: Own computation based on EHDA data (2012b)

(Table 5.14). This positive outcome was attributed to the joint efforts of both
government and industry.
This author’s survey (Oqubay 2012) results show that there has been progress in cool
chain management and bush management, while productivity (the highest was only
6.3 kg/square metre, with fifteen persons per ha.) lags. The availability of horticulture
professionals has improved over time, with many of them replacing expatriates from
Kenya and India. Research and development into seed varieties or farm practices, and
the expansion of propagation facilities, appear to be at rudimentary levels, and an
appropriate policy response appears to be lacking. Various standards and
certifications have been introduced in the industry and most firms are participating
and increasingly complying. This project has benefited from the support of the Dutch
embassy, which has sponsored the use of experts. Such standards include MPS-A,
MPS-SQ, Global Gap Certification, Codes of Good Practice, and Integrated Pesticide
Management (IPM). However, there are differences in the levels firms have been able
to achieve. For instance, fourteen firms were gaining bronze MPS certification, while
three had already achieved silver certification.
Another major concern is the decreasing role of domestically owned firms. Due to the
potential benefits, their share should be expanded, and supports should enable this.
Reversing the current trend requires a strategic response. In addition, domestic firms
still use intermediaries rather than dealing with their final customers and focus on
low-value varietals. The key challenges of involving new and the right type of
domestic entrant and of speeding up learning and catching up remain. Moreover, the
size of the sector could be increased to strengthen its international competitiveness
and fully exploit its potential for employment generation.

(p.191) Emerging Challenges in the Growth Stage

The floriculture sector has been at an incipient stage until recently, facing problems
specific to that stage. For instance, land, industrial finance, air transport by EAL, and
incentive packages played critical roles during this stage. Although these instruments
may continue to be important during the growth stage, they may not be sufficient.
The growth stage has its own unique constraints and challenges and they demand
appropriate policy. For instance, technological and marketing upgrading could be
critical. Productivity levels in Ethiopia are far behind industry benchmarks, and need
to catch up with Kenya and Ecuador, and also the Netherlands. Maximizing the
potential of existing and new linkages is critical. This requires policy adjustments
focused on new strategic issues, and developing more appropriate institutions.
It should be noted here that there are two distinct aspects: the expansion of new
projects and of existing firms. So far, only 1,500 ha. have been developed. For firms
to expand, critical support in terms of land, credit, and cargo transport should be
provided. Second, different policies and institutions are required to address the
evolving role of domestic firms, technology, skills, productivity, market
diversification, etc.

5.7 Conclusion
The Ethiopian floriculture sector has shown rapid development in terms of foreign
exchange earnings, export diversification, employment generation, and linkage
effects. A range of policies and institutions were applied, and the evidence suggests
they were largely effective. Implementation has been better than in other sectors, and
support institutions have been relatively effective. Despite this progress, the full
potential of the sector has not been realized because of implementation constraints
and emerging challenges.
Second, the industry’s narrow latitude for poor performance helped shape the
behaviour of interest groups in the sector and complemented policy interventions.
Linkage effects were multifaceted, including backward linkages (packaging
products), forward linkages (air transport and cold chain logistics), horizontal growth
of horticulture, development of new corridors, and other spill-over effects. Ethiopian
floriculture has (p.192) indeed fostered linkages and stimulated economic
development through various mechanisms.
Third, the industry was first ‘pioneered’ by small domestic firms. Other firms,
particularly foreign firms, followed suit. This is in sharp contrast to the widely held
view that FDI and large firms play the pioneering role.
Fourth, FDI nonetheless played a key role by contributing to technological
development and market access. Foreign-owned firms are predominantly medium-
sized, family-owned enterprises with prior experience in the sector. Despite continued
government support, the domestically owned firms’ share has been limited.
Ethiopians are increasingly filling farm management and other technical positions.
This calls for further indigenizing the sector in terms of ownership and human
resources.
Fifth, the policymaking process followed a distinctive path (as it is not widely applied
in other industries in Ethiopia) of collective learning and discovery. The story shows
that not only has government picked and made winners, but the ‘winners’ have also
picked the government. The government, through multiple agencies such as the
NECC, EHDA, DBE, and EAL, has played key and exceptional roles in developing
the sector. Intermediary institutions, in particular Ethiopian Horticulture Producers
and Exporters Association (EHPEA), were also active in promoting floriculture.
In sum, the central role of the state in developing the sector through appropriate
policy and institutions, and the presence of active industrialists, firms, and a sectoral
association were key success factors. While the transformative potential of this sector
is immense, the process of achieving this also offers substantial learning
opportunities for government and other economic actors. For the record, the findings
also show that there are multiple shortcomings in industrial policy. Examples include
the inability to sustain the growth of the sector after 2008, the lack of technological
deepening, and inadequate participation by domestically owned firms.
At first glance, floriculture, with its rapid expansion, seems to evidence the huge
success of an activist state in designing and applying industrial policy. The clarity and
consistency of this policy seems to point to a strong political commitment to it.
However, the findings suggest that there is more to this sector’s experience than a
neat ‘success or failure’ or indeed ‘comparative advantage versus activist state’
dichotomy. The relatively successful sectoral performance was neither a simple
product of well- (p.193) designed policy nor achieved in spite of industrial policy.
Instead, the following themes emerge.
First, the notion of ‘political will’ is of itself unhelpful, especially when it suggests an
abstract (‘autonomous’) and coherent commitment acting on a sector. Rather,
‘political will’, or sustained commitment to a clear strategy, is endogenous to the
political economy of a sector (and to the array of other sectors, institutions, and
interlocking interests). In the example of Ethiopian floriculture, and especially in
comparison with other sectors covered in the book, the evidence suggests that policy
was especially effective because of the relatively small (and recent) set of interest
groups. This simplified communications and relations between government and
sector interests. However, as the sector has evolved and new interests and challenges
have developed, these relations have become more complex and the cracks in policy
coherence have become more evident.
Second, while it is clear there have been significant successes in sector performance
and in policy effectiveness, the success on both counts is far from unmitigated. The
research highlights a number of problems for the industry and serious weaknesses in
policy design and implementation. It is as much from these shortcomings as from the
successes that scholars and policymakers can learn.
Third, this circumstance underscores the need for industrial policy to be sector-
specific: what works for cement may not work for floriculture. Policies that were
appropriate at one stage may have to be adapted later. Understanding the interplay of
linkages, political economy, and industrial structure does offer the potential for
effective policymaking.
Fourth, the research highlights the significance of learning by making policy, of the
dynamic emergence of new and shifting challenges, and to some extent of
unanticipated consequences that provoke varying responses over time. This makes it
virtually impossible, at least in the absence of a highly detailed understanding of
political economy and industrial structure, to predict the outcome of industrial
strategies. One cannot, that is, ‘read off’ likely success or failure from generalized
models, be they inspired by laissez-faire or structuralist development economics.
A final word: government–business dialogue or the collective learning of state and
industry is essential to industrial policy’s success, but only succeeds if the state plays
a leading and activist role. Mazzucato’s quote (p.194) below beautifully encapsulates
the spirit of Ethiopia’s industrial policy in the floriculture sector:
When not taking a leading role, the State becomes a poor imitator of
private sector behaviours, rather than a real alternative. It is a key partner of
the private sector—and often a more daring one, willing to take the risks
that business won’t.…The State cannot and should not bow down easily to
interest groups who approach it to seek handouts, rents and unnecessary
privileges like tax cuts. It should seek instead for those interest groups to
work dynamically with it in its search for growth and technological change.
(2013a: 5–6)

Notes:
(1) In late 2013, many Ethiopian migrants returned from Saudi Arabia because of ‘permit’-related
charges.

(2) Interview with DBE president in May 2012.

Curing an Underperformer? Leather and
Leather Products
Chapter:
(p.195) 6 Curing an Underperformer? Leather and Leather Products
Source:
Made in Africa
Author(s):

Arkebe Oqubay
Publisher:
Oxford University Press

DOI:10.1093/acprof:oso/9780198739890.003.0006
Chapter 6 focuses on the same issues as the previous two chapters, but in relation to
leather and leather goods, where performance has been very different from that of
floriculture and cement. Why has a country with the largest livestock population in
Africa and with a century of manufacturing experience in the sector behind it been
unable to achieve more than sluggish and erratic growth? Industrial policy relating to
leather and leather products has been unable to reverse this poor performance and
fully exploit the potential for linkage effects and competitive integration into the
global value chain. However, recent developments have begun to yield more
investment, better quality, and more exports of higher-end products. This chapter
explores this history and the recent changes, providing insights into the policy
process, the constraints on the sector’s expansion, and the structure of the sector.
Keywords: Ethiopia, leather, livestock, industrial policy, manufacturing, global value
chain, path dependence, political economy, industrial structure, linkages

6.1 Overview and perspectives

The Ethiopian leather and leather products industry has existed since the 1920s and
dominated the country’s limited manufactured exports until the 1990s. Almost all
firms in this sector were originally owned by foreigners, mainly Armenians. The first
shoe factory, Darmar, was founded in 1927, and the first tannery (Asco) was
established in 1925. Growth has historically been slow, and lower-end products
dominated the product mix. Export earnings amounted to merely tens of millions of
dollars, with close to 95 per cent of such earnings coming from semi-finished hides
and skins, and the balance from finished leather and leather products (ERCA 2012a).
In 2011, Africa’s export earnings from this sector accounted for about 1 per cent of
global export value, and Ethiopia’s share of Africa’s export earnings was less than 1
per cent. Ethiopia’s share of African footwear production was less than 0.3 per cent in
2011 (FAO 2011). The African export market is entirely dominated by Tunisia (61 per
cent) and Morocco (37.5 per cent). In terms of exports of semi-finished and finished
leather, Ethiopia stood second to Nigeria (40 per cent), and was followed by Kenya
(17 per cent).
The main puzzle is that even with prolonged manufacturing experience and plentiful
livestock resources, performance in Ethiopia has been persistently poor. This sector
suffered from inertia and path dependency that arguably acted as a brake on policy
changes. Evidence also suggests that obvious linkage effects from the rich
endowment of primary input materials are far from automatic.
(p.196) A number of explanations are offered for this poor performance, but they are
generally unsatisfactory. Several of these will be familiar to the reader from earlier
chapters. The first standard explanation relates to factor endowment, in terms of
which Ethiopia and other African countries should focus entirely on leather and
leather products, given abundant cheap labour and livestock. Ethiopia’s livestock
population ranks sixth in the world and first in Africa, and the country possesses a
quarter of Africa’s cattle. Its goat and sheep population is the second biggest on the
continent, and respectively the eighth and tenth largest worldwide (FAO 2011). Yet the
sector has underperformed. Although natural endowment is an important
consideration in policymaking, this argument fails to recognize that sector
development and exploitation of advantages are shaped by effective policy choices
and institutional adjustments. This is better demonstrated by comparing the sluggish
growth of the leather sector in Ethiopia with Italy’s leadership in the industry. Italy,
an industrialized country with higher labour costs and shortages of raw materials,
continues to be the industry’s leader and biggest exporter (exporting more leather
products than the total exports of all developing countries bar China). Italy has
maintained its position by constantly enhancing its technological leadership and
competitive edge. Thus, the endowment factor is inadequate in explaining the
circumstances of Ethiopia’s leather and leather products industry.
Another explanatory approach relates to the GVC. Its proponents argue that the
division of labour is determined by a country’s position in the GVC. Schmitz (2004:
1) argues that, ‘the upgrading prospects of local enterprises differ according to the
type of GVC they feed into’ (see also Gereffi and Fernandez-Stark 2011;
Kaplinsky 2005; Humphrey and Schmitz 2004). Although the GVC literature is useful
in understanding the industry, it does have serious limitations. To the extent that it
sometimes places the weight of ‘structuring’ and influence on core industrialized
countries, it risks inculcating a kind of fatalism with regard to developing-country
agency (thus echoing dependency theory). It also diverts attention from the arguably
more important issues of domestic policy, in which the interaction between political
economy factors and GVCs features prominently.
In contrast to these explanations, a political economy and structuralist perspective
(discussed in earlier chapters) may provide a much sounder explanation. In particular,
the political constraints on policymaking and (p.197) institutions, and the role played
by various interest groups and their relations with the state, can open a new avenue of
analysis. In addition, Hirschman’s linkage approach (1958, 1981, and 1992) offers a
sensible conceptual framework for understanding linkage dynamics in terms of
potential, and converting this potential into material outcomes. Guided by this
perspective, this study uses extensive primary and secondary data. The analysis is
based on disaggregation of the industry into subsectors, namely the tanning sub-
industry or the manufacture of leather, and the leather products sub-industry,
comprising the manufacture of footwear, gloves, jackets, belts, and other accessories.
As with the other sectors in this study, the key research tools were a quasi-census
survey covering the whole industry (the fifty-seven respondents represented almost
90 per cent of firms in the sector); qualitative and in-depth interviews (seventy-five
interviews); site visits and observations in factories (mainly tanneries and leather
products); and document reviews. The industry classification is based on the
International Standards of Industrial Classification, Revision 4 (hereafter ISIC). The
statistical data of CSA, FAO, International Trade Centre (ITC),1 and UN
COMTRADE have been used for historical and comparative purposes.2
The research yielded the following main insights. First, by comparison with
floriculture or cement, the industry’s growth in terms of output, exports, and
employment has been sluggish. Thanks mainly to inappropriate policies, the sector
has underperformed badly. Industrial policy in the leather and leather goods sector
has long been unable to improve performance or to fully exploit linkage potential and
insertion into the GVC. However, matters have started to change very recently.
Second, one of the main reasons for underperformance is the lack of effective policy
response to the challenges of raw material supply. This industry is material-intensive,
so a reliable supply of low-cost and good-quality raw materials is the key determinant
of its growth. Input analysis reveals that material costs amount to more than 70 per
cent of production costs. Despite the comparative advantage suggested by abundant
herds and flocks, the sector experienced shortages, rising prices, and low-quality
inputs (hides and skins for tanneries, finished leather for leather (p.198) products).
For instance, prices more than tripled between 2009 and 2011, and quality
deteriorated from 1991 to 2012 (FAO 2009; LIDI 2012d; ELIA 2012). The principal
failure here has been the lack of a consistent and effective policy that works for the
large number of scattered rural producers. This is despite of a political commitment to
small producers and the high priority afforded the industry in the GTP.
Third, interests within the industry have been contradictory, contributing to the
disappointing performance. At least until recently, tannery owners dominated the
industry but have shown little interest in technological upgrading of the sector. This
has been compounded by an inadequate policy response (for instance, in supporting
value addition and introducing a multi-modal transport system) and inappropriate
institutions (notably, the absence of institutions for inputs). This failure to invest in
industrial upgrading has meant the sector has fallen prey to a buyer-driven GVC
(Schmitz 2004). This situation may not preclude local initiatives to accelerate
upgrading, but does present policy challenges.
Fourth, as shown below, recent developments have begun to yield more investment,
better quality, and more exports of higher-end products. In the 1990s, the shoe sub-
industry was on the verge of extinction in the face of stiff competition, in particular
from China. However, it has been able to survive and improve its market share in the
domestic market. Nevertheless, the evidence is still too thin to give rise to unbridled
optimism about this sub-industry, since there are still contradictory interests (for
instance, around path dependency and the low value trap) that are poorly managed by
the state, and the fundamental input supply problem remains unresolved.

6.2 Sector performance and policy outcomes

The industrial development strategy emphasizes the integrated development of
animal resources, meat processing, and the leather and leather products industries.
Development of animal resources is underscored as the cornerstone of the meat and
leather sectors. The strategy further highlights the aim to develop local capacity to
export finished leather by improving the quality and productive capacity of existing
tanneries as well as through new investments. Also emphasized by the strategy is the
employment-creation potential of the leather goods sub-sector and the (p.199)need to
enhance production capabilities by developing and managing human resources,
improving input supply, and by gradually developing a domestic input and
accessories industry. In reviewing this sector’s performance, it is relevant to assess
the effectiveness of industrial policies over the past two decades. As this sector is
export-oriented and labour-intensive, appropriate performance indicators are output,
exports, employment, and labour productivity.

6.2.1 Output growth

Volume trends have been erratic, with periods of expansion followed by substantial
declines and then rapid recovery (Figure 6.1). Production volumes in the tanning sub-
sector, measured in square feet, increased on average by 20.6 per cent per annum
between 1992 and 2011. The footwear sub-sector, on the other hand, grew on average
by about 13 per cent per annum in the same period. Nevertheless, these average
growth rates conceal significant fluctuations. Tanning production was close to 101
million square feet in 1992; declined to about 70.2 million square feet by 2000; and
bounced back to145.7 million square feet in 2001. It then continued to decline before
it reached a new peak in 2009, at over 160 million square feet. Growth in the sub-
sector appears to have declined again in recent years.
Likewise, the average growth rates for footwear mask fluctuations. Some 874,000
pairs of shoes were manufactured in 1992. Production generally

Figure 6.1. Total production of semi-finished and finished leather by volume, 1992–
2011
Source: Own computation based on CSA Report of Surveys on manufacturing, 1992–
2012
(p.200) continued to increase in subsequent years, reaching a new peak of close to 1.6
million pairs in 2000. Production then plummeted to 846,000 pairs of shoes in 2005,
even fewer than in 1992. Thereafter, production recovered, rising to some 2.2 million
pairs in 2011. This figure is minuscule in comparison with Africa’s total output (91
million pairs) or the world’s (4.4 billion pairs) in 2010. Morocco alone exported 29.3
million pairs and Tunisia 25.1 million pairs in that year.

6.2.2 Export performance

Export Volume and Export Earnings

The leather and leather products industry has been a significant source of export
revenue for Ethiopia for many decades. The export performance of the leather
industry is consistent with the production performance of the industry described
above. Although export earnings increased on average, they have experienced
significant fluctuations over the years. Leather and leather products exports from
1999 to 2012 totalled more than $1 billion. Although this figure seems large, it
becomes less striking given the century-long existence of the sector and the country’s
acknowledged potential in this sector. Ethiopia’s share of the global market and of the
African market is almost nonexistent.
Export revenue between 1999 and 2011 showed significant instability and variability.
For instance, exports by the leather sector ranged between 3,000 tons in 2011 and
16,000 tons in 2007. Export earnings were $60 million in 2004 but only $56 million
in 2010 (ERCA 2012a; LIDI 2012g). There was, moreover, a significant gap in actual
export revenues and targets during 2002–12 (Figure 6.2). Actual performance fell
below 50 per cent of target performance, possibly because of shortcomings in
planning and implementation. This mismatch highlights both the government’s desire
to develop the industry and its inability to do so. On the other hand, there is no real
evidence on the magnitude of transfer pricing/under-invoicing, tax avoidance/capital
flight, or smuggling.
Because of the industry’s relatively slow growth, between 1999 and 2012 its share of
total earnings declined from 7 to 3.5 per cent, and its share of manufactured export
earnings decreased from 86 to 52 per cent. Moreover, this decline is not the result of
exceptionally performance by other sectors.
(p.201)

Figure 6.2. Export target and performance of leather sector (in million $)
Source: Own computation based on unpublished LIDI data, December 2012

Figure 6.3. Share of semi-finished and finished leather (based on export earnings),
1999–2012
Source: Own computation based on unpublished data from ERCA, October 2012
The transition to higher-value added leather and leather products exports
was slow, and has only recently begun (Figure 6.3). There was a modest increase in
the export share of finished leather after 2006. In 2011, it reached close to $50
million (1,357 tons) and leather products exports reached $11 million (586 tons).
Exports of raw hides and skins dropped significantly (from 15,000 tons to 5,000 tons)
after 2009, following the government’s export ban on crust (LIDI 2011). Nonetheless,
overall export revenue for the industry has increased, thanks to the shifts towards
value-added products. Export earnings from raw hides and skins were $5.3 per kg in
1999, while earnings from processed and value-added leather and leather products
increased 4.7 times to $25 per kg in 2012 (LIDI 2012g). (p.202) However, not all
tanneries could upgrade their product mix, so the increase in total earnings was
unevenly spread.
Comparison with other developing countries in this regard is instructive, and Vietnam
is an appropriate comparator for many reasons. It has the same population size as
Ethiopia and made the transition from a centrally planned economy after 1986.
Vietnam, which was never on the global leather and footwear map, has since recorded
impressive growth to become one of the top five leather and footwear exporters in the
world. In 2000, its leather and footwear exports exceeded $1 billion, and by 2007
they surpassed $4 billion. They have recently topped $10 billion. Between 2007 and
2011, shoe exports increased by 164 per cent and bag exports by 2,005 per cent
(LEFASO 2012). The industry contributes 10 per cent of total export earnings, and
leather products are the sixth most important export commodity. The sector employs
more than 650,000 workers, about 10 per cent of national employment. Major trade
partners are the EU (accounting for 48 per cent) and US (30 per cent), and the
country also exports to East Europe, Brazil, and China.

Export Destinations
Ethiopia’s exports of leather and leather products are concentrated on a few
destinations. The top ten destinations in 2011 were Italy (32 per cent), China (22 per
cent), India (14 per cent), Hong Kong (12 per cent), UK (10 per cent), Indonesia (2
per cent), Germany (2 per cent), Turkey, Romania, and Thailand (ERCA 2012a).
Nonetheless, changes are happening. In 2009, Western European countries absorbed
more than 55 per cent of the total exports in the sector, while Asia’s share was below
37 per cent. By 2011, these shares stood at 43 per cent and 51 per cent respectively.
This increased trade between Asia and Ethiopia seems to be caused by increased
Chinese and Indian investment in the sector. While Italy, a traditional trading partner,
is still the single largest buyer (one-third of Ethiopian exports), the shares of China,
India, and the UK have shown some increase. Although diversification has its own
benefits, existing evidence is insufficient to show advantages in terms of price or
technological or skills development.
Industry Concentration
In the exports of crust and finished leather, the top five exporters, accounting for 56
per cent of the total, were Ethiopian Tannery (p.203) (a privatized firm owned by
Pittards, a UK company), 15 per cent; ELICO (a privatized firm currently owned by
the Midroc Group), 12 per cent; followed by China-Africa Tannery (11 per cent),
COLBA, and DIRE (which are locally owned), 9 per cent each. In shoe exports,
German-owned ARA (29 per cent) and four local firms dominated in 2012. Lately,
changes have come to the leather products sub-sector with the advent of new players
such as Huajian, the largest Chinese manufacturer of ladies’ shoes. So far, industrial
concentration in Ethiopia is much lower than in Vietnam, where five firms account
for about 70 per cent of exports (GIZ 2009). With increased FDI in this sector,
however, concentration in Ethiopia is expected to increase.

Competitive Edge
Poor quality also contributed significantly to weak export performance. The industry
is incapable of timely exports of good quality products at competitive prices.
Customer orientation and competitive spirit were below the desired level.
Respondents cited the poor quality of raw materials and lack of skilled personnel as
the main causes for this. Lack of cost competitiveness was primarily due to shortages
and the high prices of raw hides and skins. As Table 6.1 starkly shows, material inputs
account for about 85 per cent of production costs in both leather and leather products.
Hides and skins or finished leather accounted for 66–70 per cent of production costs.
This finding on inputs as binding constraints is in line with the World Bank’s study
(Dinh et al. 2012) and Cramer’s (1999d) observations. By 2014, the crisis in the raw
material sector for leather remained fundamentally unresolved.
Table 6.1. Major inputs in tanning and leather products
Major inputs Leather sub-industry Leather products sub-industry
(in %) (in %)

2010 2011 2010 2011

Local raw materials and semi-processed 60.6 66.4 44.8 70.6

inputs

Imported raw materials and semi- 24.5 26.9 31.6 16.2

processed inputs

Utilities, transportation, and logistics 11.2 4.2 3.2 3.3

Labour and related service 3.7 2.5 20.3 9.9

Total 100 100 100 100

Source: Oqubay (2012)

(p.204) 6.2.3 Employment generation
This industry is labour-intensive, especially the leather products sub-industry, which
also employs more female workers (about 46 per cent of the total workforce) than the
tanning sub-industry. According to CSA Survey Reports (1991–2012), annual average
employment growth in the leather and leather products industry is 4.54 per cent.
Between 1992 and 2003 the workforce grew by only 1.04 per cent to 7,352
(CSA 1993, 2012c). An annual average growth rate of 3 per cent was recorded between
2003 and 2007, while it doubled between 2007 and 2012 because of new investments.
The number of firms seeking investment certificates increased from eighty-seven in
the fifteen years after 1992 to 133 firms between 2007 and 2011. However, actual
investment did not exhibit such high rates of growth.
Total employment in the fifty-seven medium and large firms sampled was 15,443 in
2012. More than 57 per cent were permanent workers, 20 per cent were on contract,
and the remainder were temporary. By contrast, permanent employees accounted for
more than 70 per cent of the workforce in the cement industry and 76 per cent in
floriculture.

6.2.4 Labour productivity and capacity utilization

Labour Productivity Trends

Labour productivity is measured as production per person per day, where production
is in terms of square feet in tanneries, and pairs per day in footwear. For leather
garments, reliable data were unavailable. Many existing leather products
manufacturers do not specialize in leather garments, and only produce them in small
quantities. Data on footwear had to be disaggregated into leather and non-leather
footwear (canvas, rubber, plastic). Labour productivity trends in leather footwear
show that the sub-sector has remained stagnant for a very long time. As
Figure 6.4 illustrates, it declined following liberalization in the early and mid-1990s
and has never recovered. Production per person was 1.00 pairs per day in 1993, 1.17
pairs in 2002, and 1.06 pairs in 2011. Compared to other countries, current labour
productivity in the leather footwear sub-industry is low. Intra-country productivity
comparisons have their own drawbacks, as many variables distort the comparison.
However, approximate average productivity in men’s leather shoes was 6 pairs in
China, 4 pairs in (p.205)

Figure 6.4. Labour productivity in footwear, 1979–2011

Source: Own computation based on CSA Report on Large and Medium Scale
Manufacturing and Electricity Industries Survey, 1978–2012

Figure 6.5. Labour productivity in tanning, 1977–2011

Source: Own computation based on CSA Survey of Manufacturing, 1978–2012
Vietnam and India, 3.3–3.5 pairs in South Africa, and according to recent observation
by India’s Footwear Design and Development Institute (FDDI), 3–3.5 pairs in
Ethiopia (FDDI 2012).
Tanneries appear to perform better, but labour productivity has been highly erratic
(Figure 6.5). The production per person was 77.85 square feet per day in 1993, 61.32
square feet in 2002, and 63.64 square feet in 2011. At times, however, labour
productivity has doubled, as in 2001 and 2009.

(p.206) Capacity Utilization

Another indicator of efficiency and productivity is utilization of installed capacity.
There can be biases in this indicator if it is not measured properly. Survey results
show that the sector suffers from low capacity utilization. According to the survey for
this study, more than two-thirds of firms operated below 60 per cent of capacity
(Table 6.2). The main reported causes for this were raw materials shortages, delays in
input supply, market or demand problems, lack of working capital, and lack of skilled
workers.
The tannery sub-industry, which is more capital intensive, has very low capacity
utilization: only 67 per cent for soaking and 57 per cent for finishing. Moreover, the
ratio of finishing to soaking capacity for sheep and goatskins was 68 per cent
(Table 6.3).
In sum, it is clear that the overall leather sector has a long way to go to meet the
GTP’s target output of $500 million by 2015. Employment has remained below
20,000, insignificant in comparison with the 650,000 employed in this sector in
Vietnam. China’s five million workers in the sector underscore the employment scope
of this sector.
Table 6.2. Capacity utilization of firms (whole sector)
Capacity utilization (%) Number of firms Per cent

<20 6 13.0

20–40 12 26.1

41–60 12 26.1

61–80 12 26.1

>80 4 8.7

Total 46 100

Source: Oqubay (2012)

Table 6.3. Capacity utilization of tanneries (2011)
Product type Soaking capacity Finishing capacity Ratio of finishing to soaking
utilization (%) utilization (%) capacity (%)

Sheep and goat 70.63 57.87 68.17

skins

Cattle hides 57.29 54.80 93.33

Total 66.53 56.82 74.83

Source: Own computation based on unpublished LIDI data, October 2012

(p.207) 6.3 Industrial structure, economic, and technological characteristics

This analysis of the structure of the leather and leather products industry of Ethiopia
is based primarily on the quasi-census conducted by the author (Oqubay 2012). The
results highlight the dominance of small and medium firms, the low entry and exit
barriers, and the wide latitude for performance.

6.3.1 Industrial structure of the Ethiopian leather and leather goods industry

Geographical Concentration of Firms and Industrial Cluster

Industrial clusters are ‘the geographical concentration or localization of enterprises
producing similar or closely related goods in a small area’ (Sonobe and Otsuka 2006).
Marshall’s (1920) pioneering work states that agglomeration economies provide the
advantage of information spill- over, division of labour and specialization, and labour
market development. Porter (1998), for his part, emphatically highlights the role of
industrial clusters in many industries: ‘A nation’s competitive advantage in industries
is often geographically concentrated. Geographic concentration is important to the
genesis of competitive advantage, and it amplifies the forces that upgrade and sustain
an advantage…Government policy has an important role in nurturing and reinforcing
clusters’.
The Ethiopian leather and leather products industry comprises sixty-five medium-
and large-sized firms clustered in and around Addis Ababa. Fifty-six per cent (thirty-
two firms) are in Addis Ababa, 36 per cent in Oromia (seventeen firms), and 11 per
cent in each of Amhara and Tigray (six firms each). Other than Addis Ababa, Mojo, a
rapidly growing industrial and transport hub on the Addis–Djibouti corridor, had the
highest concentration of tanneries. Of the firms outside Addis Ababa, more than 70
per cent lay within a 200-km radius. Historically, all the old tanneries and shoe
factories were established along rivers in Addis Ababa and Mojo because of the local
availability of skilled labour, because Addis Ababa is also the biggest domestic
market in Ethiopia, and because of better infrastructure and services. Similar
agglomerations have been observed in Brazil, China, Italy, and Vietnam, a strong
indication of the role of industrial (p.208) clusters in the leather products industry.
Although there have been recent efforts in Ethiopia to encourage cluster
development, this is at an early stage and requires further study.

Size of Firms
According to CSA survey data (2012b), the average size of manufacturing firms in
2010 was eighty-one persons, while 149 persons were employed per firm in the
leather and leather products industry. This trend appears to be associated with the
firms’ origins as SOEs, which were larger than privately owned enterprises. Some of
the new FDI-owned firms are, however, much bigger than the privatized SOEs. For
instance, Huajian had 1,200 workers in 2012. Moreover, Vietnamese factories on
average employ four times as many workers as their Ethiopian counterparts
(GIZ 2009; LEFASO 2004, 2012). One of the largest footwear manufacturers in
Vietnam (owned by the Taiwanese Chin Luh Group, which is the major manufacturer
for Nike) employs 24,000 workers and manufactures 20 million pairs of shoes per
annum. While the subcontract production system is not present in Ethiopia, 70 per
cent of firms in Vietnam operate on this basis.

Ownership and Corporate Structure

SOEs dominated the sector until the first decade of the new millennium. All large
firms were owned by the state under the Derg regime’s command economy. The
leather and leather products industry was opened to private investment following the
transition to a market-led economy in the early 1990s. By 2012, all nine of these
SOEs had been fully privatized, and two-thirds of them had been bought by domestic
private firms. Unlike in cement and floriculture, Ethiopian nationals now dominate
the leather and leather products sector. This was not a direct result of government
policy, since the investment law opened the sector to FDI. Again unlike floriculture,
the Ethiopian diaspora has played an insignificant role in the leather industry,
investing in only four firms. About three-quarters of the sector’s businesses are sole
proprietorships or private limited companies, while ‘share companies’ have little
presence. Nearly two-thirds of the firms surveyed agreed that they were ‘family-
businesses’. Moreover, the general managers of these firms were family members,
owner-managers, or partner-managers, a pattern reminiscent of floriculture, in which
72 per cent are owner-managers.
(p.209) Of the total number of firms in the industry in 2011, foreign-owned firms
accounted for 23 per cent in the leather manufacturing sub-industry and 28 per cent in
leather products (footwear and gloves). Foreign-owned firms produced less than 25
per cent of the leather, and around 30 per cent of leather products, and accounted for
about 35 per cent of leather exports and 55 per cent of leather products exports. Thus,
foreign investors are more active in leather products and oriented towards exports.
Domestically owned firms sold 38 per cent of tannery output and 68 per cent of
footwear production to the local market. The origin of FDI was less concentrated,
with China, the UK, and a Chinese–Italian joint venture each having a 21 per cent
share, and India a 14 per cent share.
In Vietnam, by comparison, FDI plays a more important role (45 per cent), and state-
owned firms have a presence, albeit diminishing, in footwear production. In 2009, the
sector had 465 medium and large footwear manufacturers and forty-two tanneries.
Domestically owned firms accounted for 50 per cent (256 firms), while seventeen
firms were jointly owned, and six were SOEs. The balance were foreign owned
(GIZ 2009; LEFASO 2012). In 2004, FDI accounted for 44 per cent of footwear;
domestically owned firms 28 per cent; SOEs 21 per cent; and jointly owned
companies 6.5 per cent.

6.3.2 Economic and technological characteristics of the industry

Technological Characteristics and Latitude for Poor Performance

The leather and leather products industry has distinctive technological and economic
characteristics. It is classified as light industry and labour-intensive. Based on
technological intensity, Lall (2000b, 2003) classifies leather production as low
technology (LT-1). The production process is product-driven, and depends on labour
skills and variable market orders. Economies of scale and scope are smaller in leather
and leather products than in industries such as cement. However, there is an
increasing trend towards concentration.
Given the limited capital requirements and the divisibility and low-tech nature of the
machinery, barriers to entry are low. However, entry into international markets,
driven by fashion houses and global retail chains, is very competitive (Schmitz 1999;
Schmitz and Knorringa 2000; Gereffi 1994; Gereffi, Humphrey, and Sturgeon 2005).
Delivery times are (p.210) becoming shorter and product varieties are increasing,
while cost-competitiveness is stiffening. The latitude for poor performance is
therefore narrower in leather products, and the competitive pressures facing Ethiopian
leather products manufacturers are stronger than for tanneries. The latter have wide
latitude for poor performance in comparison not only with leather products, but also
industries such as cement or floriculture. This is in line with the argument ‘that export
structures, being path-dependent and difficult to change, have implications for growth
and development. Low technology products tend to grow the slowest and technology
intensive products the fastest’ (Lall 2000b: 1). This may imply the need for a top-down
approach, where the leather products industry becomes stronger and pulls the tanning
sub-industry up (UNIDO 2012). This is also in line with capitalist globalization in all
industries after the 1970s, including the dominance in this sector of global actors such
as Nike, Adidas, Reebok, and Puma (Nolan 2012). These four brands dominate the
$75 billion global athletic footwear market and have a 20 per cent market share. The
GVC in this sector is characterized by increasing buyer dominance (UNECA-
AU 2013; Gereffi et al. 2005; Schmitz and Knorringa 2000; Kaplinsky and Morris 2000;
Schmitz 2007).
The key stages in the leather and leather products value chain are the recovery and
supply of hides and skins, their conversion into semi-processed and finished leather,
the manufacture of leather products, and marketing. The sector is material-intensive,
and has much stronger backward linkages, especially with agriculture. In terms of
linkages ‘calling forth’ activities, it may be worth establishing a sequence here: the
industry was created because of the existence of the livestock sector, which implies a
forward linkage from livestock. As the leather sector develops, it has the potential to
stimulate huge livestock production (productivity and quality) through a backward
linkage. This shows there is the potential for a ‘feedback loop’. However, this is not
evident in Ethiopia due to the leather products industry’s lack of dynamism. In
addition, agriculture (specifically, livestock) is vulnerable to drought and variable
rainfall, thus constituting immense ‘supply side’ technological uncertainty. This has
significant implications for policy design in the leather sector in terms of institutional
support, extension and veterinary services, credit, access to grazing and watering
facilities, fodder cultivation, post-slaughter care, and marketing. Moreover, livestock
owners and leather producers, on account of their dispersal, lack of organization, and
sometimes conflicting (p.211) interests, are poorly placed to bring pressure to bear on
government. This is relevant to government policy commitments and concerted
effort.
Another key dynamic relevant to policy in the leather sector is the pressure to
improve environmental compliance. Tanning has significant environmental effects,
including air, water, and ground contamination. Leather substitutes are expanding,
while buyers and regulators are increasingly insisting on strict environmental
protection and the application of eco-friendly labels, such as LITE standards (Low
Impact to Environment) and Leather Working Group. A current major issue is the
lobbying to ban the use of chrome in tanning. This represents a significant challenge,
as chrome is very economical and is used by nearly 80 per cent of the world’s
tanneries. Decreased applications of salt are also being strongly encouraged.
Furthermore, traceability is becoming important, as is the tracking of carbon
footprints. And at the same time as the leather industry is being pressured by buyers
and others to improve its environmental compliance and promote its environmental
image, it is under constant pressure to reduce production costs and expand the fashion
range (FAO 2008; Kaplinsky 2005; OECD 1976).

Technical Capacity and Staff Composition

A firm’s competiveness and productivity is largely driven by its pace and scale of
learning, and one aspect of the technical capacity of firms is staff composition,
specifically the extent of the professional and technical core. The number of
professionals in fifty-two firms was 440, or 2 per cent of the total workforce, while
the share of technical experts and engineers was below a quarter of total professional
staff (or 1 per cent of overall employees). This is believed to be very low. Moreover,
only 4 per cent of all workers have technical or vocational training. In view of the
industry’s skilled worker shortage and low retention levels, firms may have less
motivation to invest in skills development.
6.3.3 Market structure and international competition

Global Market Structure

The global market for leather and leather products was $60 billion in 2011
(FAO 2011). Growth of the leather products sub-sector is determined by the global
economic situation and consumer demand. On the other hand, the leather sub-sector’s
growth is determined by demand from the leather (p.212) products industry, and
production in the meat and dairy industries (FAO 2008; OECD 1976).3 In terms of
global footwear production, China is the leader, dwarfing all countries with 42 per
cent of world production and employing five million workers (FAO 2011). The
Chinese footwear sub-industry is the biggest employer in the world (China Leather
Industry Association 2012). Vietnam, Brazil, and India follow far behind in global
exports of footwear. While Italy leads in higher-end products, China, India, and
Brazil are focused on the lower end.4
The main source of hides and skins for China, Europe, and the Far East has been the
US. The US exports some 750,000 tons of bovine hides and skins per year, while
China is the biggest importer of bovine hides and skins (FAO 2013). Brazil’s annual
domestic consumption of hides and skins stood at 12 million in 2011. Brazil has 213
million head of cattle (14 per cent share of the world’s cattle population), an increase
of 23 per cent in ten years. This remarkable growth is attributed to Brazilian policy
and the Brazilian leather promotion agency’s initiatives to improve sustainability,
quality, innovation, and creativity.

6.3.4 Summary
In reviewing the Ethiopian leather and leather products sector, certain features stand
out. First, small and medium, mostly family-owned, firms dominate both sub-sectors.
Economies of scale and scope are low, as are entry and exit barriers. Because the
sector is not demand-constrained, there is the opportunity to engage more firms. The
firms’ limitations suggest the need for increased support from government agencies;
marketing, research, and development institutions; and industrial associations.
Learning from Brazilian and Vietnamese experience may be important. Second, as a
labour-intensive industry, labour productivity plays a central role, implying a need to
focus on developing production capability. The dominance of high productivity and
low cost countries (p.213) (such as China, Vietnam, and India) and the increased
buyers’ bargaining position in the GVC pose formidable challenges for entrants like
Ethiopia. Industrial clusters, which have played important roles in Italy, China, and
Brazil, could do the same in Ethiopia.
Third, the latitude for poor performance is wide, implying that policy effectiveness is
hampered by the lack of inherent pressure in the production process. Firms prefer to
sell on the domestic market rather than in the more competitive export market.
Moreover, the small size and limited capital base of the domestic firms limits the
scope of specialization, and the firms view diversifying to other business as a better
option to mitigate their business risks. This implies, arguably, the need for innovative
policymaking and institutional arrangements. Relatively, the leather products sub-
industry exerts greater pressure of export discipline.
Fourth, just as energy is critical to cement, and air transport to floriculture, the input
supply issue is a binding constraint on this industry. This is due to the material
intensity of the product. In view of the factor endowments of Ethiopia, which also
became a key factor in attracting investment, the key policy challenge will be
addressing this binding constraint. With these premises, the following sections review
the policy and institutional responses of the government.

6.4 Linkages, path dependence, and industrialization

6.4.1 Path dependence and low value trap in leather and leather products industry
Path dependence ‘characterises…those historical sequences in which contingent
events set into motion institutional patterns or event chains that have deterministic
properties’ (Mahoney 2000). Path dependence is locked in and hard to escape (Vergne
and Durand 2010). The Ethiopian leather and leather products industry has
experienced path dependence in the form of a low value trap. This dependence has
been reinforced by many factors. First, pressures from the GVC have been less
acutely felt, as Ethiopian leather and leather product output is based on tanneries.
Footwear production and exports have lagged leather production. The tanning
industry was dependent on raw hide and skin exports until 1975, and on semi-
finished leather exports until December 2011. Under (p.214) the command economy
(1975–91), government was forced to prohibit hide and skin exports in order to
secure supplies for nationalized state-owned firms that were experiencing shortages
of, and high prices for, raw materials. Evidence suggests this intervention was not
made with a view to upgrading the industry.5
Until 2007, the main export was wet-blue and pickle. It was only in that year that the
government imposed a high tax to discourage exports of low-end semi-finished
leather and encourage exports of crust. After 2011, crust exports were also
discouraged by a high export tax, and exporting finished leather was promoted.
Despite these interventions, the sector faces major challenges in terms of value
addition, localization of inputs (local content is below 60 per cent), industry
upgrading, and product design and development (Oqubay 2012). This stands in
contrast to Vietnam, which has managed to break into the GVC, competing primarily
on cost (against China and others) after liberalization of the centrally planned
economy in the late 1980s. It is now the second largest manufacturer of footwear in
the world.
Second, exports of semi-finished leather suited the Ethiopian tanneries because they
only had to deal with a limited number of European tanneries (unlike the footwear
and other leather products manufacturers). This reduced buyer uncertainty and
facilitated good relationships with higher value leather manufacturers. The prices of
semi-finished leather were stable between 1991 and 2008, and the trend changed only
after the end of 2008 because of falling retailer demand. For instance, in the US hide
prices decreased by more than 50 per cent in 2009 (FAO 2008). Changes in
consumption patterns for leather products do not directly transfer to crust level
manufacture. The purchasers of finished leather are leather products manufacturers,
who put heavy pressure on leather processors and insist on stringent specifications,
costs, delivery times, and quality. The logistics of acquiring chemical inputs is more
complex and requires more working capital. After the oil price hike in 2008,
operating costs in Ethiopia increased as chemical costs rose. This situation further
worsened when raw hide and skin prices increased threefold on the local market
between 2009 and 2012 due to the establishment of new—mostly (p.215) foreign-
owned—tanneries. As a result, tanneries came under increasing pressure and the
leather products industry was hard hit.
Third, with rising demand for leather footwear, the tanneries and footwear industry
were able to sell profitably on local markets. In this they were assisted by the
increased price of imported leather footwear because of devaluation. This
strengthened the dominance of family-owned firms, some of which had also been
involved in footwear manufacture. Those who invested in tanneries after 1992 were
hides and skins traders, some of whom saw manufacturing as an extension of their
hides and skins trade. They also regarded their monopoly of the supply of raw inputs
as advantageous and were not interested in the government’s (half-hearted) efforts to
improve the supply of raw hides and skins. Instead, they considered that the status
quo offered better advantages in terms of access to raw materials, price, and quality.
Fourth, the historical evolution and composition of the Ethiopian Leather Industries
Association (ELIA) further consolidated the tanneries’ position. Tannery interests
shaped the association’s activities as well as how government was lobbied. Eight
firms, all tanneries, originally founded ELIA in 1994. In 2012, ELIA’s membership
reached thirty-nine, of which 61 per cent were tanneries, and tannery owners also
owned some of the twelve footwear manufacturers. The firms do not share cohesive
aims and seem to wish to avoid competition. For instance, they effectively lobbied to
prevent new foreign firms from investing in tanneries, as predicted by Hirschman
in The Political Economy of Import-Substituting Industrialization in Latin
America (1968). This policy decision was made on a provisional basis in 2012, and
became part of the new investment code. In the absence of a stronger reciprocal
control mechanism, the organizational capacity of these private sector investors
seems likely to consolidate industry path dependence rather than to break it.
Fifth, the low value trap and path dependence involves professionals and
management as well. Most managers and professionals had long been operating
under a quota and central planning system, which explains their lack of enthusiasm
for competition. An anonymous observer remarked that the chemists in the sector
were not aware of or receptive to technological innovation (interview, August 2012).
He also added that the owners were less interested in specializing in, for instance,
sheepskin or cattle hides, crust production, or finished leather.
(p.216) Some of the constraints in this sector could have been tackled through
appropriate policies, institutions, and organization. This did not happen for several
reasons. First, other than the broad definition of the strategy, no comprehensive
roadmap was developed to chart how the sector could upgrade its position in the
GVC. GVC analysis focuses on the ‘dynamics of inter-linkages within the productive
sector, especially the way in which firms and countries are globally integrated…’ It
also examines ‘the full range of activities which are required to bring a product or
service from conception, through the different phases of production (involving a
combination of physical transformation and the input of various producer services),
delivery to final consumers, and final disposal after use’ (Kaplinsky and Morris 2000:
4). Most of the policy decisions have not been informed by an integrated approach
and knowledge of the industry. For instance, data on the local or global outlook of the
industry are not available to MOI. Most of the reports submitted to the NECC
focused on export performance and urgent constraints rather than underlying
structures. Nor was the design of a multimodal transport system informed by
appreciation of the features of the global market in leather and related products. The
policy decision to upgrade products, though important and long overdue, was not
supported by coherent long-term plans.
Second, pressure from powerful lobbies (in particular tanneries) weakened policy
coherence and consistency. Some policies appeared incompatible, for instance, those
promoting live animal exports versus those promoting the meat processing industry.
In addition, some decisions were not followed through and firm accountability and
monitoring was lacking. For instance, the import of raw materials and semi-processed
leather was raised as far back as 2004, but was never seriously addressed. This was
mainly due to pressure from existing tanneries happy with the status quo, and
policymakers who did not insulate themselves from this pressure. Even the upgrading
of products was only implemented much later (in October 2012).
Third, the institutional setting did not effectively support the sector. There was no
strong institution working on the input side, and there has been a failure to coordinate
trade facilitation and logistics. This is critical to an industry needing to import
accessories and components, and to export under tight delivery and cost schedules.
Research and vocational training institutions have weak links to the industry. The
lead institution for the sector was strengthened only after 2010. Overall, this lack
of (p.217) institutions has further solidified the inertia. This situation stands in stark
contrast to Brazil, where research and development support is effective (see Di
John, forthcoming).
Further evidence of inertia includes the inability to diversify the product range, in
part because sub-contracting is almost nonexistent. For instance, Ethiopian footwear
firms have focused on men’s shoes, while foreign-owned firms produce women’s
shoes, for which there is a bigger and more rapidly changing market. Government has
put insufficient effort into designing policy instruments and institutions to lead firms
towards higher productivity and industrial upgrading. Rather, a perceived easy way
out has been taken in the form of misguided and detrimental efforts to continue
protection and restrict FDI, as well as the weak interest in exporting and industrial
upgrading and lack of inter-firm linkages and specialization (see Section 6.4.2 for
details). Porter (1998: 598) points to the wider significance of problems such as these
by emphasizing that ‘the most serious mistake is to support policies that will
undermine true competitive advantage, reduce the impetus to improve and innovate,
and create an attitude of dependence on government’.
In other countries, (for instance, the Sinos Valley footwear cluster in Brazil and
central eastern Italian footwear clusters) industrial associations, firms, and local and
national governments have played an active role in industrial upgrading. The Sinos
Valley cluster employs 153,400 people in 1,821 firms (tanneries and footwear and
leather article manufacturers). It exported 100 million pairs of shoes worth
approximately $900 million in 1991. The dynamism of this cluster arose in response
to international competition and increased costs. Cooperative and collective action
was the key to this success (see Schmitz 1995a, 1995b, 1998; IDS 1997).

6.4.2 Backward linkages: neglected constraints and opportunity

The evidence shows that the quality, price, and shortages of raw hides and skins were
the binding constraint on the leather and leather products industry. A number of
observations support this finding.

Inputs as Key Constraint on Productivity and Export Competitiveness

Most representatives of the firms surveyed stated that the major causes of low
capacity utilization were shortages of working capital and the (p.218) shortage and
poor quality of raw materials. These are obviously interrelated issues. In terms of
constraints on export performance, close to 40 per cent stated that the delays, high
prices, and poor quality of raw materials were the major constraint. In response to the
question on the main cause of lack of competitiveness in terms of price and timely
delivery, the firms pointed to the prices and inadequate supplies of raw materials as a
prime reason. Most of the firms also identified these factors as the major cause of the
lack of competitiveness in terms of quality. According to recent ELIA data, less than
20 per cent of skins and hides qualified as Grades 1–3, while Grade 4–6 account for
above 80 per cent (MOI 2012a; ELIA 2012). This is consistent with the findings of
other studies (see USAID 2008; Dinh et al. 2012; Global Development Solutions 2011).
A comparative study on livestock development in Botswana and Ethiopia indicates
that Ethiopia’s cattle productivity is among the lowest in the world, and that the
supply chain is dominated by ‘numerous intermediaries and actors’, a situation that
increases transaction costs (UNECA 2012).

Linkage Effects in the Leather Sector

The above evidence is consistent with the analysis of linkages, particularly the input
and output side of tanneries and the input side of the leather products sub-industry.
Tanneries and leather products have high material intensity. Based on the responses
of thirty-eight firms, hides and skins are the key components in the input–output
linkage. In tanneries (in 2011), 97 per cent of local raw materials were hides and
skins, with sheepskin alone having a 69 per cent share. Similarly, in leather products,
79 per cent of local inputs are finished leather. Hides and skins account for the lion’s
share of the cost of inputs. In leather products, finished leather has a share of 56 per
cent. Clearly, for tanneries, the supply, quality, and prices of hides and skins are
crucial to their growth and competitiveness. The same is true of finished leather in the
leather products industry.
The price of hides and skins has increased over time, constraining the growth and
competitiveness of the industry. The cost of sheepskin increased twelvefold from
ETB 7.71 to ETB 91.53 between 1993 and 2012, and threefold between 2010 and
2012 alone as the capacity of tanneries expanded. Similarly, although less
spectacularly, goatskin prices increased sevenfold between 1993 and 2012, and by
183 per cent between 2010 and 2012. The corresponding figures for cattle hides are
fivefold and (p.219) 166 per cent. The quasi-census shows that over the same period,
other input prices were fairly stable, although some showed less dramatic increases.
According to the CSA survey of livestock, Ethiopia had more than 52 million cattle
(almost half of all livestock), 24.2 million sheep, and 22.6 million goats (CSA 2012a).
Livestock is concentrated in the most populous regions of Oromia, Amhara, and
Southern Nations, Nationalities, and Peoples’ Region (SNNPR). There have been
fundamental problems in livestock development. For instance, the level of
commercialization (surplus to subsistence sold) ranges from 12 per cent for cattle to
25 per cent for sheep and 18 per cent for goats, and 10 per cent overall. The most
extensive use of cattle was for draught, and in the case of sheep and goats, breeding.
Livestock contributes up to 45 per cent of Ethiopia’s agricultural GDP (25 per cent
according to MOFED estimates) (Behnke 2010: 7). In low-income countries, livestock
has broader social significance than as a financial asset or food source, and represents
the basis of livelihood and cultural identity (Economist at Large 2011).
The number of slaughtered animals has been very low, and insufficient to supply
tanneries. The off-take rate for Ethiopian livestock was also low, 40 per cent for
sheep, 27 per cent for goats, and 14 per cent for cattle (MOA 2012). According to the
CSA Agricultural Sample Survey, 61 per cent of skins and 48 per cent of hides were
not marketed but used within producer households (2012a). Moreover, due to
traditional animal husbandry practices, the estimated reproductive rate is 37 per cent
(which means 28 million sheep produce only 9 million lambs), one of the lowest in
the world. At the other end of the spectrum are New Zealand, Australia, and the UK,
with a reproductive rate of 150 per cent. Improving the agricultural extension system
and providing a stronger incentive package is essential, but not sufficient. This has to
be supported by other interventions, such as specialized infrastructure development.
Further, the quality of raw hides and skins has deteriorated in the last twenty years.
Although Ethiopian highland sheepskins (from so-called Abyssinian sheep) are of
good quality, this quality is undermined by traditional slaughter methods and poor
handling of raw hides and skins during collection. According to ELIA, in the 1980s
and early 1990s 50 per cent of tanneries received Grades 1–3, a result of the
compulsory dipping of animals, as opposed to 20 per cent in 2012. Studies show that
the major reason for the defects is ectoparasites, also colloquially known as
‘cockle’ (p.220) or ‘Ekek’ (USAID 2008; MOA 2012; ELIA 2012). This widespread
infestation, which can be controlled through appropriate veterinary services, damages
skins and undermines the effects of improved rearing practices. These findings are
consistent with interview responses by Ministry of Agriculture officials, tanneries,
and leather products firms. Although traceability is difficult because of the inter-
regional movement of animals and skins, sheepskins from Gojjam and Gondar are
held to be of better quality and sell at a premium, while poor quality skins from
Wollo sell at a lower price.
Poor livestock health, estimated to cost more than $2 billion per year in SSA in the
1980s, is a major constraint and requires government to take an active lead (de Haan
and Umali 1994). Unfortunately, veterinary services and agricultural extension
services in Ethiopia are poor and neglected (Embassy of Japan 2008; USAID 2008;
UNIDO 2005), a situation that has been highlighted in various Ethiopian government
documents (CSA 2012a; MOA 2012; MOI 2012a, MOFED 2010; see also Little et
al. 2010 and Mahmoud 2010). Studies and pilot projects have been conducted by
various organizations, including USAID and Pittards, which show that the technology
is available to improve skin quality by up to 80 per cent, and that reproductive
performance can be immensely improved.
The introduction and expansion of ranches should be explored as a complementary
policy with medium-term relevance for the meat-processing industry and supply of
raw hides and skins. There is limited experience of ranches in Africa, although they
are common in Latin America. The type of ranches is related to political
factors.6 This policy option was, however, not considered in the development of the
livestock sector in Ethiopia, possibly to avoid the risks that smallholder farmers
might face by becoming more dependent on livestock.
Currently, the livestock sector is viewed as supplementary to crop production by
smallholder farmers, rather than as a sustainable source of livelihood (which also
contributes to structural transformation and industrialization). The necessary
interventions, from livestock development to the leather products industry, require a
coherent and integrated approach and high priority in terms of resource allocation.
Evidence suggests that (p.221)

Figure 6.6. Export comparisons: live animals, leather, and meat processing, 1998–
2011
Source: Own computation based on ERCA (2012a)
although there was a clear strategic articulation of this priority, no comprehensive
policy approaches were formulated.
There is also a lack of compatibility in policies. As previously mentioned, one major
policy contradiction regards exports of live animals, which grew faster than meat
exports. The former’s share of export earnings increased from a negligible 0.16 per
cent to 6.57 per cent in fourteen years (see Figure 6.6). However, domestic
slaughterhouses and meat exporters are on the verge of closure due to high prices and
lack of supply. In November 2012, the association of meat processing plants and
abattoirs requested the government to ban the export of live animals (MOI 2012a).
The structural problems related to smallholder livestock management need to be
addressed with more effective and coherent policy instruments. Exploring the
experiences of Brazil, Botswana, and South Africa can help in addressing these
constraints. For instance, in Botswana there are still some 700 ranches, but their share
of the national herd (3 million in 2010) has dropped from 30 to 10 per cent because of
low returns. Feedlot operators face similar problems (FAO/GOB 2013). Ranches in
some African countries have been increasingly constrained by overgrazing and lack
of harmonization with pastoralist community norms (Behnke and Scoones 1992; Ash,
Bellamy, and Stockwell 1994). In Ethiopia as well, (p.222) periodic environmental
disasters erode livestock numbers and productivity, and the effects of overgrazing are
considered to be immense (Hardin 1968; Economists at Large 2011; Aklilu and
Wekesa 2002). It is noteworthy that there is no inherent conflict between continued
smallholder production and the development of, or experiments with, larger scale
arrangements such as ranches: the answer is not ‘either/or’, but leveraging
complementarities.
To conclude, the meat and meat processing industry and the supply of hides and skins
are complementary and could have induced the rapid development of a new industry.
This is a typical example of Hirschmann’s ‘neglected problem’:
…I have distinguished between privileged and neglected problems. I
defined as ‘privileged’ those problems whose victims have adequate access
to the policymakers so that the latter are obliged to pay attention, for the
sake of political stability in general and of their own political survival in
particular. ‘Neglected’ problems, on the other hand, do not enjoy this direct
access, but they can be brought to the attention of policymakers in various
indirect ways…How privileged a problem is depends on answers to such
questions as: how numerous and how concentrated are the problem’s
victims, how important is the issue to them, and how much influence do
they have? (Hirschman 1981: 150)

6.4.3 Political economy of value-addition and industrial upgrading

An important illustration of the policymaking process is the experience of trying to
promote greater value addition in leather. Upgrading may refer to process upgrading,
product upgrading, or chain upgrading. Studies suggest that product and process
upgrading can happen within the GVC, including under multinational corporations.
However, it is unlikely that chain upgrading can develop under the existing global
governance system for value chains (Humphrey and Schmitz 2004; Gereffi et al. 2005;
and Kaplinsky and Morris 2000). This governance system mirrors the relative power
of actors and their ability to assert their interests, and it is the global retail chains that
are at the apex of the industry, with their reach extending right down to the tanneries.
The value addition is closely associated with the technological level of the industry.
For instance, Italy’s leather and leather products industry is much more advanced
than Kenya’s or Ethiopia’s (see Table 6.4).
Ethiopian raw hides and skins were exported during the imperial period and under the
Derg (1974–91) before being banned in 1983. After 1983, (p.223)
Table 6.4. Value chain comparator of Ethiopia, Kenya, and Italy
Comparative factors Kenya Ethiopia Italy

Availability of raw hides and skins Abundant Abundant Low

Quality of raw hides and skins Generally Low–high High

poor

Sustained capital investment Low Low High

Technological sophistication of facilities and Low–medium Low–medium Very high

equipment

Process skills Limited Limited Very high

R&D Limited Limited Very high

Product development Limited Limited Very high

Unique skills within the sector Rare Rare High

Degree of vertical integration Low Low High

Product perception by the global market Poor Poor (high for Very high
sheepskin)

Source: Adapted from UNECA and AU (2013)

wet-blue and pickle hides and skins dominated exports, predominantly to Italy, where
they were processed into finished leather. The export of pickle and wet-blue hides
and skins was prohibited in 2007. Exportable products then had to be processed up to
the higher crust level. In September 2008, Proclamation No. 567 imposed an export
tax on enterprises ‘exporting hides and skins without adding significant value’
(FDRE 2008a). The tax rate was 150 per cent of the value of raw hides and skins and
5–20 per cent for wet-blue and pickle. In November 2011, MOFED Directive No. 30
increased the rate to 150 per cent of the value of all raw hides and skins and semi-
finished leather. It was only in December 2011 that crust exports were also
discouraged through a high export tariff (MOFED 2011b). At present, exportable
products need to be processed up to the level of finished leather to avoid these steep
taxes. MOI states that thorough discussion and consultation with the industry
(through ELIA) preceded the decision on crust exports. However, although it may
seem straightforward to induce domestic value addition in this sector, in fact several
fundamental changes are required.
Upgrading within the semi-processed stage is important, but the key leap occurs in
upgrading to finished leather. This requires upgrading of the industry in terms of
skills, technology, quality and supply of inputs, market channels, and of the
mentalities of firms and industrial actors. The decision to ban semi-processed leather
exports was discussed at NECC meetings from 2005 onwards. For instance, at its
eighteenth and twenty-second meetings, it was decided that tanneries would not
receive incentives and access to loans if they did not produce finished leather. At
the (p.224)

Figure 6.7. Monthly exports of leather sector ($ million), July 2011–November 2012
Source: Own computation based on LIDI data, 20 December 2012
thirty-first meeting, it was also decided to tax tanneries that did not start producing
finished leather.
The final decision to ban crust exports was made in August 2011, with December
2011 given as the cut-off date. When firms realized this decision was irrevocable,
they maximized their crust exports between September and November 2011,
temporarily increasing export revenues (Figure 6.7). On the other hand, prices of raw
hides and skins had been increasing rapidly even before, but now skyrocketed. After
January 2012, however, finished leather exports dropped for many months, reflecting
the industry’s lack of readiness to make such exports. Many of the preparations were
incomplete, and a comprehensive package was not yet in place. Many in the industry
also resisted the change. In short, although the policy was important, its
implementation resulted in resource wastage and havoc.

Developing the Leather Products Industry

Another aspect of upgrading is development of the leather products industry. In the
early 1990s, there were only two footwear firms. As NECC meeting minutes reveal,
the government was committed to building up the leather products industry by
attracting new actors to the sector (p.225) (mainly after 2008) and by developing the
capacity of domestic footwear firms through a benchmarking programme. Among the
foreign firms attracted, a few (for instance, ARA from Germany and Huajian from
China) were industry leaders. Since 2008, government had made special efforts to
attract such firms, and has since cleared some of the obstacles in their way.
Between 1992 and 2001, most investment was in tanneries (91.5 per cent). From
2002 to 2011, investment trends and numbers of firms changed. Up to 2006, 50 per
cent of firms in the industry invested in leather products. This number peaked at 106
firms between 2007 and 2012, accounting for 80 per cent of all investment. This
clearly demonstrates the focused investment promotion of the leather products sub-
industry. Indeed, NECC minutes reveal that of the 348 issues discussed after 2004, 90
per cent focused on leather products, as did thirty-five of its thirty-nine (on average)
annual decisions. Although gradual and limited, value addition has increased recently.
In the tanning sub-industry, the value of local raw materials and semi-processed
products showed a bigger increase than imported inputs.

Views of Tanneries and Leather Products Firms on the Ban of Crust Exports
It should be clear that the ban of crust exports affected different actors differently and
their responses mirror their divergent interests (UNIDO 2003). During interviews, it
became apparent that many foreign-owned tannery firms that also owned tanneries in
their home countries preferred to export crust and add value at home. Their interests
did not match those of the Ethiopian government. Many of the domestic firms
disliked the decision because their interlocking interests with buyers (foreign
tanneries) were jeopardized and they were unable to compete in the changed
circumstances. The majority of the firms stated that the preparation time was too
short, and only half indicated that the policy has had positive effects on tanneries. By
contrast, 88 per cent of the leather products firms maintained that the policy had
positive effects, as it allowed them to buy finished leather on more favourable terms.
This may have been based on the assumption that the tanneries would be unable to
meet export standards of finished leather. In fact, the price of finished leather
continued to rise between 2012 and 2013, thanks to the continuing mismatch between
supply and demand.

(p.226) Views of Firms on Temporary Ban on Licensing New Tanneries

In response to the shortage of raw hides and skins, domestic tanneries put increasing
pressure on government to ban new investment in tanneries. Although MOI did make
this decision, the Leather Industry Development Institute (LIDI) did not agree on the
need for it. LIDI was concerned that such a ban would reduce domestic competition,
affecting the leather products industry. In this it appears that LIDI was right, as the
decision mainly benefited existing tanneries. Government could have provided
information on the circumstances in the industry and left the decision on whether to
invest to firms. This decision also undermines the expansion of capacity for future
growth, in response to supply once the input problem is addressed. On the effects on
firms, more than 70 per cent of tanneries agreed it was positive, as did 53 per cent of
leather products firms. Experience elsewhere shows that the outcome of such
decisions is contingent on specific conditions (for instance, domestic competition and
processing capacity). It is evident that the policy decision was not based on a
comprehensive sectoral study.
In conclusion, the policy decisions seem to have had some positive effects, in that
exports of finished leather gradually increased. Nevertheless, the process was painful
in terms of reduced capacity for almost a year, and the finished leather capacity still
does not match the soaked capacity. This has two policy implications: industrial
policies could work in this sector, but success is not automatic. It entails extensive
planning, monitoring, negotiation, sanctioning, as well as temporary pain. Moreover,
such a policy would have been more successful if it had been supported by richer
analysis of the sector and a comprehensive package with a longer term perspective at
a much earlier stage.

6.4.4 Binding constraint: Logistics and trade facilitation

Particularly for landlocked countries, international trade requires efficient logistics
and trade facilitation. This is very much the case for Ethiopia. World Bank studies
point out that the logistics and trade facilitation constraint has become more
significant in the last few years. For instance, customs and trade facilitation is the
major problem encountered by Chinese FDI in Ethiopia (World Bank 2012a). Other
major obstacles are trade regulation, tax administration, access to finance, tax
rates, (p.227) macroeconomic instability, labour regulation, and electricity. According
to Doing Business 2011, 2012, and 2013 (World Bank and IFC 2011, 2013), Ethiopia
was ranked 152 in 2009, 157 in 2010, and 161 in 2013 in cross-border trade.
Arguably, compared with Ethiopia, other countries have made more improvements
that are appropriate. In the survey responses, the firms not surprisingly identified
import–export logistics and trade facilitation as major constraints. In relation to
export delivery times, 55 per cent stated that logistics were the foremost constraint.
In short, external trade is much more costly in Ethiopia than in many countries, and it
takes much longer to import and export. One twenty-foot container costs $2,160 to
export and $2,660 to import. In terms of delivery times, it takes forty-two days on
average to export and forty-seven days to import such a container. The current
industrial norm for delivery of footwear and other leather products is less than forty-
five days from the date of order. This clearly highlights the challenge. Due to intense
competition and the dominance of GVCs, an export market has to be carved out in
conditions of stiff competition from Asian manufacturers. The most successful
exporting economies (Korea, Germany, and Singapore) have the shortest delivery
times and lowest costs (as low as four days and $439 per container). Examples of
developing countries include Vietnam (twenty-one days and $600), and Egypt
(twelve days and $625/755). It is also believed that inland costs can be significant in
landlocked countries, depending on mode of transport.
Experience elsewhere shows that successful logistics and trade facilitation policies
involve an integrated or multi-modal transport system, economical rail transport,
enhanced competition among logistics providers, and harmonization of customs and
transport among neighbouring countries. Single windows linked electronically,
automation, and factoring risks have also been adopted in many countries. The key
ingredient in this exercise is effective institutions.
Many efforts have been made by government to improve logistics. These include
establishment of dry ports, a multi-modal transport system in 2011, industrial zones,
and economic operators authorized by the customs authority. Not all these efforts
have been effective, and logistical obstacles remain. According to many in the
industry, the failed introduction of the multi-modal transport system had adverse
effects on manufacturing firms. This system, found to be effective in many countries,
was unsuccessfully implemented in Ethiopia. It proved (p.228) impossible to trace
containers, and shipments took up to four months (double the earlier period). The
new system failed from the very beginning for many reasons. First, it was designed
without sufficient knowledge or study of fundamental issues such as the problems
associated with inland freight transport and port related aspects. Second, the
implementation agency was a merger—still incomplete—of three SOEs: ESL
(shipper), Ethiopian Maritime and Transit Service, and the Ethiopian Dry Port
Authority. Moreover, none of these individual entities had good track records.
Third, there was no proper piloting of a complex system involving new technology,
which would have enabled officials to learn by experimenting and partial execution,
and nor was there sufficient consultation with customers, industrialists, and other
actors. For successful policy implementation, timing is important. For instance, it has
been argued that the system’s introduction should have been linked to the
commencement of the new Djibouti–Addis rail corridor. Fourth, hiring experienced
partners or consultants or adopting a management contract was not considered. Fifth,
in the middle of the project, all transit and clearing agents were forbidden to pursue
their activities. This may have reflected the government’s broader economic and
political concerns about middlemen, market intermediaries, and traders.
A related issue was the monopoly rights granted in May 2000 to ESL to ship all
import cargoes, except where the line does not have the capacity or serve the route.
This directive, written in the form of a letter, has significant policy implications.
First, while ESL was granted a monopoly, no government office was named to
regulate the company. This flies in the face of Stiglitz’s argument in the mid-1990s
that privatization should not proceed without prior establishment of effective
regulatory bodies (Stiglitz 1998). Second, ESL was insulated from competition, and
the outcome was negative. Even after enjoying a monopoly for twelve years, ESL
was still uncompetitive and weak in capacity. This was in stark contrast to state-
owned EAL, which had to contend with stiff international competition. EAL has
grown and expanded to become one of the leading airlines in Africa. Third,
manufacturers and exporters experienced delayed deliveries, as ESL had a limited
number of vessels capable of calling all ports of origin. Such delays resulted in
missed orders and increased costs, eroding the leather/leather products industries’
already thin margins.

(p.229) 6.5 Policy instruments in leather and related industry

This section discusses direct policy instruments, including export and investment
promotion, industrial financing, and privatization.

6.5.1 Investment and export promotion instruments

Export Promotion Schemes

The export promotion schemes include foreign exchange retention for all exporters,
and other manufacturer-targeted schemes such as duty-drawbacks, vouchers, and
bonded warehouses. Of these, only foreign exchange retention seems to have been
remotely effective. This is confirmed by 73 per cent of the firms surveyed, which also
reported no major implementation problems. By contrast, 85 per cent of firms stated
that the voucher system was ineffective and entailed enormous bureaucratic
difficulties. According to the unpublished report of the Taskforce on Export
Promotion Incentives Review, 187 firms were registered as beneficiaries of the
voucher system, of which fifty-two (28 per cent) are in floriculture, forty-two (23 per
cent) in leather and leather products, thirty-eight (20 per cent) in textiles/garments,
and twenty-eight (15 per cent) in agro-processing. Only forty-two firms (about 60 per
cent of all firms in the leather sector) were registered as beneficiaries, showing the
low interest in the scheme by manufacturers in this sector. Even those which were
registered did not benefit from the scheme. Duty-drawbacks and bonded warehouses
were also ineffective. The reasons for these failures lie in the weak design of the
instruments and low implementation capacity of government agencies.

Exchange Rate Policy

Ethiopia devalued its currency in 2010–11 by about 25 per cent. The government has
used exchange rate policy to promote exports, and, as already noted, leather/leather
goods were prioritized for export. In this case, the policy clearly failed, as the
interviews with firms and export data for 2005–11 confirm. Only 43 per cent of
survey respondents indicated that devaluation had a positive impact, in contrast to
two-thirds of floriculture firms. This deviates from the expectation that devaluation
should have a significant positive effect on exporters. A possible explanation is that
most of the firms failed to work at full capacity and mainly sold (p.230)
Table 6.5. Domestic-owned firms’ share of export and local sales (by volume), 2005–11
Year Export sales (in %) Domestic sales (in %)

2005 95.73 4.27

2006 97.41 2.59

2007 71.23 28.77

2008 50.52 49.48

2009 67.71 32.29

2010 62.14 37.86

2011 62.47 37.53

Source: Oqubay (2012)

locally, resulting in smaller export earnings. Firms are increasingly producing for the
domestic market, which is more profitable. For instance, the quasi-census results
show that the share of exports in total sales decreased by 33 per cent between 2005
and 2011 (Table 6.5). This implies that the policy instruments and incentives have not
been calibrated to influence the behaviour of industrialists, and to link export
promotion with exploitation of the domestic market.

Investment Promotion Instruments

Initially investment incentives were designed to attract any type of new investment, a
practice that continued until 2012. The incentives were neither designed to monitor
reciprocity, nor to attract the best industrial performers, who would advance
Ethiopia’s technological and marketing base. The total number of investment
certificates issued between 1992 and 2012 was 220. However, of those registered
between 2003 and 2007, only 12 per cent became operational, 7 per cent were at the
implementation phase, while 81 per cent were still in the pre-implementation phase.
Investment increased gradually after 2002, peaking after 2007. Between 1992 and
2001, more than 86 per cent of investors were domestic, and they were primarily
interested in tanneries. The primary reason for investing in this sub-sector was its
familiarity in Ethiopia (see Table 6.6).
After 2005, the government began to target investment promotion. Of the thirty-one
investment certificates provided to FDI, 68 per cent were registered between 2007
and 2011. Moreover, increased labour costs in many emerging economies (including
China) may have served as a pull factor. There was also an investment shift away
from tanneries, which had dominated in the 1990s, towards leather products after
2007 (see (p.231)
Table 6.6. Domestic-owned firms’ reasons for investing
Reasons Ranked first

No. of Share in
firms %

Natural resources 16 30.2

Experience and knowledge of the sector 14 26.4

Family background 12 22.6

Cheap labour 5 9.4

Growing domestic market 2 3.8

Others: Availability of incentives, political stability, cheap land, ease of 4 7.6

access to USA/Europe market

Total 53 100
Source: Oqubay (2012)
Table 6.7. Investment certificates in leather and leather products (1992–2011)
Period All Average Foreign firms Tanneries sub- Leather products
firms firms/Year (FDI) sector sub-industry

Firms In % Firms In % Firms In %

1992–2001 (10 47 4.7 4 13 43 91.5 4 8.5

years)

2002–2006 (5 40 8 6 19 20 50 20 50
years)

2007–2011 (5 133 26.6 21 68 27 20 106 80

years)

Total 1992– 220 11 31 100 90 41 130 59

2011(20 years)

Source: Own computation based on unpublished FIA data (July 2012)

Table 6.7). Consequently, the share of investment in leather products increased to 80
per cent. In terms of geographical distribution, more than 50 per cent of the
investments targeted Addis Ababa and 34 per cent the Oromia region.
Despite the increased number of investments, the administration was cumbersome
and there were difficulties in acquiring land and other services. Profit-tax holidays
were easier to implement, as the firms confirmed. More than 55 per cent of firms
complained that customs duties and land deliveries were problematic, causing delay
and additional costs. Two-thirds of the firms complained about poor coordination
among federal agencies as well as among federal, regional, and local administrations.

6.5.2 Industrial financing in leather and leather goods

DBE has provided industrial financing to various industries in the manufacturing
sector (DBE 2012a, 2012d). Major beneficiaries were textiles and (p.232)
Table 6.8. Summary of DBE’s industrial financing (in ETB million)
Sector Number of Beneficiaries in Average loan Total Loans in
loans % amount portfolio %

Textile 32 10 156 4,995 46.2

Non-metallic mineral 16 5 238 3,816 35.2

Food processing and 221 71 4 869 8.1

beverage
Sector Number of Beneficiaries in Average loan Total Loans in
loans % amount portfolio %

Chemicals and chemical 22 7 30 669 6.2

products

Leather and leather 21 7 22 465 4.3

products

Total manufacturing 312 100 35 10,814 100

sector

Source: DBE (2012d)

garments (46 per cent share), non-metallic minerals (35 per cent), and food
processing and chemicals (close to 15 per cent). With a share of 4.3 per cent of the
total loans to manufacturing, less than half a billion ETB went to leather and leather
products (Table 6.8). The average size of loans to firms in this sector was ETB 22
million, while the average for the manufacturing sector was ETB 35 million. The
limited share and loan size for leather and leather products was partly a function of
the lower investment requirements, but does also raise the important question of
whether the sector got what it required to fulfil its export and employment creation
potential.
Eighty-eight per cent of firms in the sector indicated that inadequate credit facilities
posed major challenges, and pointed to the gap between working capital requirements
and available industrial financing. Working capital plays an important role in this
sector, as firms are usually forced to carry a large inventory. This is due to the high
material intensity of the sector, inefficient trade and logistical infrastructures, and
inefficient management within firms. The period for holding stock and inventory is
double that of the overall manufacturing sector (CSA 2012c). The leather and leather
products sector’s poor performance may have discouraged banks from extending
loans to the sector. For instance, in an interview with the author in 2012
(Oqubay 2012), a senior CBE official, it was indicated that the bank was unwilling to
provide working capital because the firms would not repay within the stipulated
twelve months. Thus, financing policy seems to have been ineffective in promoting
this sector. The leather and leather products industry was also unable to benefit from
the export credit guarantee scheme (Table 6.9). (p.233)
Table 6.9. Export Credit Guarantees, 1 September–30 November 2011
Commodities or products No. of loans Amount (ETB million) Share %

Oil seeds 17 316 61

Cereals pulses 6 88 17

Textile and garments 4 64 12

Commodities or products No. of loans Amount (ETB million) Share %

Livestock 3 37 7

Gum, incense, bee products 2 14 3

Total 32 519 100

Source: DBE (2012d)

6.5.3 Privatization in leather sector

Another government policy in the leather and leather products industry was
privatization. All nine large state-owned firms were privatized and remain in
operation. The previous Derg regime had made major investments to expand the
capacities of these firms. Privatization in this sector was implemented over fifteen
years, and was guided by the government’s pragmatic interests rather than ideology.
Domestic firms were given priority and more favourable loans (the central bank’s
saving rate). Accordingly, three-quarters of the firms went to domestic buyers
(PPESA 2012). Valuations were revised when the exaggerated estimates became less
attractive to investors. There is no evidence that privatized factories were transferred
to interest groups connected to the ruling party. Even the regional endowment funds
were not among the buyers: for instance, EFFORT established a new tannery in
Tigray under the name Sheba Tannery. In addition, the government used two transfer
modalities: most of the firms were sold by open tender, while the biggest tannery,
Ethiopian Tannery, was transferred to Pittards through direct negotiated sale. Pittards,
which initially opted for contract management, improved productivity, made an
additional investment, and strengthened production management. In 2011, it also
established a glove factory, with a second leather products factory following in 2012.
Between 2007 and 2011, additional investment of ETB 234 million was made by the
new owners to equip the privatized firms. Capacity utilization of privatized tanneries
and footwear factories stood at 69 per cent and 56 per cent respectively in 2011,
which were not below the industry’s low average capacity utilization. This does not
allow us to conclude that all privatized firms have improved their performance. For
that, a more detailed study is required. Measuring the outcomes of privatization
raises (p.234)
Table 6.10. Major problems of privatized firms
Problems Percentage

Shortage, high price, poor quality of raw 49

materials

Shortage of skilled personnel and capacity 20

Shortage of finance 9
Problems Percentage

Logistics, customs, and trade facilitation 13

Others 9

Total 100

Source: Oqubay (2012)

several methodological issues, including differentiating direct and indirect factors.
Some of the privatized firms were still among the best performing top five exporters
in the sector in 2011. Privatized firms faced problems similar to those faced by other
firms (see Table 6.10).

6.5.4 Summary
In sum, first, the policy instruments were only partly successful and were not
comparable in scale to the support provided to the cement or floriculture sectors.
Second, many of these instruments suffered from design deficiencies, insufficient
implementation capacity, as well as coordination problems. Third, the firms in this
sector played a largely passive and insignificant role in the process, an issue with
deeper implications.

6.6 Drivers of transformative change: specialized institutions

This section focuses on three relevant institutions, namely export coordination
institutions, the lead organization for the leather and leather products industry, and
intermediary institutions.

6.6.1 Export coordination and NECC

Some of the most significant NECC decisions included the decision to privatize state-
owned tanneries (May 2004) and that to prohibit new tanneries from receiving
government incentives (February 2006). Other equally important decisions were, for
instance, the one stipulating that tanneries be supported in producing finished leather
(September 2005) and that DBE loan allocations be guided by this decision and not
be made (p.235) to new investors in tanneries (October 2005). Moreover, in January
2007, it was decided to impose the new tax on tanneries that did not start producing
finished products. However, in May 2007, it was decided to encourage Turkish
investors in the tanning sub-industry, thereby reflecting inconsistency in policy
decisions. There were clearly other policy inconsistencies, including the decisions to
export live animals while also importing raw hides and skins to alleviate shortages.
Similar decisions were repeatedly made, suggesting that government institutions were
complacent and unresponsive, or lacked accountability or capacity.
Although NECC did treat this sector as a priority, there were problems in the
decision-making process. First, reports to government lacked depth and a strategic
analytical perspective, and were limited to tackling immediate constraints.
Consequently, policies and decisions were not well informed. For instance,
discussions on input constraints were not based on detailed study and a
comprehensive approach to the development of this strategic sector. Second,
decisions were not effectively implemented. Third, the producers, enterprise owners,
and trade unions were not engaged in the decision-making process, an approach that
exacerbated the information gaps and divergences in the implementation of decisions.
Moreover, an opportunity to build trust between state and the industry was missed.
Government should play a role in promoting active private sector involvement, as the
industrialization process depends on such activism.

6.6.2 LIDI as lead agency

The lead agency for the leather and leather products industry is LIDI. Its precursor,
the Leather and Leather Products Technology Institute, had been established in 1998
by the Council of Ministers. Its main tasks focused on training and information
services and improved productivity and quality standards. In addition to the resources
allocated by the Ethiopian government, the Italian government, the Common Market
for Eastern and Southern Africa’s Leather and Leather Products Institute, and UNIDO
provided technical assistance and financial support. Building of the Leather and
Leather Products Technology Institute facilities took more than five years. In June
2010, this institute was re-established with expanded responsibilities and duties as
LIDI. The new organization’s mandate included:
(p.236) …to study and recommend policies to the government; conduct
research; promote and support investment in leather and leather products
sector; provide training services; enhance technological and know-how
transfer; provide laboratory services; assist in market development;
promote input-output linkages; provide design and product development
services; network with all potential stakeholders and institutions. (FDRE,
Council of Ministers, Regulation No. 181, 2010)

Led by a director-general, there were directorates for leather manufacturing

technology; footwear manufacturing technology; leather garments and goods
manufacturing technology; market support; research; and investment support (project
engineering). The institute has 265 personnel, almost equally divided between line
and support staff. Among line staff, 64 per cent were professionals and 32 per cent
had technical certificates and diplomas. Among the technologists and technicians, 34
per cent were in leather manufacturing, 31 per cent in footwear manufacturing, and
34 per cent in garment and goods manufacturing.
LIDI is equipped with state-of-the-art technology and modern facilities. The latter
included models of a tannery, a footwear factory, and a leather garment and goods
factory. It also has an effluent treatment plant, fully equipped laboratories, a
computer-aided design/manufacture centre, a library, computer facilities, and training
halls. LIDI’s main deficiency was its lack of capacity to use these facilities to support
the industry. In 2011, government made an agreement with two Indian institutes, the
Central Leather Research Institute to support the leather sector, and the FDDI to
support the footwear and leather products industry. There is evidence that the
twinning arrangement is contributing to improvement in LIDI’s capacities and those
of the leather and leather products industry (LIDI 2012a, 2012b, 2012c).
Eighty-four per cent of firms agreed that LIDI has made a valuable contribution in
supporting the sector, and affirm that the institute works closely with industry (86 per
cent). Eighty-one per cent of firms agreed that the institute focuses more on
supporting firms than on regulation, a higher rating than that for EHDA in
floriculture. Evidence does not suggest that learning had been rapid or to the required
level. Lack of capacity remains LIDI’s single greatest challenge (see Table 6.11).
Moreover, coordination among government institutions was weak, contributing to
LIDI’s ineffectiveness. According to the survey results, 69 per cent of the firms view
coordination among federal government institutions as (p.237)
Table 6.11. Firms’ response to ‘What is the key limitation of LIDI?’
Key limitations Per cent

Lack of skilled personnel 24.6

Lack of implementation power 22.9

Lack of focus on domestic market 13.1

Poor training support 9.8

Limited mandate 9.8

Lack of technological up gradation 8.2

Others 11.5

Total 100

Source: Oqubay (2012)

poor, while 60 per cent regard coordination among federal, regional, and local
administrations in the same light.
A review of firms’ requests to LIDI highlights the coordination challenge. A review
of 316 letters (LIDI 2012e, 2012f) showed that 40 per cent of those from foreign-
owned firms related to permit requests for expatriates from the labour affairs office.
Delegating authority to provide these permits to LIDI on behalf of the labour office
would have eased this problem. Domestically owned firms’ most frequent requests
(37 per cent) related to sending sample shoes or raw skins overseas, which should not
have required such permission to begin with. Again, a simple directive would have
solved this problem. This also shows how administrative hurdles that should never
have arisen in the first place can overwhelm an agency, weakening its focus on key
aspects of the industrial policies to promote the development of the sector.
The data also show that these problems became more frequent over time (eight in
2010, fifty-five in 2011, eighty-eight in 2012). Usually two to four weeks are lost
because of such administrative bottlenecks, significantly delaying delivery times and
increasing costs. This conclusion is also supported by the author’s survey results
(Oqubay 2012): the majority of firms indicated that they spend significant
management time on government-related issues, and even more than firms in
floriculture and cement (Table 6.12).
This shows the failure of such government agencies as MOI and LIDI in providing
coordinated and effective support. During interviews, most firms regarded LIDI and
the ministry as ‘toothless lions’. This in turn shows the limitations of NECC in
guiding the parties and enforcing decisions. It also confirms that logistics and trade
facilitation, as well as government bureaucratic procedures, continued to be critical
constraints. (p.238)
Table 6.12. Time spent on government-related issues
Management time (%) Floriculture Leather Cement Total

Firms In % Firms In % Firms In % Firms In %

≤15 19 35 19 34 5 31 43 34

16–30 22 41 23 41 6 38 51 41

>30 13 24 14 25 5 31 32 25

Total 54 100 56 100 16 100 126 100

Source: Oqubay (2012)

In the long-term, it is perhaps the competitiveness of domestically owned firms that is
most affected by such constraints.

6.6.3 Effectiveness of ELIA as intermediary institution

ELIA is the industrial association of the sector and the main intermediary institution.
It was founded in 1994 as Ethiopian Tanners Association and reorganized itself in
2004 into Ethiopian Tanners, Footwear and Leather Products Manufacturing
Association. In 2004, the association had twenty members, mainly tanneries. In 2007,
it became ELIA and had forty-four members, more than 60 per cent of which were
tanneries. This group continues to dominate the association. By contrast, in Vietnam
it is the leather products (footwear, bags etc.) manufacturers that dominate the
national industrial association.
Twelve of the fifty-five survey respondents were not members. Although ELIA’s
capacity mirrors the capacity of its members, responding firms admit the association
is ineffective. They said it played a weak role in market development (61 per cent), in
target setting and monitoring (78 per cent), and in investment promotion (66 per
cent). Regarding its role in training and knowledge transfer, 90 per cent of the firms
gave it a poor rating. Less than 60 per cent were satisfied with ELIA’s role in policy
development and government lobbying (see Table 6.13).
Although members rate the dialogue between government and the industry as
satisfactory, they recommended more discussion forums, improved participation by
stakeholders, and further government support. Many of these views, however, are
based on exaggerated expectations concerning government’s role. Significant parts of
the industry are comfortably locked into the status quo and are unwilling to engage.
Sixty-four (p.239)
Table 6.13. Views on ELIA’s role
Degree of Lobbying Market Training/kn Target Policy Investment
ELIA’s role government development ow-how setting/moni initiations promotions
transfer toring

No. % No. % No. % No. % No. % No. %

Excellent 1 2 – – – – – – 4 9 2 5

Satisfactory 25 58 17 39 4 10 9 22 23 52 11 29

Poor 17 40 27 61 36 90 32 78 17 39 25 66

Total 43 100 44 100 40 100 41 100 44 100 38 100

Note: No. represents the number of firms

Source: Oqubay (2012)
per cent of firms admitted that the association’s members do not play an active role.
Furthermore, the association does not include recent entrants into the industry,
including foreign-owned firms. In addition, hides and skins suppliers are not part of
ELIA or any other association, which further compounds the challenges. All of this
indicates that government has failed to address important institutional aspects of the
sector.

6.7 Conclusions
The disappointing performance of the leather and leather products sector in Ethiopia
is a reminder that industrial policy can fail. A number of conclusions stand out from
the research into this industry. First, in terms of policy outcomes, sector performance
was characterized by erratic and sluggish growth throughout. Compared to cement or
floriculture, performance was very disappointing. Notably, massive animal resources
were not productively used. The evidence strongly suggests that comparative
advantage in natural endowments does not automatically lead to sustained
competitive advantage. Moreover, potential linkage effects are converted into actual
linkage promotion through effective policies, which were lacking. Nonetheless, the
sector is not quite the abject failure it is sometimes made out to be. In recent years,
there have been signs of recovery and positive initiatives such as benchmarking.
Policies did achieve some value-adding in existing tanneries (from semi-finished to
finished) and in attracting new entrants into the leather products sub-sector.
Furthermore, it should also be emphasized that one of the most (p.240)important
macroeconomic conditions for maximizing linkages may be ensuring a competitive
exchange rate.
Second, in terms of policymaking and policy capabilities, it is evident that policy
improvements and huge investment in institutional development were made. The
policies and institutions rightly identified the key constraints: inputs and logistics, for
instance. Nevertheless, they could not address them due to problems in
operationalizing policies and, in some cases, inappropriate policy choices. The
institutional deficit in input development (livestock development and marketing chain
for raw materials) was a significant failure. In fairness, dealing with small, scattered
producers/input suppliers is always more difficult than dealing with a few ‘modern’
enterprises, such as in the cement industry.
Third, the main economic actors in the sector are not well organized and have
conflicting interests. As a result, they played a much less prominent part than in other
sectors, contributing to the industry’s slow growth and lack of competitiveness.
Unlike the intermediary institution in floriculture, ELIA played an insignificant role,
and lacked internal cohesion and a jointly developed agenda. Firms suffered from
inertia, preferring to limit themselves to the domestic market and shy away from
exports. This also shows that the incentives partly failed to induce the required
behaviour and export discipline. Thus, government–industry dialogue forums need to
be institutionalized more effectively.
Finally, it is noteworthy that this experience is associated with path dependence in the
leather and leather products industry. Path dependence influences and shapes
understanding, problem solving, policy design, and implementation. Instituting
desired changes in a long-established sector with an accumulated network of interests
is more difficult than in newer sectors. Path dependence in this old sector at least
partly explains the long history of stagnation, the limited deepening of the value
added, the inertia among actors and their intermediary institutions, and the
ineffectiveness of policies. Operationalization of policy was insufficient or policies
were followed for far too long. There are two possible policy responses to change the
sector. One is to allow foreign investment in tanneries; the other is to continue to
protect them but to introduce a reciprocal control mechanism, with clear targets as the
condition for the protection.
Notes:
(1) For example ITC (2012).

(2) CSA list of firms involved in 2012 survey of manufacturing was referred to
investigate active firms.
(3) The global leather market showed sustained growth between 1981 and 2006, from
$13 to $60 billion, but was slowed by the 2008 global economic crisis (FAO 2008). Of
the total sales, the hides and skins segment, tanning, and leather footwear accounted
for 12, 30, and 58 per cent respectively in 2007 (FAO 2008).
(4) Brazil has one of the fastest growing leather goods sectors in terms of exports and
production. Brazil’s export earnings increased from $700 million in 2000 to more
than $2.2 billion (352,222 tons) in 2011. This shows that this is not a ‘demand
constrained’ sector, but one in which new competitors have expanded to take larger
slices of the global market.
(5) The decision was also partly influenced by the Derg’s need to produce boots for
its soldiers under extreme balance of payments constraints.
(6) See Mwangi (2007) on the political factors that transformed property rights and the
ranch system.

Failing Better
Political Economy and Industrial Policy in Ethiopia
Chapter:
(p.241) 7 Failing Better
Source:
Made in Africa
Author(s):

Arkebe Oqubay
Publisher:
Oxford University Press

DOI:10.1093/acprof:oso/9780198739890.003.0007
What are the insights that emerge from the comparative analysis of the three sectors
in previous chapters? Synthesizing the research material on Ethiopian manufacturing
and industrial policy, the chapter explores uneven growth and unevenly effective
policies despite a common foundation in industrial development strategy and
industrial policy. It shows the limitations of conventional views in explaining this
dilemma, and provides an alternative perspective. This chapter shows that overall,
what matters for the evolution and effectiveness of industrial policy is the way that
three key factors—industrial structure, linkage dynamics, and politics/political
economy—interact. The interrelational dynamics among these variables have
significant implications for policy design and implementation. And from its emphasis
on ‘failing better ’, it draws out and highlights areas with significant implications for
policymaking. This chapter also shows that industrial policymaking in Ethiopia is a
work in progress, and shows the challenges of catching up in twenty-first-century
Africa.
Keywords: Ethiopia, comparative analysis, uneven growth, linkages, industrial
structure, politics, political economy, failing better

7.1 Comparative performance

7.1.1 Performance, policy learning, and Ethiopian ‘anti-fragility’

Without wishing to provoke charges of policymaking hubris, I believe there is
evidence of ‘anti-fragility’ in Ethiopian industrial policy. Anti-fragility is the term
coined by Nicolas Taleb (2012) to describe the ability of a system to be strengthened
by stress rather than collapsing under its weight (fragility). This is distinct from
robustness, defined by Taleb as the ability to fend off threats and remain unchanged.
To understand the sources of anti-fragility in policymaking, one must appreciate
policy processes based on learning-by-doing and adaptability, characteristics rarely
explored in studies of policy and performance in developing countries.1 Through
sector case studies, previous chapters have shown examples of these policy dynamics.
It is also important to understand the conditions that allow for anti-fragility rather
than robustness or fragility. In Ethiopia, these conditions seem to include the ruling
party’s experiences as a (p.242) liberation movement and in the protracted military
struggle to oust the Derg regime. In the post-liberation period, the EPRDF has
survived under enormous external threats; succeeded in developing a cohesive
ideology and mechanisms for resolving internal differences; often revitalized itself
after internal crises; and blended experiment, pragmatism, and long-term vision and
principles. By contrast, many other successful liberation movements have
degenerated after assuming power, often plunging their countries into destructive
violence (see EPRDF 2011b; Young 1997; Clapham 2009; Tareke 1990, 1991).
At first glance, it is easy to identify successes and failures in the case studies explored
earlier in this book, but this is not, per se, a study about success or failure, a too
simplistic binary evaluative framework to be useful. At the very least, the evidence in
preceding chapters shows how frail the categories of success or failure can be when
applied on an aggregated sector-wide basis. For example, despite disappointing
overall performance in leather and leather products, there have nonetheless been
important advances in this sector from which lessons may be learnt (for instance,
value addition). Floriculture appears to be an obvious and clear success (UNIDO-
UNCTAD 2011: 63), yet there are areas of failure and emerging challenges that
necessitate new policy responses. The emphasis is better placed on challenges and
problems and the logic behind them, an approach that is also a healthy safeguard
against self-congratulation. While success brings new problems and challenges,
problems have the potential to generate new and unexpected opportunities.
Reinvigorating growth ultimately depends on leveraging learning in policymaking,
and having the political resolve to take decisive action and ensure coordination
among multiple actors.
Earlier chapters have shown that the outcomes of Ethiopian industrial policy have
been distinctly uneven. This is a puzzle for policymakers and scholars alike and
requires explanation. As the discussions in Chapters 4–6 illustrate, floriculture,
cement, and leather and leather products feature prominently in national development
and industrial development strategies. Likewise, the broader context has affected
general conditions in each of these sectors in similar ways. This book has argued that
the Ethiopian state has played an important developmental role within a market
economy, a role that goes beyond merely ‘facilitating’ comparative advantage, as has
been advocated by some. The country’s political landscape is shaped by a federal and
nascent multiparty system, and also the continuing (p.243) dominance of the ruling
party since 1992. These factors are believed to have been favourable to the growth of
all three sectors, and for the first time in its history, the country has witnessed a
decade of rapid economic growth. What has not thus far been explained is why,
within this broader context, there has been such marked variation in performance.
This chapter provides a comparative overview of the three sectors before outlining
such explanations as have been put forward. These are more or less standard
explanations—that outcomes must be a function of comparative advantage; are a
product of neopatrimonialism; or are a product of a ‘discovery process’. While not
without some value, none of these explanations is sufficiently convincing empirically
or logically. Instead, the chapter goes on to argue that the Ethiopian experience of
policy and performance is better understood in light of Albert Hirschman’s insights
into ‘linkages’ and the unevenness of developmental processes. From this
perspective, bottlenecks and constraints can be creative. As Samuel Beckett put it
in Worstward Ho: ‘Ever tried. Ever failed. No matter. Try again. Fail again. Fail
better.’2

7.1.2 Comparative analysis of the three sectors

Performance across the three sectors reveals several facts. First, sectoral
performance was markedly uneven: disappointing and erratic growth in the leather
and leather-working industry, and higher growth in both the cement and floriculture
sectors. This can be seen in the comparative growth in production, employment, and
exports (see Figure 7.1).
Second, output growth was much stronger after 2002 than between 1992 and 2002.
This coincides with and may to a significant extent be attributed to the government’s
policy learning and the refinement of its development strategies and industrial
policies after 2003. This in turn signals the importance of a relatively stable political
settlement since the removal of the Derg in the early 1990s. It has to be noted that
this initial phase was also characterized by economic transition from a command
economy to a free market economy, the imperatives of postwar reconstruction with
exceedingly scarce resources, and political reform. This transition served as a
springboard for economic stability and recovery

Figure 7.1. Comparative performance in three sectors (2003–12)

Source: Summarized from previous chapters (all in per cent)
(p.244) after the late 1990s. It is also evident that growth in post-1991 Ethiopia was
much faster than in pre-1991 Ethiopia. Learning by doing is as much a feature of
policymaking as of a firm’s management and production techniques. Arguably, a key
feature of the overall policymaking context (and its effectiveness) was the
combination of relative political stability/continuity and a sharp political shock. On
one hand, political stability and continuity since 1991 created an environment
conducive to learning by doing, similar to that allowed by industrial protection. On
the other, the 2005 elections produced a ‘shock’. The unexpected political challenge
in 2005 acted almost as a ‘hiding hand’ by prompting a renewed developmental
imperative and refreshing the coherence of policy commitments.
The significance of policy learning by doing or an evolving adaptive capacity
(Giezen et al. 2014) is not commonly emphasized in the literature on the political
economy of industrial policy in Africa. But in another respect, the Ethiopian evidence
does confirm the findings of Whitfield (2011) and Buur et al. (2012), who argue that
political continuity and coalition stability matter greatly for the outcomes of
productive strategies. Whitfield (2011), for example, argues that support for Ghana’s
potentially strong horticultural export sector was undermined by the incentive
structure built into the political settlement after the advent of democracy. Effectively,
elites were not compelled to support new productive sectors, however great their
long-term potential: on one hand, emerging horticultural producers were not strong
enough to represent a threat; on the other, (p.245) the returns to government would
not flow quickly enough to warrant state support. In Mozambique, where there has
been greater political and party continuity since independence, there are nonetheless
factional divisions that have undermined the coherence and stability of state support
for particular sectors (Buur et al. 2012), although in other sectors, such as sugar,
support has been more sustained and effective.
The record in Ethiopia suggests that a development-oriented state indeed performed
better over time than states without an activist industrial policy and that its industrial
policy has contributed to higher growth. The facts regarding the policy measures,
institutional changes, and sectoral performance underpin this assertion. The important
implication is that an appropriate industrial policy can work even in a very poor
African country like Ethiopia. Moreover, this growth and industrial expansion took
place in a landlocked economy in a ‘bad neighbourhood’, thereby distinguishing the
experience even more clearly from the predictions and typologies of some influential
analysts (Collier 2006, 2007, 2009; Collier and Gunning 1999; Easterly 2002; Easterly
and Levine 1997; Fosu 1992, 2012).
Third, despite higher growth during this period (1992–2013), the outcome in terms of
structural transformation was underwhelming, indicating the limitations of industrial
policy. The share of manufacturing in the economy and in export earnings remained
low throughout. This reflects an inherent weakness in current industrial and other
related policies in Ethiopia, which, as noted in earlier chapters, can be contrasted with
the policies adopted to transform Korea economically from 1960 to 1980 and China
between 1985 and 2005. The implication is that industrial policy capability in
Ethiopia remains rudimentary, and needs further development. Nonetheless, the
relatively strong performance of the overall economy and some parts of industrial
production underscores the importance of the Ethiopian government’s efforts to
maintain policy independence, since the policies pursued often represented a direct
challenge to Bretton Woods prescriptions, prescriptions routinely followed by most
African countries.

7.2 Conventional arguments and standard explanations

7.2.1 Comparative advantage

One influential explanation views factor endowments (particularly cheap labour and
natural resources) as the most influential determinants of the (p.246) success of
industrial policy. A World Bank publication (2012), based on this approach, tried to
demonstrate how light manufacturing could be promoted in Africa, by taking
Ethiopia as one of the African case studies, with Vietnam as a comparator, and China
as benchmark. Justin Lin, former chief economist of the World Bank, champions this
approach (Lin and Chang 2009). The key argument is that countries such as Ethiopia
can successfully industrialize if they stick to their comparative advantages and focus
on light manufacturing, such as leather and apparel, and on taking the prescribed
actions to address key constraints identified in the value chain. This approach fails to
consider the political economy determinants of industrialization, such as the nature
and role of the state and the role of policies in determining or avoiding path
dependency. It falls into the all too common trap of casting economic development as
a largely technical challenge. Buur et al. (2012) and others in the Elites, Production
and Poverty programme at the Danish Institute for International Studies provide one
alternative exploration of productive success and failure that is much more realistic in
its appreciation of the political economy of productive expansion (or stagnation).
While factor endowments are clearly important, there are difficulties with a policy
approach that relies exclusively on them. According to Lall (1999: 9) ‘the traditional
theories of comparative advantage that are noted as the foundations of government
policy are based on highly simplified models of “perfect” markets with no scale
economies, costless and full information flows, no risk and so on.’ Not only are the
assumptions unrealistic, often in the extreme, but many regression analyses provide
no convincing empirical evidence for the postulates of comparative advantage theory.
Indeed, as has been noted, one renowned macroeconomist, Greg Mankiw ‘was not
surprised by the lack of robustness in the cross country results, given the large
number of candidate variables relative to the number of country observations. He
conjectured that economists support free trade because they believe Ricardo, not
because they have been convinced by regressions.’3In other words, there is a strong
element of faith among economists in Ricardian comparative advantage theory, but
this does not make it the best guide for policy. The assumptions in the theory are a
gross oversimplification of real (p.247) economies that are characterized by
ubiquitous market and information ‘imperfections’ (as they appear, at least, to
neoclassical economics). These circumstances in turn require active industrial
policies in order to accelerate industrialization and structural transformation of
economies. However, industrial policies should be based on market realities, rather
than on abstract models of ‘perfection’. Korean and Taiwanese economic
transformations, for example, were possible because they successfully moved up the
ladder from simple products to advanced skills and technology using active industrial
policies, rather than merely specializing in their traditional comparative advantages
(Lall 1999; Chang 1999; Singh 2011).
Factor endowment provides an inadequate explanation of the puzzle at hand for many
reasons, and factor endowments alone do not account for the success or failure of
industrial performance across sectors in Ethiopia. The comparative advantage
argument might predict that Ethiopia would be competitive in basic leather
production but not necessarily in leather products (in particular, higher-end gloves,
accessories, etc.), or in high productivity flowers. This is because some versions of
comparative advantage thinking (Owens and Wood 1997) look more at the ratio of
skills to cultivable land—evidently higher in Holland (a high-productivity flower
producer) or in Italy’s leather products sector. Yet there has been a sharp rise in
productivity and skills in some of these sectors in Ethiopia, whereas even in the more
obvious area of basic production of leather from hides and skins, Ethiopian
performance has historically been very poor (under a variety of different policy
regimes). Further, there is plenty of evidence from a range of countries of their
accelerating their growth and structural change by challenging factor endowment
constraints.
Likewise, growth in the Ethiopian cement industry did not depend on factor
endowments. The cement industry is a capital-intensive heavy or basic industry, with
large economies of scale and dependence on professional management. In Ethiopia, it
is also a strategic import-substitution industry that was driven by a fast-growing
domestic market. It was neither a light industry nor labour-intensive, and was not
based on the country’s endowment of cheap labour or natural resources. Further,
expansion in this sector was dominated by domestically owned firms and indeed by
state-owned and EIGs, thereby deviating from international trends in which cement is
typically dominated by massive global firms. This evidence suggests that the advice
to sub-Saharan Africa to focus (p.248) on light industry and to avoid building up such
capital-intensive industry is based on unwarranted assumptions.
Of course, there is some common sense in adopting the ‘Ricardian’ comparative
advantage approach, at least partially, as almost all developing countries have done at
some stage of their development. A ‘Ricardian’ strategy essentially follows and
supports existing comparative advantage. It effectively means, for a low-income
country, specializing in the export of primary commodities and perhaps some light
manufacturing that is technologically undemanding and requires little skilled labour.
Such a strategy typically will be supported by a ‘light touch’ state and by liberalized
trade, which indeed will provide the price signals to illuminate comparative
advantage. Nevertheless, historical facts also show that ‘Ricardian’ strategies can
create self-sustaining growth only by shifting to a second type of strategy, namely, the
‘Kaldorian’ strategy (Schwartz 2010).4 A ‘Kaldorian’ strategy is more complicated
and prioritizes the expansion of manufacturing, given the idea that manufacturing
plays a unique role in the growth and structural change process. Such a strategy also
rests on the premise that it is not possible to accelerate industrialization and to speed
up the gains from economies of scale and learning-by-doing in manufacturing
without protective policies.
In the Ethiopian context, static comparative advantage was not the key driver of
industrial policy or performance in any of the three sectors featured in this book. For
instance, despite abundant Ethiopian livestock resources, the leather and leather
products industry has remained stagnant: this suggests a form of impotent or
unrealized comparative advantage. The government has long failed to upgrade the
industry or attract dynamic private investors into it; and the strategy has lacked
coherence and remained largely ineffective. By contrast, leather industries in Brazil,
China, and Italy have continued to play a major economic role, and not without
significant state intervention to support their growth or, in the case of Brazil,
resurgence. In Ethiopia, the sector has continued to be uncompetitive internationally.
Even cheap labour has not helped reverse the industry’s stagnant and falling
productivity, though the latter may have begun to change with recent inward
investment by a major Chinese women’s shoe producer, as well as other foreign
investors.
(p.249) By the same token, in floriculture, despite the existence of natural
endowments (water, soil, land, climate, relative market proximity, etc.), the industry
did not flourish until the early 2000s. What facilitated the rise of this new industry
were the government’s readiness to ‘pick’ this industry (and to allow it to be ‘picked’
by entrepreneurs in line with the government’s overall development strategy) and to
implement appropriate policy and policy instruments, combined with the readiness of
investors (foreign and national) to participate in this industry.
These examples also suggest another area where political economy, rather than
abstract technical economic principles, is critical. As Buur et al. (2012) argue in regard
to sectoral variation in Mozambique, one factor that matters is sustained political
support and another is the organizational capacity within a given sector or industry.
Further, foreign investment can raise such organizational capacity, as in the recent
cases of foreign investment in the Ethiopian leather sector. The flower sector is a very
good example of both domestic sector organization and a huge contribution to this by
foreign investors.

7.2.2 Patronage and neopatrimonialism

Another version of industrial policymaking is propounded by the neopatrimonial
school. The patronage and neopatrimonial interpretation has enjoyed much currency
in the donor community and neoliberal circles since the early 1990s and comes in
several shades. A key inspiration in this interpretation is Weber’s identification of
three ‘ideal’ or ‘pure’ types of legitimacy, namely traditional, charismatic, and legal.
According to Clapham’s (1985: 48) widely used definition, ‘neopatrimonialism is a
form of organization in which relationships of broadly patrimonial type pervade a
political and administrative system which is formally constructed on rational-legal
lines.’ Thus neopatrimonialism is a system of government that is a mix of the
rational-legal Weberian-type state, with separation of powers and impersonal rules,
and surviving personal relations that guide decisions, resource allocation,
appointments, etc. Weber (1947) applied the term to traditional authority in ‘the
Orient, Near East, and Medieval Europe’. However, on the basis of Clapham’s
definition and personal observation of political processes and behaviour in advanced
industrial countries, including the US, UK, and France, neopatrimonialism is a
globally pervasive and enduring feature of governance. This would seem (p.250) to
reduce even further the concept’s analytical value for understanding African
economic policy and performance.
However, it is to precisely the enduring and pervasive ‘chronic failure’ of economic
growth in Africa until recently that current proponents of neopatrimonialism apply
the concept. According to this pessimistic view, there is no foundation or scope for
developmental states in Africa. Industrial policy, this school holds, cannot succeed in
Africa, since it will only reinforce, and be confounded by, neopatrimonialism.
According to Altenburg (2011: 8):
Industrial policy plays an important role in stabilizing neopatrimonialism as
it creates political space for politicians and bureaucrats to allocate
government resources to specific groups of beneficiaries. These can be
employed to strengthen ties of loyalty between individual politicians or
bureaucrats and private beneficiaries, but also to buy political support from
specific social and ethnic groups that are considered important for the
survival of the incumbent regime.…Consequently, politicians and
bureaucrats who want to employ industrial policy for patronage and
clientelism can easily find technical justification to mask their political
objectives.

Successes in some African countries are seen as exceptional or unsustainable, and

generate a supplementary literature on ‘developmental neopatrimonialism’. Kelsall et
al. (2010: 28) argue that ‘centralized, long-horizon rent-seeking, when combined with
broadly pro-capitalist policies, can generate dynamic growth’.
Industrial policy is thus viewed as a vehicle to benefit party members and supporters;
party-affiliated businesses, such as endowment funds; and specific ethnic groups.
Ethiopian government programmes such as the micro and small enterprises
development programme are seen as a channel for consolidating the ruling party’s
position. Influential NGOs, the donor community, political parties, and vocal
dissidents very much subscribe to this view (Kelsall 2013). Nonetheless, no strong
theoretical grounds or empirical evidence are adduced to explain how this is different
from other political systems, for instance in the UK and US.
Conceptually, the neopatrimonial view has fundamental flaws and cannot offer a solid
analytical approach. First, it reflects Afro-pessimism and is imbued with an
entrenched prejudice that views African culture or Africans as incapable of making
progress (Mkandawire 2001; Padayachee 2010). All ills are attributed to the failures of
African leaders or societies. Consequently, it is blind to country-specific peculiarities
and the diversity among (and within) African countries. Yet, historical evidence
shows that (p.251) many African countries had the vision and capacity to secure high
economic growth, for instance, during ‘the golden period of Africa from 1960 to
1980’ (Soludo et al. 2004). Even after the 1980s, Mauritius and Botswana succeeded
in sustaining growth, and since 2000 a dozen countries, including Ethiopia, have
achieved high economic growth.
Second, it is a static, simplistic view that ignores how economic and political forces
are in constant motion, and the complexity of policymaking. According to Ottaway
(2003), neopatrimonialism is ‘an ill-defined code word for the political ills that afflict
the continent’, and for Clapham (1996: 820) ‘…explanation in terms of culture should
be regarded as deeply offensive’. Mkandawire further refutes the notion in conceptual
and analytical terms.
[a]…while providing descriptions of the styles of exercise of authority…
the concept has little analytical content and no predictive value with respect
to economic policy and performance…[b] Economic policymaking is a
highly complex process involving ideas, interests, economic forces and
structures, ‘path dependence’ and institutions, and cannot be reductively
derived from the ‘logic of neopatrimonialism’. It will require serious
attempts to understand the ideas, interests and structures, which shape or
hinder Africa’s development efforts. The neopatrimonial approach is too
simplistic and too formulaic to help in understanding the complex drama of
development taking place in Africa.

Furthermore, Altenburg (2011) relied on a limited number of selective interviews,

which may not lead to objective observations. For instance, individuals do not—as
Altenburg suggests—own endowment funds, and the law on endowment funds does
not allow resource channelling to political parties. The presence of political leaders
on the boards of state-owned or endowment fund–owned firms does not necessarily
support this conclusion. Moreover, loans have been available to most floriculture
firms, including foreign-owned firms (some of whom have ‘over-picked’ the state).
The beneficiaries of the biggest DBE loans were private firms with no political ties or
connections, including foreign firms. If supporting the chaebols in Korea was not
neopatrimonialism, how can support for regional, publicly owned endowment funds
in Ethiopia be counted as such? The evidence on the allocation and performance of
concessionary loans to the floriculture sector, where there was at first little or no
discipline (Amsden’s ‘reciprocal control mechanism’) on borrowers, suggests, if
anything, naivety on the part of state agencies rather than calculated patrimonial
misallocation.
(p.252) Despite neopatrimonial assertions, state-owned enterprises have also not
necessarily or uniformly nurtured predatory groups and clienteles. The Mugher
Cement Enterprise was a pioneer and played a leading role in the cement industry. As
we have seen, other state-owned enterprises such as EAL and DBE have served as
key vehicles of industrialization and change in the political economy of Ethiopia.
Indeed, international evidence suggests that ownership—state versus private—is not
the key factor in enterprise performance, and that, if anything, state-owned
enterprises have most often been fundamental to industrial catch up (Musacchio and
Lazzarini 2014a, 2014b).5 Support to the private sector in Ethiopia’s priority industries
does not demonstrably produce the patronage predicted by neopatrimonial theory.
Again, there is nothing exceptional in the blurred lines between government-
controlled state enterprises and private sector enterprises. As Milhaupt and Zheng
(2014: 8) argue: ‘The…response to the 2008–2009 financial crisis in the United States
illustrates that even in countries where private ownership of enterprise has strong
ideological and historical roots, the boundary between government control and
private control’ can be blurry. In addition, the focus of state-owned and endowment
fund enterprises on large-scale manufacturing and strategic projects shows their
developmental role in overcoming ‘market failures’. Getting prices and property
rights wrong (Amsden 1989) has, as the evidence suggests, been critical to the, albeit
inchoate, process of structural change in Ethiopia. Endowment funds made critical
investments—and took or socialized fundamental risks—in cement, beverages,
leather and leather products, textiles and garments, and transport. In sum, claims by
proponents of the neopatrimonalist approach appear in large part to be unfounded and
not empirically grounded.

7.2.3 Limits of ‘industrial policy as a process of discovery’

Another approach to industrial policy and performance that has had some influence in
recent years is ‘industrial policy as discovery process’. This hypothesis would
attribute success in the floriculture sector to collective (p.253) learning and the
discovery process. Industrialization is very much about learning. And the idea of a
discovery process captures something relevant, something not dissimilar to the
emphasis in this book on adaptation, on Hirschman-like dynamics of creative
problem-solving. This is, therefore, a relevant and interesting perspective, though one
that like so many mainstream perspectives seems strangely unaware of a whole
tradition of development economics and engagement with industrial policy. At the
same time, it seems to be an over-simplistic, descriptive, and post hoc ergo propter
hoc approach. The level of private sector development, existing and potential tensions
between conflicting interests, the risk of political capture by sector interests, etc., are
important variables that are given little attention in this kind of approach. The
predictions of this framework are also slightly fatalistic, calling for a less proactive or
more passive industrial policy until after a new and promising industry is discovered.
This is fully consistent with the pessimism Rodrik (2014) has expressed elsewhere: ‘I
come down on the pessimistic side, due to what I think are poor prospects for
industrialization.’6 But if everything boils down to a mysterious process of discovery,
how are policymakers to operate? Should they simply sit and wait for discoveries to
unfold?
This so-called discovery framework also fails to consider that the private sector may
not, left to itself, always represent the best interests of the broader economy. Its
entrepreneurs may be inclined to make ‘discoveries’ that are not really in the interests
of overall transformation or welfare improvements or, without appropriate
government intervention, may be a realm of rentier opportunity (Mazzucato 2013a). It
is more likely that the private sector focuses on short-term returns rather than the
long-term interests of industry; just as, on the flipside, governments may be so beset
with factional divisions or frequent democratic turnover they too focus excessively on
the short run and fail to ‘facilitate’ investors in promising new sectors
(Whitfield 2011).
The nascent private sector in Ethiopia has in the past demonstrated a tendency to
focus on short-termism and on the temptations of engaging in speculative activities
rather than productive sectors, and to prefer light industry over riskier intermediate or
heavy industry. This should not be taken as evidence of uninformed judgements or
hasty decisions. (p.254) Rather, it is a product of their experience and resources, and
of prevailing incentive structures. Given the high risks involved in investing in the
manufacturing sector, and high rents in services and trading, there is nothing ‘wrong’
or surprising in this preference. Specific sectoral features and, indeed, broader
political economy factors also shape interests and investment patterns. Unlike in
floriculture, the government was the pioneer and key player in the development of the
cement industry throughout its initiation and growth stages (through its state-owned
enterprise). The government gradually induced the private sector to play an active
role in the industry, through the demonstration effect of the state-owned enterprise
and the inducement or linkage effect of the rapidly expanding market for cement
(principally in the construction sector). The government had to ‘show’ the sectoral
prospects to private sector investors rather than wait for them to ‘discover’ these for
themselves. In a nutshell, the narrative in the cement industry shows that its
development required an active and leading role by government, despite the prompts
and hints given by an industrial strategy that favoured large-scale infrastructure
investments, and by the rapid growth of the construction sector.
Although industrial policymaking is indeed a learning process, and there is a need for
state and private sector to communicate (and often industrialists will know better), the
‘discovery process’ approach (Rodrik 2004) reduces the process to an unrealistic fable
whose moral is that everyone should be nice and listen to each other, the typical
charm of liberal assumptions. Nor is it always just a matter of the kind of creative
imbalance that Hirschman (1958) emphasized, though that is important: it is also a
matter of conflicting interests and tensions. Often these conflicts are too intense and
the scarcity of resources compounds the tension. Hence, states have sometimes to be
more interventionist. The South Korean state, in the period of early rapid
industrialization, was far from ‘nice’ and infamously put people in jail when they did
not perform. Although this might not be a ‘lesson’ to encourage, it is an indication of
the intensity of conflicts of interest. For instance, it is precisely in the leather and
leather goods sector where these conflicts have created an obstacle to sectoral
expansion and transformation.
The ‘discovery process’ interpretation was also based, in its application to Ethiopia,
on biased and incorrect evidence, as is shown in earlier chapters. What mattered in
floriculture was not dialogue per se, but the (p.255) decisive government response to
the constraints of the industry. The government took swift political decisions to make
land available, to make loans accessible, and to ensure that air transport and logistics
infrastructure were provided. In view of the resource constraints and political
sensitivity, the state showed its capability to manage rents, to socialize risk, and to
make critical political choices. There was almost inevitably some corruption and
wastage in this process, and some individuals had clearly taken advantage of
incentives without adequate ‘reciprocal control mechanisms’ in place, especially
early on. But overall, the evidence suggests a net gain in terms of foreign exchange
earnings, employment creation, structural changes, and (unevenly) the development
of enterprise governance capabilities. The government’s readiness to listen to the
private sector was an outcome of its strong political commitment and strategic
orientation to develop this sector. Land provision was a highly contested political
issue, and the political decision to subsidize air transport at a time when government
had withdrawn subsidies on kerosene and fertilizer was potentially dangerous, as
were concessional loans in the context of very scarce credit. These key decisions
could not have been made but for the state-owned development bank, the state-owned
air transport firm, and publicly owned land (including former state-owned farms).
These issues say a lot about a developmental state and the policy independence of the
government, rather than its capture by patronage. The backdrop was that in contrast
to many countries in sub-Saharan Africa, the Ethiopian government resisted
privatizing the development bank and held land under public ownership, despite
pressure from international financial institutions and strong neoliberal prescriptions
(what Joseph Stiglitz calls ‘market fundamentalism’). These were not primarily
matters arising out of a dialogue process, but rather out of political constraints and
political economy.
In addition, the frequent narrative is that the private sector pioneered the floriculture
industry. This narrative is associated with the implicit assumption that foreign
investors were the key drivers of this sector. This is far from the truth. For instance,
the state explicitly encouraged the development of sectoral organizational capacity by
advising the two Ethiopian pioneers to establish their association EHPEA in order to
overcome obstacles and voice their interests. EHPEA is now one of the most effective
intermediary institutions, and an example to other industries. (p.256)Therefore,
development of this sector was mutual and complementary, rather than one-
directional.

7.3 Linkage approach to policymaking

7.3.1 Unfolding of linkage effects in the three sectors

As discussed in previous chapters, from a linkage perspective the pace of
development in general and industrialization in particular depends on industries with
strong backward and forward linkages. What the evidence from the research suggests,
though, is that there is considerable variety in the way in which apparently obvious
‘pressure mechanisms’ or ‘pacing devices’ (Hirschman 1992) unfold, or fail to unfold.
There is nothing automatic about linkage effects. They depend on policy, but they
also depend on politics, and on the structure and particularities of specific sectors.

Linkages are Neither Homogeneous nor Automatic

The linkage approach should not be taken to mean that linkages seamlessly reveal
themselves (perhaps a version of the ‘discovery process’), automatically generating
new productive activities, and taking effect in a uniform way. Nor should it be
assumed they reveal themselves in similar shape and scale and with similar
dynamism. This is not the case in the real world, as the sectoral chapters show. In
fact, the diversity among linkage effects has been striking, and there was no instance
in this research where they repeated themselves or where they appeared in one form.
The first important distinction is whether the linkages emanate from inside the
industry or from outside. Some linkages may initially originate in another industry,
sector, or activity to create the conditions for the activity under observation. A typical
example is the backward linkage from the construction industry to the cement
industry that enticed the growth of cement production. In contrast, in floriculture, the
linkages were ‘transplanted’ from elsewhere (notably from the export market, not
from domestic market dynamism). There was clearly already a latent, readily
accessible source of autonomous export demand. But that of itself did not stimulate
investment. A combination of regional demonstration effect (Kenyan success),
exploratory contacts with potential investors, and a raft of policies was required to
generate momentum in the flower industry, which in (p.257) turn developed its own
logic of opportunity and linkage to other investors and to other related activities like
packaging and logistics.
A second aspect of linkage effects is that the direction and dynamism of the linkages
may change gradually. At the initial stage, the prime linkage for the leather sector
may have been the forward linkage from livestock, that is, raw hides and skins, to
tanning and leather manufacture. The push from this linkage became too weak, and
the linkage feedback loop shifted in the opposite direction from the leather sector
back to the development of raw skins and hides production, in short, a backward
linkage. But this linkage in turn only (and for a long time very slowly and fitfully)
unfolded because of the identification of the sector’s potential by successive
governments (supported by a string of external consultancy reports).
The third aspect is that exploitation of potential linkage effects depends not only on
policy design and execution, but also on politics. Arguably, one of the key lessons of
this study is that almost all linkage effects required government policy to be
effectively exploited, and the outcome mirrored the quality and effectiveness of the
adopted policy and institutional response. It is noteworthy that some policy responses
depended on the state’s posture on addressing the issue. The linkage to air transport
and cargo logistics in floriculture, the linkages to the energy sector, and those from
the construction industry to cement, and the critical financing of the expansion of all
three sectors by the state-owned development bank, were possible because of
the decision-making of the state(in the form of political imperatives and cohesion,
policy design, and institutional settings) at that particular time.
In some cases, such as the development of a packaging industry initially to service
floriculture, the key was the emergence of new entrepreneurs induced to invest by the
rapid growth of flower exports, but even they were supported by government
policies. Yet in the cement industry, manufacturers directly diversified into a
backward linkage to the manufacture of cement bags (economies of scope). In the
leather sector, the response to linkage possibilities was left to smallholder livestock
owners, but without fundamental structural and social transformation they were not in
a position to take advantage of the opportunities created by an emerging leather
sector. Developing a dependable input industry based on many and scattered
producers is perhaps more difficult than supporting a few modern manufacturers.

(p.258) Failures and Successes in Facilitating Linkage Mechanisms

It is therefore sensible to suggest that an important explanation for the variability in
policy outcomes was the differences in policy approach to developing the linkage
effects. The evidence so far shows the following:
1. a) In the floriculture sector, downstream (forward) linkages were more
important than upstream linkages. The industry’s growth would have been
arrested if the solutions in cold chain logistics and dependable air services were
not put in place. This binding constraint was removed primarily through state
support. EAL was there to fully exploit the opportunity created. It made
strategic moves including expensive aircraft purchases to modernize its fleet,
and subsidizing its services when necessary. The government also, through
EHDA, successfully supported the corrugated packaging industry, and
improved the local availability of inputs, thereby contributing to the sector’s
competitiveness. These efforts helped not only address binding constraints but
also to fully exploit opportunities to develop new industries. In addition,
government made land available and provided close to ETB 1.5 billion to the
majority of firms in the flower sector, including foreign firms. Even realizing a
competitive advantage based, in part, on factor endowments (land and agro-
climatic conditions) required political and policy intervention in the form of
 land policy, leases in particular. This was also symbolic in terms of
demonstrating the government’s commitment to the sector, as well its readiness
to welcome FDI.7
2. b) In the cement industry, the backward linkage (from the construction boom)
was strong and effective. The growth of the construction industry was itself an
outcome of government policy intervention, including macroeconomic policies
that produced sustained economic growth and programmes such as housing and
infrastructure development and capacity building in the construction industry.
Sustaining this rapid growth was important politically (the
2005 (p.259) election revealed narrow latitude for failure). This is similar to the
situation where internal and external threats provoked ‘political will’ in
industrialization policies in East Asia (Doner et al. 2005). Housing development
was the flagship programme in urban centres. This drive (with its direct effect
on cement demand and on political commitment to resolve key constraints of
the cement industry) was instrumental in the exponential growth of this sector.
Energy and freighting constraints were addressed through large-scale
government resource allocation to infrastructure development. The capital-
intensive character of the industry, and its scope for economies of scale, also
required government investment in the sector through the state-owned Mugher
Cement Enterprise.
3. c) The analysis of the leather and leather products sector clearly showed that
linkage effects are not always automatic or as compelling as they would appear
to be. The growth of the sector depended on a sustainable supply of cheap and
good-quality hides and skins, something that would appear possible given the
huge livestock numbers in Ethiopia. The strategy in this sector was built on
comparative advantage (cheap labour and livestock resources), and this sector
was singled out for promotion in the overall development strategy, i.e.
Agricultural development-led industrialization (ADLI) and Industrial
Development Strategy of Ethiopia (IDSE). Nevertheless, livestock resources
are scattered across the country and agricultural transformation is required to
meet the growing demand and requirements of the leather and related products
sector. Considering that livestock herds are dispersed among a large population
of often very low-income smallholder farmers, it may be a long time before
livestock production and husbandry is fully commercialized and transformed.
Traditional agriculture has diminishing returns, but the feedback loop from
leather manufacturing was too weak to exert the necessary pressure on
traditional farming. The government’s capacity to mobilize its own resources
(directly through the state-owned enterprises) in shaping this linkage was too
limited, which is surprising in light of the country’s ADLI strategy. Hirschman,
in The Strategy of Economic Development (1958), refers to backwardness in
agriculture by quoting Gerschenkron’s hypothesis that ‘the more backward it
is, the more the state intervention is (p.260) needed’. Yet, in Ethiopia, what has
happened is the reverse: the leather and leather products sector should have
enjoyed a much more forceful and effective industrial policy. This does not,
 however, preclude the possibility of promoting a more realistic policy option,
namely modern ranches and modern slaughter and meat-processing firms.

7.3.2 Linkages as coping mechanisms and policy learning

It is useful to draw out the parallels and direct connections between linkages, in the
sense of one economic activity ‘calling forth’ or making viable another, on the one
hand, and the policymaking and institutional process, on the other. The core of the
linkage and ‘unbalanced growth’ approach is the recognition that bottlenecks and
imbalances can be creative—they can induce recognition of a problem that requires
an innovative solution, and signal incentives to provide such a solution.8 However,
this does not always happen. When it does happen in economic activities, it is called
a linkage. In policymaking, this response is more about learning by doing, adaptation,
and so on. In addition, the learning and accumulation of policy experience can, of
course, be applied to other problems as they arise. It is difficult for outsiders to see
this clearly. Instead, they come as external experts (Hirschman’s ‘visiting economic
expert syndrome’) with blueprints, models, and best practices, which are often
limited and inappropriate, undermining the scope for institutional/policymaking
innovation (they may divert Hirschman’s ‘pressure mechanisms’ and clog up the
‘pacing devices’).
Clearly, this situation becomes very interesting where the institutional/policymaking
process takes place at the point where an economic linkage does or does not
materialize, or a sector does or does not undergo its own development and learning
by doing. One example is cheap loans to flower farms. Subsidized loans were made
available, with two negative results: some of them were diverted into very different
uses by some investors (possibly including transfer abroad through over-invoicing
and so on) and some of them were invested, but in unnecessarily expensive
equipment (p.261) and technology. Part of the problem lay in DBE’s lack of
appropriate and effective mechanisms either to choose between borrowers or to
monitor or discipline them. But it seems that there was at least partial learning here:
the development bank started imposing more ex ante discipline by, for instance,
forcing borrowers to put up cash rather than just a fence around the land. Monitoring
was also strengthened (involving joint action with EHDA, the lead government
agency in floriculture) and measures included transferring assets and loans to a third
party. This indicates that there has been adaptation, but only partially. One concludes
that a reciprocal control principle and monitoring/disciplining have not been fully
developed.
Another such example is the interactions between the EHPEA, EAL, the government,
and the massive Dutch-owned flower farm. Here, too, government adapted. At first,
government subsidized rent for no good reason or purpose, allowing the firm to
monopolize forwarding of output to the airport for other flower producers in Ziway.
However, there have been changes since then, including direct service provision to all
growers by EAL. Here, the question arises: to what extent and how is business
decision-making and ‘soft’ technology or knowledge of running a capitalist enterprise
developing among Ethiopian producers? And how is it spreading (if it is spreading)
from Ethiopians who play a lead role to others, and from foreign investors (such as
the relatively small French-owned Gallica or Sher-Ethiopia) to Ethiopian owners? In
addition, how does government influence this ‘soft’ technology transfer? Evidence
suggests that farm productivity has improved and the gap with Kenya is narrowing,
while changes in terms of product and market development and technological
advancement are insignificant. This requires further research.
Against this backdrop, it is possible to suggest that part of the explanation for the
failure in the leather and leather products industry was that government policy lacked
adequate appreciation of, or responsiveness, to the scope for and challenges to
stimulating linkages, despite the many studies conducted (for instance, UNIDO,
Japanese, and Cramer). The challenge of profoundly transforming agriculture (the
time needed, scale, and politics of intervention) was beyond the scope of the policy
response. This can also be associated with the view that agriculture, unlike
manufacturing, has fewer linkage effects. The challenge was compounded by the
institutional inertia of, in particular, the Ministry of Agriculture (p.262) and regional
extension services, and the intermediary institution, especially ELIA, the industrial
association for the leather industry, and the absence of institutions that equally
represented the interests of all firms and stakeholders in the input chain. In view of
the structural rigidities in agriculture, alternative solutions focused on development of
modern ranches and strengthening modern slaughterhouses might have helped
address input constraints.

7.3.3 Path dependence in leather and leather products

Lall (1999, 2003) emphasizes that learning is ‘path dependent’, and adds: ‘Once
embarked on, technological trajectories are difficult to change quickly, as
specialization patterns tend to persist over long periods.’ This is particularly true of
the Ethiopian leather and leather products industry. This industry has been
characterized by low-value products for many decades. Tanneries were dominant and
they have been interlocked in mutual interest with Italian tanneries, an unholy
alliance whereby Ethiopian tanneries supply semi-finished leather. This created a
comfort zone and a false shield for the sector. These circumstances were a reflection
of the long tradition of Ethiopian tanneries under the command economy, where
tanneries did not have to compete among themselves or with others. In the footwear
sub-sector, the moribund, stagnant productivity equilibrium was suddenly disturbed
by liberalization and the subsequent competition from imported goods and entry of
new firms. The sub-sector faced fierce competition from imported Chinese footwear
(from the mid-1990s to the mid-2000s in particular), and almost all the footwear
firms were on the verge of shutdown, as they were uncompetitive even in the
Ethiopian market. Most firms then embarked on investment in technological and
skills accumulation. The entry of many new firms has positively reinforced this
transition to a better path.
In the leather sub-sector, however, no such exogenous shock arose, as liberalization
did not automatically lead to the inflow of imported semi-processed hides and skins.
Even privatization of state-owned tanneries did not break old path dependence.
Consequently, the old equilibrium remained undisturbed. Since 2005, the tendency to
sell only to the domestic market and shy away from exports has become dominant in
the sector. Further, some key players in tanneries had backgrounds in the speculative
skins and hides trade. Hence, these tanneries did not face input shortages, (p.263) and
were less interested in fundamental transformation of the input chain. Tannery
interests dominated ELIA, making the association less receptive to new government
policies focused on exports. Efforts to deepen the industrial structure met passive
resistance, forcing the government to postpone many policies for a long time. In
addition, existing firms perceived new investments in the tannery sub-sector as a
threat, and lobbied policymakers concertedly and with some success. In this sector,
piecemeal intervention and changes that depended solely on existing actors did not
bring about the required change.
Different policy decisions may have had varying results in breaking path dependence
and the development of a new growth path. For instance, the government pursued and
strongly adhered to a policy of developing the leather products sub-sector (such as
footwear and garments). Leather products provide narrower scope for poor
performance, as their production forces firms to confront global competition. The
presence of new actors with such experience enhances the chances that interventions
will bear fruit, and contributes to breaking the inertia and forging a new path. This
may compel backward linkage activities, for instance, in the leather sub-sector. The
ban on exporting semi-finished leather and the requirement for increased value
addition may also have a transformative effect in breaking path dependence. In
addition, the import of inputs (raw skins and hides, semi-processed, and finished
leather) could have helped break path dependence.
Yet, the banning of new tanneries (despite the government’s good intentions to
strengthen domestic firms) may further solidify existing inertia, as it reduces
competition within the leather sub-sector and reduces pressure to transform the input
value chain. Arguably, given the strength of path dependence in this sector, the best
policy option would have been a targeted promotion of firms based on their past
successes, rather than on the origin of investment. Moreover, the introduction of
performance standards (including ex-post) could be more effective in channelling the
dynamics in the right direction.

7.4 Variations in industrial structure and policy implications

Industrial structure by itself is not everything. Nevertheless, the findings suggest that
industrial structure and industry-specific factors were (p.264) important for
performance variation and policy effectiveness across sectors. Knowledge of
industrial structure aids selectivity and targeting in industrial policies, guides
understanding of the roles and characteristics of economic actors, and highlights the
economic and technological characteristics and requirements of the specific industry
and the opportunities they afford. Industrial structure reveals the underlying forces
and basic characteristics of sectors, the ‘type of the economic activity defined by a
combination of capital intensity, economies of scale, production flexibilities, and
asset/factor flexibility’ (Evans 1997: 61–87). The three aspects that have significant
implications for industrial policymaking in Ethiopia are the characteristics of firms
and investors, the latitude for performance standards, and the stage of
industrialization.

7.4.1 Economic actors and policymaking

The characteristics and behaviour of firms and producers vary across sectors. For
instance, the cement and floriculture sectors have been dominated by a few medium
and large firms. This makes it easier for government to support and interact with
sector leaders. In the leather sector, actors are heterogeneous, with varying sectoral
interests: tanneries (mostly medium and large firms), leather products (dominated by
many small enterprises, in addition to medium and large firms), unorganized input
traders, and small, scattered skins and hides producers. These actors operate under
varied institutional settings, and often have conflicting interests.
Until late 2003, efforts to attract FDI were driven by the need for capital investment
and, hence, there was no selective sectoral targeting or stringent monitoring of the
process. There have been some shifts towards selectivity in the three sectors more
recently. In floriculture, investment promotion targeted industrial leaders, for instance
the Netherlands, which plays a dominant role internationally. The majority of foreign-
owned firms had the marketing and technological capability needed to secure
Ethiopia a foothold in the global market. The government arranged finance, which
added to its leverage on foreign investors. Local firms were supported with training
interventions (by government, the industrial association, and the Dutch embassy) that
facilitated expertise and skills transfer. This is an example of how aid, foreign policy,
and multinational interests partly converge; of how a bilateral state agency (the Dutch
embassy), acting primarily in the interests of Dutch capital, (p.265) smoothed entry
for Dutch firms, and subsequently provided assistance and technology transfers to
Ethiopia, probably beyond what Dutch firms themselves would have done; and at the
same time, an example of how multinationals depend on their country of origin to
promote their global reach. The foreign-owned floriculture firms were family-owned
and medium-sized, as well as quick to adapt. It appears that the political risks and
pressures that come with these firms have not been unmanageable. There have been
political challenges in the flower sector relating to compensation issues, the
environment, and labour, all of which required government responses. This situation
has shaped the pattern and momentum of the sector throughout its history (2004–12),
as has the active role played by industrialists (domestic and foreign) and their
intermediary institutions (the sector’s industrial association and the Dutch embassy).
The leather and leather products sector continued to be dominated by domestic firms
with institutional interests that created hostility to competition and a strong
dependency on government. Until after 2008, designs to target investment promotion
were inadequate. Domestic enterprise owners were less interested in seeing
significant FDI or in learning from counterparts. After 2008, the government took
important steps to engage firms such as Pittards (in 2009), Huajian (in 2011), ARA
(in 2010), and New Wing Addis (in 2010) through a managed privatization process
and promotion of FDI. Nonetheless, it appears that government acted only after a
long stalemate and when the damage had been done. In this industry, the policy
approach promoted inertia, while in floriculture it produced a dynamic state–
business partnership that enhanced collective action and collective learning.
In cement, state-owned Mugher, and later Messebo, played a leading role in terms of
taking risks and being first-movers. Others then followed, facilitating collective
action. Further, there is a lesser presence of FDI in the Ethiopian cement sector than
in the corresponding sectors in any other African and many other non-African
economies. This may well have been important to the domestic evolution of firm and
technology capabilities, but thus far has undoubtedly come at the cost of efficiency.
The point is that policy, and indeed the larger issue of industrial policy and the
‘developmental state’, is not independent of either industry structure or the pattern of
interests among investors and other relevant actors in specific sectors. Just as
‘governance capabilities’ may vary across (p.266) sectors within a country, so too will
the degree of ‘state autonomy’ vary across sectors and over time.
Finally, the industrial structure of the cement industry permitted and called for the
government’s direct role through state-owned enterprises. This has allowed the sector
to benefit from the critical role played by the state. This was not the case with the
other two sectors, which were more favourable to small and medium, family-owned
firms. The role given to and played by state-owned enterprises in Ethiopia has been
driven by the belief that they are essential to industrialization, which requires an
activist role by the state. The neoclassical view emphasizes that markets would
function efficiently and that state intervention is bad. For instance, according to
World Bank (2013), ‘Private sector-led activities appear to be constrained by the
policy choices favouring public investment’. This view rejects the state’s role in
direct economic activities through large state-owned enterprises (World Bank 1995;
Shirley 1997) and holds that the state should confine itself to protecting private
property rights and reducing transaction costs (North 1990). Consequently,
privatization of state-owned enterprises has been included as a key component of the
Washington Consensus (Williamson 2008) and its successors. The neoclassical view
fails to see that markets are neither always superior to states, nor immune themselves
from politics (Stein 2006; Chang 2006). Bureaucratic decision-making in state-owned
enterprises is wrongly blamed, although bureaucracy is a feature of all large firms,
private or state-owned (Chang and Singh 1997; Chandler 2004).
State-owned enterprises are founded with multiple aims, for instance, economic
transformation or catch up (Amsden 1989; Chang 1994). There are also no grounds for
assuming that state-owned enterprise performance is weak because of the size of the
enterprise or state ownership (Chang and Singh 1997; Jalilian and Weiss 1997).
Indeed, the research does not support the argument that privately owned firms are
more efficient than comparable state-owned enterprises (Mühlenkamp 2013). There is
certainly huge variation in the performance of state-owned enterprises, which is still
poorly understood (Musacchio and Lazzarini 2014a, 2014b), but so is there variation
among private firms. The differences between large private and large state-owned
enterprises often evaporate on close inspection: ‘Large, successful firms—regardless
of ownership—exhibit substantial similarities in areas commonly thought to
distinguish SOEs from POEs: market dominance, receipt of state subsidies, proximity
to state power, and execution of the state’s policy objectives’ (p.267) (Milhaupt and
Zheng 2014). A good example is Brazil, where ‘thanks to SOEs Brazil developed
large sectors that initially were not funded by the private sector alone, such as steel,
airplane manufacturing, telephony, national oil, gas, petrochemicals, mining, and an
integrated electric grid…’ (Musacchio and Lazzarini 2014b: 16). Moreover, most
applied innovation efforts were also developed either by state agencies like Embrapa
in agriculture or by large state-owned enterprises such as Petrobras and Embraer
(Musacchio and Lazzarini 2014a).

7.4.2 Latitude for performance standards

The prevailing incentive structure for private sector actors, including the persistence
of rent-seeking opportunities and lack of incentive to accelerate learning and
accumulate technological capabilities, creates intense pressure on a society or
government seeking to speed up structural transformation. Hirschman (1967: 87)
notes that latitudes for performance standards are ‘the propensities and pressures to
which the decision makers are subject’, and that some projects or activities are ‘so
structured that latitude is severely restricted or completely absent: in these cases I
shall speak of lack of latitude or positively, of the presence of “discipline” imparted
by the project.’ However, such pressure mechanisms or disciplining opportunities are
not equally realized in all sectors. For instance, developing countries typically have
better advantages in process-centred production (such as refineries or cement
manufacture) than in product-centred production.
Technological and economic characteristics impose different latitudes on
performance standards in the three studied sectors in Ethiopia. Floriculture’s narrow
latitude for failure arises from the perishability of the product, and the care it requires
from farm to points of consumption. This is reinforced by constant pressure from
supply-side technological uncertainties. This is one reason why there is a
concentration of family-owned medium-sized firms in the sector. The exclusive
orientation to export markets imposes the need to cope with international
competition. This has had an important disciplining effect, in particular, on the many
domestic firms that were new entrants into the industry. It also helped to liquidate at
an early stage the exclusively rent-seeking firms that were unwilling to change.
The same was true of the cement industry, whose economies of scale, capital
intensity, high entry and exit barriers, and continuous production (p.268) process
acted as a positive force to improve capacity utilization. The risk of failure was too
great. This was reinforced by the nature of the product, which is perishable and bears
immense risks and legal implications if quality standards are not met. In contrast, in
the leather sector such pressures did not exist, particularly in the tanning and basic
leather sub-sector. The industry was built on the dominance of tanneries and not on
the leather products sub-industry. By contrast, the leather products sub-industry has
much narrower latitude for failure. Producers in this sector also ‘exploited’ the
loophole of being able to supply the domestic market, rather than the much more
competitive export market. This wide latitude interacted with the inadequate and
inconsistent policy approaches in the sector.
According to Lall (1999), the ‘domestic market plays an important role in national
capabilities’, as local demand affects the development of products, quality, etc. This
is crucial in countries like Ethiopia, where the domestic market is, given population
size, potentially large. This requires intensification of competition in domestic
markets, which helps to dissipate rents not based on performance. Porter (1998: 119–
20) argues that
domestic rivalry not only creates advantages but also helps to avoid some
disadvantages…The stock of knowledge and skill in the national industry
accumulates as firms imitate each other and as personnel move among
firms. Domestic rivalry not only creates pressures to innovate, but also to
find ways that upgrade the competitive advantages of firms. Toughened by
domestic rivalry, the domestic firms are equipped to succeed abroad.

Where really open and fair competition is unlikely, other devices need to be instituted
by the state, such as reciprocal control mechanisms or other disciplining measures.
The role of the domestic market in building national capabilities also suggests that
there should be no dichotomy between export-led industrialization and import
substitution: the value of exporting does not preclude nurturing production for the
domestic market. This is opposite to the view associated with the approach that
undermines import substitution, hindering integration between both strategies to
maximize their complementarities. In the three Ethiopian sectors, export discipline
and competitive pressure seem to have played a critical role, especially in flower
production. Lack of intense competition in the domestic market (p.269) for leather
and leather products, and the weakening of export discipline are major factors in the
sector’s disappointing growth. The disparity between the success of state-owned EAL
(unlike many other state-owned airlines), which has had to compete in the
international aviation industry, and the disappointing performance of the state-owned
Ethiopian Shipping Lines, which has enjoyed a monopoly and unjustified rents, is a
typical example of the critical role of competition in industrial performance and
competitiveness. Competition is not a magic ingredient, as it may not breed success
(as witness the failed airlines around the world). Though it is rarely ‘perfect’ and is
neither the only nor always the most effective means of promoting learning,
adaptation, or productivity, competition is, nonetheless, one ‘disciplining’ mechanism
that narrows the latitude for failure.

7.4.3 Stage of industrialization

Variation in outcomes of industrial policy across the three sectors is partly a function
of the challenges and opportunities that come with the level of development of the
industry. It appears that, in some ways and contrary to what one might expect,
initiating a new industry is easier than stimulating an existing industry. For instance,
Hirschman (1968) highlighted how the challenge in the early stages of import
substitution is much easier than in the second stage of industrialization, when the
interlocking interests of different groups and political structures have become
established. There may also be a problem if there is insufficient capacity among
policymakers to follow and understand changes in industrial structure over time and
to adapt to the sector’s dynamic challenges. For example, floriculture, after a decade
of initial growth, appears to have entered a new industrial stage since 2010, but the
state has largely maintained the existing policies. There is evidence that the slower
growth of the flower industry after 2009 reflected policies that have ‘run out of
steam’, requiring new policies suitable for the next growth stage. The same risk
seems to confront the cement industry, which has entered a new phase since 2012.
Each industrial stage brings its own constraints, challenges, and opportunities that
call for appropriate responses. Arguably, initial success in an industry may create the
illusion of wider latitude for failure, and entrench institutional patterns of behaviour
and sets of interests, all of which pose fresh challenges for policymakers. This
highlights the need for policy (p.270) capabilities that include the ability to sustain a
long-term vision and constant monitoring.

7.5 Unevenness of policy instruments and organizations

Policymaking is a dynamic and complex process beset by dilemmas, tensions, and
uncertainties. It involves diagnostic policy design, implementation, and evaluation of
impacts. For instance, if policy outcomes are not monitored, how can policymakers or
others know whether there are blockages or what the concrete policy effects are?
Policies are also the product of negotiations between conflicting interests, rather than
merely technical endeavours or the seamless translation into action of higher-level
visions or strategies. This implies that policies adopted as feasible and relevant may
not necessarily be the most effective ones.
It is often difficult to separate the outcome of a policy from other relevant variables,
and policy outcomes may need a long time horizon for their full effects to become
clear. Successful policies would make themselves redundant over time, due to
emerging constraints, challenges, and opportunities. Although plans are necessary
elements of industrial policy, their effectiveness is undermined by the uncertainty of
data and the reality of constant change. Within this complex context, it is likely that
mistakes will be made. Policy-learning capabilities may help reduce mistakes and
improve the pace and scope of policy learning. Ultimately, what matters is that
benefits outweigh costs, and that lessons are learned to minimize future mistakes and
develop the capacity to adapt to and recover from mistakes.
More importantly, the choice of policy instruments reflects political choices and
political constraints (domestic and international). While the same instrument may be
used for different aims, divergent instruments can be used to achieve a specific goal.
For instance, privatization or FDI policy may serve different purposes, and many
incentives can be used with an inbuilt reciprocity principle. In Ethiopia, the
government’s cohesive strategic direction and political orientation has infused most
specific policy instruments, contributing to a consistency. Federalism, the relative
weakness of the state bureaucracy and stronger institutional strength of the party, the
relatively underdeveloped private sector, and the state’s activist orientation, have all
shaped political choices. For instance, active (p.271) use of state-owned enterprises
where ‘the market’ was reckoned unlikely to lead to the most productive and
developmental outcomes was compatible with the government’s approach to
privatization. This can be contrasted with policies followed in other African countries
such as Mozambique and South Africa (see Cramer 2000). The policy on FDI has
been consistent, given the relative weakness of the domestic private sector and the
government’s priority to strengthen it. The political commitment to strengthen the
DBE and provide subsidized long-term loans to key industries was made possible by
the cohesiveness of the governing coalition and by the political pressures that helped
cement this cohesion. Nevertheless, the disparity in the coherence and concentrated
use of the design and implementation instruments has generated varied results.

7.5.1 Inconsistency in applying the reciprocity principle

Although most of the instruments (for instance, investment promotion, export
promotion, exchange rate policy) were applied across the three sectors, their
outcomes, importance, and applicability varied from sector to sector. Export
promotion required huge administrative capacity, while investment incentives were
easier to administer. The evidence shows that uniform incentives were easier to
implement than sector-specific instruments, and applying a ‘reciprocal control
mechanism’ was difficult: incentives were linked very weakly to performance
standards. The principle of reciprocity depended upon a more sophisticated capacity
to administer the incentives than incentives without such a requirement. Export
promotion policy was less effective in leather and leather products, as it partly failed
to shape the behaviour of actors. The policy to attract FDI into the leather sector had
limited effect until 2006. In the absence of a strong link between incentives and
performance, outcomes have depended partly on the nature of actors, export
discipline, and competitive pressure. The message is that reciprocity and
performance-driven instruments have a central function in industrial development.
In Korea, ‘the reciprocity principle operated in almost every industry’ and ‘in return
for protection in the domestic market, the government required the enterprises to
export part of their production’ (Amsden 2007a: 96, 2001: 151). However, the political
setting in Ethiopia is different from Korea’s. Unlike Korea, where the political
economy allowed for a concentration of intermediate assets among national
champions (p.272) (Amsden 2001), the requirements of federalism and a political
commitment to equitable regional growth make such resource concentration difficult
in Ethiopia. Amsden argues in The Rise of the Rest that where there is broad equality,
it is easier to concentrate intermediate assets, because there is less political risk.
Ironically, where there is high initial inequality (as in Latin America), risk-averse
leaders do not dare to concentrate these assets. The Ethiopian experience suggests
that, despite low initial inequality, the federal system constrains the scope for
concentration of intermediate assets and necessitates a wider dispersal than may be
optimal for industrial policy. The endowment funds may contribute to narrowing
regional disparity, but also arouse envy in regions where no strong endowment funds
exist.
An important condition for the reciprocal mechanism to work is the introduction of
export discipline, as it constantly forces firms to improve their productivity. Ocampo
(2007: 2), for instance, emphatically argues that ‘…a successful export strategy is
highly dependent on how the export sectors are integrated with other domestic
economic activities, not least in terms of employment generation’. Export discipline
must be supplemented by increased competition and rivalry in the domestic market.
According to Porter (1998: 119–20), toughened by domestic rivalry, it is more likely
that ‘domestic firms are equipped to succeed abroad’. When imposing reciprocal
control mechanisms or equivalent forms of discipline becomes too difficult,
governments are left with the ruthless discipline of export competition as the chief
mechanism to promote adaptation (or to destroy its prospects). That is why it is
suggested that competition policy be an integral part of any industrial policy (see also
Roberts 2004). In contrast to the mainstream approach, these authors argue that
competition policy ought to be seen as the set of measures providing
incentives as well as the “sticks” fostering innovative behaviours.…In
short, competitive pressures on individual firms must be strong enough not
only to dissipate monopolistic rent but, more importantly, to induce firms to
adopt active competitive strategies instead of just profiting from the
incentives provided by industrial and technology policies’ (Possas and
Borges 2009: 450)

Domestic rivalry within the leather and leather products industry continues to be low,
allowing firms to operate in a comfort zone while undermining the incentive to
improve competitiveness and export orientation. In contrast, competition in the
cement industry has latterly (p.273) intensified, while competitive pressure has been
high throughout in floriculture. This does not mean that designing and implementing
competition policy is easy, as it is usually captured by specific interests.
Export target-setting and monitoring were put in place after 2006, but the evidence
shows this has not been highly successful. The most significant reasons for this
failing have been lack of reliable data and of effective participation by relevant actors
in the planning process. Nonetheless, it would be a mistake to dismiss target-setting
as a complete failure, for the research also shows that in some ways there have been
‘policy linkage’ and ‘institutional linkage’ effects. For instance, target-setting has
helped reveal bottlenecks in capability, calling forth responses, though unevenly. As
an example, the export earnings target for leather and leather goods, which is set each
year, was $0.5 billion in 2014–15. Although actual implementation lagged badly,
investigation of causes and bottlenecks resulted.

7.5.2 Compatibility and cohesiveness of policy instruments

There was variation in the use of generic and sector-specific instruments. Other than
the generic policy instruments used in multiple sectors (for instance, exchange rate
policy and investment incentives), there are instruments that have been used to
address sector-specific constraints. The varied approaches pursued to solve cold
storage constraints and packaging inputs in the flower sector, and the approach to
solve the raw materials constraints in the leather sector (supply, price, and quality of
raw hides and skins) are classic examples not only of specific needs, but also of the
variable success in addressing them. Capacity to address such constraints is
influenced by the priority given by the top political leadership, depth of analysis, and
institutional strength. In general, sector-specific instruments were more effective than
the generic incentives in addressing the constraints in each sector, and their
effectiveness is easier to assess. Integrated and coordinated use of instruments also
enhanced policy effectiveness. In the flower and cement industries, there was better
governance of policies than in leather. The heterogeneity and scattered nature of the
actors and the diverse sub-industries in leather and leather goods have thus far over-
taxed existing policy capabilities.
There were also disparities in the speed of application of instruments, and in how
much they were concerted. Instruments that were (p.274) implemented swiftly and in
concert had better effects than those implemented piecemeal. The support provided to
the flower sector was swift, coordinated, and concentrated, and had a clear impact.
Loans were provided in high density without fundamental mistakes being made.
Mistakes were indeed made, but the cost was much lower than the benefits. In
cement, support in accessing mineral deposits and energy supplies (electricity, heavy
fuel oil, coal), and the design of regulations were directly related to the requirements
of the industry. It became easier to identify the progress and drawbacks of sector-
specific instruments. In floriculture, the critical requirements were serviced land
(with transportation and energy access and proximity to the airport), availability of air
transport, cold storage facilities, and packaging options. Support has been focused on
these issues, enabling the sector to grow. In the leather and leather products industry,
sector-specific interventions were not evident until 2010, when benchmarking and
upgrading value addition were introduced. It is clear that sector-specific instruments
deserve greater focus and are more challenging. Policy decisions related to leather
were sometimes inconsistent (for instance, the export of live animals and import of
inputs), which further undermined policy effectiveness. Some of the decisions were
not taken at an early stage, as was evident in floriculture.
High-level political commitment was crucial in developing the floriculture sector.
The same was true of the cement industry, as there was strong political pressure on
government to support the construction sector. The state was able to exert strong
pressure and make bold decisions to meet requirements. In the leather sector, despite
its priority in government policy, this was not the case. The measures were too
fragmented and incoherent, producing minimal synergy. Industrial loans (investment
and working capital) were not easily available in sufficient amounts.

7.5.3 Quality and appropriateness of policy responses

The quality of industrial policy depends on the ability to make adequate analysis of
the situation, participation of concerned actors in the policymaking process, and
transparency and accountability in decision-making. This book shows that the quality
of policies has been mixed. The multimodal transport intervention is a typical
example of insufficient study or preparation and minimal involvement by
industrialists and (p.275) other stakeholders. This failed initiative created additional
disruption in the trade logistics of the leather industry. Insufficient analytical work
was observed in many decisions across all sectors, although the effects were felt less
in industries with active enterprises and intermediary institutions (mainly in
floriculture) and larger firms that leveraged the advantage of their size (mainly in
cement). The major victim was the leather sector, as key stakeholders were passive
and the intermediary institution was weak.

7.5.4 Coordination and insufficient organizational competence

1. a) Insufficient institutional capacity: Implementation failure has mirrored
implementation capacity (in terms of attitude, skills, and structure) in various
government agencies. For instance, LIDI (in the leather sector) had better
organizational capacity than EHDA (in the flower sector) in terms of facilities
and staffing. Nevertheless, the challenges and constraints in both sectors were
different, and the capacity of LIDI was inadequate and focused on
administrative tasks. This was further compounded by the relative weakness of
the fragmented actors in the value chain. There was no strong intermediary
institution of input suppliers, and the agency working on livestock
development was weak.
2. b) Institutional tensions and poor coordination: The coordination requirement
differed among instruments and sectors. Coordination among federal
government agencies (horizontal) and among governments (vertical: federal,
regional, local) has in most cases been deficient. This has been especially
important for the leather and leather products value chain. Federal and regional
priorities are not necessarily the same, a factor compounded by capacity
constraints, which are even more visible at lower levels. Further, rigid
bureaucratic practice and hierarchies in government agencies (another example
of path dependency but also the product of particular political history, see
Lefort 2007, 2013) acted as constraints on learning and adaptation.
3. c) Intermediary institutions and state–business relations: Intermediary
institutions are important in enhancing an industry’s capacity in terms of
market development, training, and technological advancement, or representing
the interests of economic actors. They (p.276) facilitate policymaking by
improving the flow and quality of information, communication of intentions,
and joint actions. There has been major variation in the roles played by
intermediary institutions among the three sectors.
EHPEA has played an active role throughout the history of the floriculture industry.
Communication with the highest level of policymakers and government agencies was
open, facilitating the timely resolution of problems and improving coordination. In
the cement industry, the absence of intermediary institutions did not hamper
communication, due to the large size of the firms. This does not necessarily imply
that this will be the case in the future though. In the leather and leather products
industry, ELIA, which mainly represented tannery interests, was weak and passive
with respect to influencing policymaking, and members lacked a common vision.
Moreover, economic actors in the input supply chain were not organized into any
association.

7.6 Conclusions
The country’s political landscape, rapid economic growth, and development strategy
have been conducive to the growth of all three of the sectors studied. What has not
thus far been effectively explained is why and how, within this broader context, there
has been such marked variation in performance. The dilemma here is why a common
industrial policy foundation produced different outcomes in different sectors. Existing
literature provides limited empirical evidence and research, as the more typical
comparisons in industrial policy and developmental state literature are between
countries rather than between sectors within individual countries.
This chapter has synthesized the conclusions and analyses in the various case studies
and sought to identify comparative insights, thereby generating a unique picture of
the patterns of policymaking and performance. The discussions have also shown that
industrial policy outcomes in Ethiopia have been distinctly uneven in recent years.
The research suggests there are three main factors shaping this variation, with
significant policy implications and a wider relevance beyond Ethiopia. First,
industry-specific factors, in particular the nature of economic actors, the industry’s
latitude for performance, and the stage of development of the (p.277) industry were
the key determinants of the pace and scope of policy outcomes. Second, the varying
scope for linkage effects in different sectors, and the appropriateness of policies in
promoting creative linkage effects, were crucial factors, with path dependence and
low-value traps acting as impediments in one sector. An additional point is that past
sectoral experience proved to be more of a constraint than a boost to industrial
prospects: the main example here is the largely stagnant and mainly export-oriented
leather and leather products sector, where a form of path dependence thwarted policy
objectives and interventions.
Third, the disparity in political commitment to, compatibility between, and
consistency of policy instruments and institutions was significant. Political factors
(for instance, the post-2005 election period and urban development programmes,
such as housing and infrastructure) played an important role in promoting the cement
industry. Political pressure from scattered producers in leather inputs and processing
played an equally important, but opposite role. The quality of policy decisions and
institutional capacity varied among the sectors. The particular tensions and trade-offs
that arise in policymaking across individual sectors, ministries, and organizations are
also very relevant. Overall, what matters for the evolution and effectiveness of
industrial policy is the way these three factors—industrial structure, linkage
dynamics, and (broadly) politics—interact.
Finally, the contribution of this research, and perhaps its most original policy
implication, is that it has unequivocally shown the importance for policymakers of
understanding and engaging with the interaction of industrial structure, linkage
potential, and politics/political economy. The interrelational dynamics among these
variables have significant implications for policy design and implementation; and for
the type of selective intervention chosen to promote industrialization, as well as for
how to guide the design of reciprocal control mechanisms and assess their viability. It
is often pointed out that selectivity is critical to industrial policy: this research
provides a way of guiding thinking regarding that selectivity.

Notes:
(1) For a discussion of ‘adaptive capacity’ in decision-making in ‘mega-projects’, see Giezen et al.
(2014).

(2) And indeed as the 2013 Australian Open tennis champion Stanislaw Wawrinka
had tattooed on his forearm.
(3) Rodriguez and Rodrik (2001:338).
(4) On Ricardian and Kaldorian strategies, see Chapter 2, Section 2.3.2.
(5) Musacchio and Lazzarini (2014b: 16) highlight ‘…thanks to SOEs Brazil
developed large sectors that initially were not funded by the private sector alone, such
as steel, airplane manufacturing, telephony, national oil, gas, petrochemicals, mining,
and an integrated electric grid…Most applied innovation efforts were also essentially
executed by state agencies (such as Embrapa in agriculture) as well as large SOEs
such as Petrobras and Embraer.’
(6) <https://www.sss.ias.edu/files/pdfs/Rodrik/Research/An_African_growth_miracle.pdf> ‘An
African Growth Miracle?’, Institute for Advanced Study, Princeton, April 2014 (p. 2).
(7) The demonstration effect of this commitment should not be underestimated.
Arguably, it helped reveal fresh opportunities in other sectors and hence to build
momentum for FDI flows into Ethiopian manufacturing. In this sense, there may be a
form of global intersectoral linkage, where the obstacles and pressure points include
multinational firms’ challenges in finding reliable manufacturing and assembly
suppliers.
(8) Here there is a direct parallel between Hirschman’s idea of linkages and his (1967)
idea, in Development Projects Observed, of the principle of the ‘hiding hand’,
whereby underestimation of the costs and difficulties of a large project are necessary
for the project to be begun at all, but then, once these problems arise, they often
provoke creative problem-solving adaptations.
Lessons from Industrial Policy in Twenty-
First-Century Africa
Chapter:
(p.278) 8 Lessons from Industrial Policy in Twenty-First-Century Africa
Source:
Made in Africa
Author(s):

Arkebe Oqubay
Publisher:
Oxford University Press

DOI:10.1093/acprof:oso/9780198739890.003.0008

Abstract and Keywords

This chapter increases the depth of the analysis by assessing the implications of the
book’s evidence and argument for Ethiopia and for Africa more widely. Industrial
policymaking in Ethiopia is a work in progress, but the evidence shows that industrial
policies can work and thrive in a low-income African country, and that the state can
and should play an activist developmental role, with policy independence an
important ingredient. However, the book also highlights how great the challenge of
catching up and industrialization is for twenty-first-century Africa—while emphasizing
that it is far from impossible. It is also necessary, if the narrative of ‘Africa Rising’ is
to translate into more than a temporary and narrowly enjoyed boom.
Keywords: Africa, Ethiopia, industrial policy, state’s developmental role, policy
independence, industrialization, catch up, industrialization, policy learning
This chapter ties together the book’s findings and arguments on industrial policy and
performance across three sectors in Ethiopia and highlights future research areas.
After summarizing the previous chapters and focusing on policy learning and lessons
for Africa and other developing economies, the chapter concludes with a discussion
of Africa’s catch up and activist industrial policy in light of the theoretical
perspectives and empirical findings presented in the book. This final section
reinforces the link to the basic premise of the book—that industrial policy should be
the vehicle for catching up and structural transformation. As has often been noted in
earlier chapters, this is a departure from the more ‘market-friendly’ interventions
favoured by conventional economists, including those who have rediscovered
industrial policy in recent years.
8.1 Summary

8.1.1 Main findings and theoretical implications

This book has examined industrial policymaking in Ethiopia between 1991 and 2013,
and has sought to shed light on why outcomes have been uneven across industries,
despite a common foundation in industrial development strategy and industrial policy.
This puzzle has received little attention in existing literature, since most studies
generalize at the national and international levels, and not at the level of cross-
sectoral comparison. Above all, the research for this book supports the
argument (p.279) that industrial policies can work and indeed thrive in a low-income
African country such as Ethiopia, and that the state can play an activist
developmental role, with policy independence an important factor. The pretentions,
lapses, setbacks, and failures of the ‘developmental state’, far from indicating its
uselessness, should be viewed, the book argues, as part and parcel of the real-world
process of accelerating structural change and development. Industrial policymaking
in Ethiopia is a work in progress, and the examples discussed in the book illustrate
the colossal challenge of catching up and industrializing in twenty-first-century Africa.
In Chapter 2, the cardinal concepts of industrial policy, structural transformation,
catch up, the developmental state, and linkage effects are examined, and the African
context is briefly laid out. In addition, debates on industrial policy and development
are presented, and the economic structure in developing countries (especially SSA
countries) and historical growth trends are outlined. Industrial policy has always been
the subject of ideological contention, and the dominant perspectives are strongly
averse to such policy in developing countries and have been better at accounting for
its failures than its evident successes. This book has deployed structuralist and
political economy perspectives (more pluralist than the ‘mono-economics’ prevailing
in the discipline since the late 1970s) to investigate the rationale for and recent
experiences with industrial policy in Ethiopia.
Ethiopia is a low-income, landlocked economy in a ‘bad neighbourhood’, and has the
second largest population in Africa, and yet it is also one of the few developing
countries to record rapid economic growth in the early twenty-first century.
Chapter 3 maps out the history and foundations of policymaking in Ethiopia,
including the industrial development strategy of the current government and the
associated institutional framework, information that is not available elsewhere in the
literature. The policy instruments include industrial financing, investment and export
promotion, trade protection, the state as direct economic actor/industrial player, and
privatization. The chapter sets out the roles of the overall export coordinating
institution and specific institutions designed to develop priority sectors, intermediary
institutions (such as industrial associations), and investment administration
institutions. This descriptive analysis highlights government adaptations of
international experience and practice, particularly in East Asian countries. Industrial
policy instruments relating to specific sectors are further discussed in subsequent
chapters.
(p.280) Chapter 4 examines growth patterns, linkage effects, and industrial policy and
policymaking in the cement industry. Cement production is a strategic industry in
many countries. The Ethiopian cement industry has undergone major changes
throughout the period under consideration, growing faster than in most developing
countries. This growth, although characterized by swings, has been dramatic and has
been driven by the expansion of the domestic market and the construction boom. The
book demonstrates how the government’s industrial policies shaped the development
of the cement industry through direct and indirect interventions. Some policies were
more effective than others. Similarly, the incentives yielded different outcomes, some
more desirable than others. The cement sector has absorbed a large share of scarce
resources, involving a number of trade-offs, tensions, and learning experiences. Also
demonstrated in the chapter is the synergy between industrial and other public
policies, the interplay between economic and political factors, as well as the
dynamics of policymaking and the significance of the narrow latitude for poor
performance. In sharp contrast to the overall African cement industry, where
multinationals preponderate, domestically owned firms continue to dominate the
industry in Ethiopia. SOEs have played a pioneering role through spill-over effects,
and continue to make an important contribution. Overall, the cement industry in
Ethiopia has been a binding agent of economic development and transformation in
multiple ways.
Chapter 5 explores floriculture, whose economic success has attracted international
interest and policy debate. This sector shares many characteristics with
manufacturing. Emerging in Ethiopia in 2004, it has since experienced sustained
growth, making Ethiopia one of the leading producers and exporters of cut flower in
the world. The standard explanations for this success are comparative advantage,
factor endowments, a discovery process, and developmental patrimonialism, rather
than the role of the state. A detailed analysis of the drivers of the growth in the
industry is undertaken in the chapter, yielding a more comprehensive explanation
than has been available before. Government policy has been critical in nurturing and
expanding the sector, as have the interplay between policymaking and institutions,
the dynamics of industrial structure, and interest groups. Ethiopian floriculture
classically demonstrates how unemployed labour and underutilized local
entrepreneurial potential, as well as natural endowments, can be mobilized for
economic (p.281) development, but do not just bloom automatically under the warm
sun of comparative advantage. State commitment to develop the sector and its use of
policy instruments was exceptionally clear and coherent. Collective learning was also
impressive, with the industry ‘picking the state’ and the state ‘picking the firms’.
Floriculture also benefited from the narrow latitude for poor performance and export
discipline. The chapter also shows how policies that helped bring about successful
take-off became insufficient as the industry matured and confronted new challenges,
which call for new policy responses. The chapter demonstrates an ongoing ‘learning
by making policy’, in spite of these new challenges.
Chapter 6 focused on the underperformance of the leather and leather goods industry
and the reasons for the disappointing outcomes. The ‘joint production’ of outcomes in
this case involved policy design, the structure of the industry, and, very much, the
sector’s political economy. What the chapter clearly shows is that factor endowment
arguments are grossly inadequate explanations for the failings in this sector. In
contrast to cement and floriculture, the performance of the sector has been
disappointing and characterized by erratic and sluggish growth. Industrial policy has
been unable to reverse this poor performance, or to fully exploit potential linkage
effects and insertion into the GVC. The main puzzle is that there has been prolonged
manufacturing experience in the sector and there is a plentiful endowment of
livestock. Yet the policies failed to generate the required behaviour and export
discipline among firms. The main economic interests in the sector have been
fragmented and often conflicting. Tanneries have dominated, while the leather
products industry (which is integrated into GVCs) has failed to exert much influence.
This demonstrates how linkage effects are not necessarily automatic, and how the
policy instruments and institutions also lacked the dynamism to promote them. Policy
also failed to address strategic input problems, while trade and logistical constraints
further weakened the sector’s competitiveness. Dependence on small-scale livestock
husbandry is also part of the explanation, as is the wide latitude for poor performance
in the sector. The sector has experienced path dependence that has perpetuated a low-
value trap. Despite these failings, recent policy on value addition and new entrants is
helping to break the logjam in the sector. Moreover, there is recent evidence of more
investment, better quality, and increased exports of higher-end products.

(p.282) 8.1.2 Implications for policymaking in Ethiopia

It is in Chapter 7 that the insights from previous chapters are synthesized and
developed. While sectoral performance has been uneven, the development-oriented
state performed better over time (especially after 2002) and its industrial policy has
contributed to higher growth. That said, manufacturing output remains extremely low.
One thing the case studies reveal is that learning by doing is as much a feature of
government policymaking as of firms’ or organizations’ performance. Comparative
advantage/factor endowments and neopatrimonial arguments do not provide
sufficient explanation for the unevenness of policy outcomes, and cannot account for
such dynamics as learning by doing, learning by failing, or ‘failing better’. The book
demonstrates that there are three main factors shaping the variation in sectoral
performance, and suggests that these may have wider relevance for other developing
countries. First, industry-specific factors, in particular the nature of the economic
actors, the industry’s latitude for performance, and its stage of development, matter
greatly. Second, success in promoting creative linkage effects was crucial, while path
dependence and low-value traps act as impediments. Third, the depth of political
commitment to, consistency of policy instruments, and compatibility of policy
institutions was significant to the success of a sector. Overall, what matters for the
evolution and effectiveness of industrial policy is the way these three factors interact.
The book also shows that even in successful economic sectors, the seeds of failure
were and are present, while there are signs of productive policymaking evident even
in the disappointing leather and leather goods sector. This has significant
policymaking implications for these three sectors, the wider manufacturing sector,
and for overall economic development in Ethiopia and possibly beyond. As the book
repeatedly shows, there are many important issues requiring policy responses. In the
cement sector, the key policymaking challenges are sustaining the sector’s dynamism
through linkages to construction and infrastructure development, and the
development of coal mining. There is also the promotion of the manufacturing of
other building materials. Although domestic market growth is the primary driver of
this sector’s performance, different instruments are essential to expanding the
industry’s market horizons and to improving its productivity and competitiveness.
Furthermore, (p.283) technological development, a neglected area, needs policy
focus, as it could serve as the foundation for equipment manufacturing.
The leather and leather goods sector deserves special mention, because of its
apparently wider scope for linkages, its contribution to export earnings and
employment generation, and indeed its potential for poverty reduction in rural areas.
Promotion of multiple linkage effects, especially backward linkages, is vital. A
comprehensive review of policies is required, as is devising policies that break the
sector’s path dependence and a constant focus on upgrading the industry. These
efforts need to be augmented by the development of market and technological
capability by attracting foreign firms that are key players in the GVC. While
inadequacies and limitations in research and knowledge are often obstacles to
effective policy design and prioritization in Ethiopia, this is less the case in this
sector. However, what became clear is that the policy challenge in this sector is at
least as much political as technical.
Meanwhile, floriculture is in a critical transition phase to a new stage of development.
Specifically, the development of domestically owned firms, technological deepening
of the industry, and the development of linkages to the wider horticulture sector in
various growth corridors have emerged as new challenges that the government needs
urgently to address.
Although this book focused on three specific sectors, it has wider implications for
manufacturing. First, a detailed study of the remaining sectors is necessary, although
they may generally resemble these three sectors. The textile and garment sector, for
instance, has been long established in Ethiopia, and is an export priority. It faces
major constraints, and its growth has not been impressive until very recently. Linkage
effects with agriculture are significant, and the domestic market is growing. Both
domestic and export markets are key drivers of its growth. Moreover, agro-industries
are playing an insignificant role in export earnings, despite their large potential.
Among import-substitution industries, pharmaceuticals, steel, and beverages have
divergent characteristics. More broadly, the point is that policymakers cannot design
a cookie-cutter policy in the expectation that it will work uniformly across sectors.
Instead, they need to design policy and to assess performance on the basis of detailed
knowledge of the sector and its political economy. This stricture applies both to state
departments and development banks.
Second, there are also cross-cutting issues affecting all manufacturing industries that
require a policy response. For instance, in view of global (p.284) competition and the
global business revolution (Nolan 2003), redefining the role of national champions is
required. Present policy has been insufficient to maximize the synergies between the
export and domestic markets. Furthermore, industrial clustering and industrial parks
have played an insignificant role till now, but could play a much bigger future role in
overall industrial development strategy. Again, some of these issues point to
dilemmas that the government will need to address. For instance, there is the tension
between industrial clustering and agglomeration and the political commitment to
spreading resources and opportunities across federal regions. Meanwhile, the effect of
joining WTO would need to be considered within this broader strategic and policy
context, and with an understanding of its implications for the industrialization process
and for preserving ‘policy space’ (UNCTAD 2014).1
Third, there is a need to promote policy capabilities and adaptive capacity, including
developing multiple policy institutions. The availability of reliable data is a major
problem in Ethiopia, as research institutions are too few and too thin. Policy learning
has to be supported by analysis based on reliable data. Many government agencies
lack full knowledge of the industry they are responsible for and their plans rely on
insufficient research. Furthermore, industrial policymaking is intertwined with the
federal political framework, and each needs to reinforce the other. Considering the
limited and unique experience of federalism in Ethiopia, it is important that the
interplay of policymaking, institutions, and federalism be properly researched. Other
countries such as Korea, Taiwan, or China, having unitary government systems, yield
little information of direct relevance, although clearly, given China’s massive
regional differences, there may be something to learn from Chinese policy
experience. Such research in Ethiopia might yield unique insights of value to Ethiopia
and other countries.
Finally, perhaps the most original policy implication of this book is its unequivocal
demonstration of the importance for policymakers of understanding and engaging
with the interaction among industrial structure, linkage
potential, and politics/political economy. The dynamics among these variables have
significant implications for policy design and (p.285) implementation, for the type of
interventions to be instituted to promote industrialization, and for the design of
reciprocal control mechanisms and assessment of their viability.2

8.2 Lessons for policy learning in twenty- first-century Africa

As argued above, industrial policy can be effective in Africa, notwithstanding the
conventional wisdom among development agencies, economists, and IFIs. One
consequence of the weight of this conventional wisdom has been the diversion of
research attention elsewhere: it is important that this be reversed. There is a need for
more research into African industrial policymaking, particularly its intersection with
structural change and its relationship with politics and political settlements in
individual countries. This would improve understanding of how industrial policies
can promote structural transformation and foster economic catch up. The author
hopes to have made the case throughout for engaging in further comparative research
within specific countries. Such research should focus on country-specific
understanding of, and practices in, policy development and implementation, as
theoretical concepts and constructs have ultimately to be based on empirical evidence
and experience. Much work has been done on African industrial sectors, but it is
often either narrowly technical or too shallow, the product of ‘quick and dirty’
studies. And it is important to encourage pluralism in research and evidence
collection. The lesson after decades of enormous amounts of research—especially
into rural economies—to support the global poverty reduction campaign is that
significant gaps in perspective and evidence are still all too common. These are
arguably the result of ideological blind spots and of the influence of entrenched, but
insufficiently questioned, research assumptions.
Three key issues of policy learning in Africa emerge from this book: the critical role
of policy independence in policy learning; the role of learning-by-doing as the main
thrust of capacity-building in policymaking; and the intelligent use of signals and
mechanisms (such as (p.286) latitude for performance, hiding hand, and linkage
effects) to steer and enhance policy learning.

8.2.1 Policy independence

Policymaking is a pattern and mode of action geared, in this case, to economic
development. Policy choice influences shifts in political economy, and is constrained
by the existing political economy. In other words, no state is wholly autonomous
from the wider political economy of a society. Policy independence above all means
the right, and political space, to make policy choices free of political pressure or, at
any rate, without succumbing to particular interests. From a slightly more unusual
perspective, it means reserving the right to make mistakes and, in the process, to learn
from them. Policy independence also means the freedom to make major policy
decisions that entail risks and bold experiments. Without this dimension, policy
decisions will only sustain the status quo. In terms of industrial policy, this may
effectively mean abandoning hopes of anything other than the very slowest and most
modest structural change. Major policy decisions will only achieve more ambitious
results if they are made within a long-term perspective. So, policy independence also
entails the right to chart long-term perspectives for development. This independence,
however, has its limits. Unlimited protection against the consequences of ‘mistakes’
may reinforce a pattern of failure, with massive social and political consequences.
Policy independence is unlikely to generate constructive experimentation and
learning from errors if applied without discipline. This discipline comes in various
highly context-contingent forms, but is fundamentally a matter of political viability.
Policymaking in Ethiopia has reflected policy independence, including occasional
pressure from IFI and some donors, usually in the form of economic threats. For
instance, the government has refused to yield to pressure to open up the finance
sector to foreign banks, to privatize utilities and telecom, to reform public land
ownership, to freeze public investment, and not to expand universities. The events in
the late prime minister’s office described in Chapter 2 show the lengths to which the
Ethiopian government is willing to go to maintain policy independence. Ethiopia has
also embarked on building the Grand Renaissance Dam on the Blue Nile, depending
entirely on domestically mobilized resources, and despite the threat and withdrawal
of support by external forces. In (p.287) areas of common interest, it has collaborated
with external forces and accepted assistance. The government has been able to
develop its own policies because of domestic political support, and by capitalizing on
the country’s important role in the geopolitics of the region.
Policy independence is a major concern in many African countries. Some countries,
despite independence from colonial rule, have little freedom to make their own policy
choices or, at any rate, have not been highly effective in using what freedom they
have. They are obliged to ‘consult’ on their policy proposals to get a green light, and
receive backing for being ‘good students’. Economic embargoes are enforced against
those who would set a bad example. As one diplomat said to a friend: ‘One of the
unique things about working as a diplomat in Ethiopia is that all feel equal, as
Ethiopia has never been ruled by colonial masters, as was the case in many other
countries.’ Another European diplomat remarked that Ethiopia is fortunate in not
having to get the green light for major policy decisions, and that it can make them
without having to consider the reaction of former colonial masters. It is difficult to
believe that such practices continue in some African countries into the twenty-
first century, after half a century of independence. It is important that African
countries be able to make their own policy choices and bold experiments. This can
only be achieved if African countries individually and collectively resist such
intrusions, and if the international community condemns them. But it is also
incumbent on those, like Ethiopia, with greater room for manoeuvre to use this
effectively, thus to signal to others what is possible and sensible.

8.2.2 Learning by doing, emulation, and role models

Learning by doing is the prime means of mastering production among late
developers. This concept is equally valid for policymaking. We also know that almost
all late developers catch up by emulation, learning by copying. According to Reinert
(2010: xxiii), emulation means ‘imitating in order to equal or excel…[this], rather
than “comparative advantage” and “free trade”, lies at the heart of successful
development…’ As the history of economic development shows, development almost
by nature is about copying and emulation.3 However, there is a balance
between (p.288)copying and learning by doing that is difficult to define. On one
hand, early catching up involved trying to repeat things that happened in the UK in
the ways they happened, although as Gerschenkron (1962) pointed out, the solutions
became different because the game had changed. This is what China and East Asian
Tigers have pragmatically done: copy technology, copy policies, etc. But, on the other
hand, copying and learning are not straightforward, and it is not always obvious
which lessons are more important or applicable.
Some lessons may be quite straightforward, but most are more complex and less easy
to ‘carry across’ from one context to another. The ‘best practices approach’ that
focuses on reviewing and applying detailed lessons to other contexts may not be
helpful, as the conditions in different countries vary in terms of history, culture,
political economy, the role of the domestic market, and exposure to international
trends. For instance, Ethiopia has always been an independent country and was never
colonized. And its history shows that fundamental political and economic changes
have occurred during regime changes, sometimes by violent means. At the moment,
the country’s ethnic federalism is the only such experiment on the continent, and the
country’s ruling party has a history of focusing on its rural political base. So, there is
much that works (or doesn’t) because of political factors that are not the same across
African countries. To adopt the currently fashionable randomized controlled trials
approach to development, or rather the critique of this approach, we cannot be at all
sure that ‘what works here’ will also and equally ‘work there’ (Cartwright and
Hardie 2012).
However, this does not mean there are no lessons applicable to certain other contexts.
Without dwelling on detailed practices, some broad lessons may be drawn. For
instance, understanding the industrial structure of different sectors, and
leveraging latitudes for performance are valuable in industrial policymaking.
Hirschman’s linkage concept (the favouring of industries with strong backward and
forward linkages as well as supporting activities with greater employment linkages
through indirect and induced mechanisms) is essential in bringing dynamics and
impetus to new activities and increasing returns and in creating economic space in
developing economies. The principle of reciprocity is important in almost all
conditions, despite the challenges of implementation and its dependence on the state’s
political clout. As East Asia shows, reciprocity has been (p.289) indispensable in
promoting exports and fighting rent seeking. According to Amsden (2007a: 94, 96):
The guiding principle of the best bureaucracies—politics permitting—was
to give nothing away for free. Reciprocity was ideal.…The reciprocity
principle in Korea operated in almost every industry.…Reciprocity helped
governments. If projects succeeded, they got more power. The elitist
development banks, flagship of the ‘developmental state’, subjected their
clients to monitorable conditionalities. (my emphasis)

Despite the efforts made, the rudimentary application, and the mixed results of
reciprocity in Ethiopia, the government needs to continue experimenting with and
mastering reciprocity.
In addition, many scholars highlight the importance of role models for late
developers. Amsden particularly showed their relevance for learning, based on the
economic histories of Argentina, Chile, and the Philippines in contrast with the
experiences of Korea and Indonesia. Amsden (n.d.) argues that:
A problem in Argentina and Chile was the absence of any hands-on role
model to guide them.…[T]hey could emulate nothing regional. Nor did…
 multinational investment from Europe or the US, offer much guidance.…
Overall, the Philippines…had good students but no teacher to guide it when
an opportunity presented itself…Off-setting the costs of emulation, role
models provide hands-on knowledge to emulators.…Korea and Taiwan…
used many of the same agents, policies, institutions and tactics [as Japan] to
build their own heavy industries.…Africa’s disappointing economic
performance in the presence of rich raw materials may be attributed to…a
continuity in foreign ownership of mines and plantations, and the absence
of a concrete role model to emulate other than abstract theory.

Applying these approaches requires trial and error (rather than the quasi-scientific
notion of experiment in the randomized controlled trial literature) and learning by
doing. It also requires adaptive capacity in decision-making (Giezen et al. 2014). If
learning by doing is important in manufacturing, it is even more so in policymaking,
where it depends on a pragmatic approach, curiosity in learning, and the boldness
(including political commitment) to experiment and make mistakes. There will be no
learning without mistakes, and the aim should be to improve the pace of learning and
to narrow the scope for and of mistakes. Ethiopian patterns of policymaking show
attempts at experimenting on mini and grand scales, often with mixed outcomes.
Learning by doing is probably a (p.290) good lesson for other African countries to
learn in making industrial policy work.
In sum, what is important is not simply the fact of having a role model, but the matter
of which role model presents itself or is chosen. There are some indications in the
basic policy documents of the ruling party and the government that the latter has been
looking to East Asia (in particular, newly industrializing Korea, China, and Taiwan)
for role models. There are also close links with Japanese and Korean scholars on
industrial policy and frequent visits to China by policymakers. The
Japanese kaizen concept has been chosen to lead the productivity and quality
improvement movement in the manufacturing sector; while vocational education and
training and higher education have been organized on the German model.

8.2.3 Applying Hirschman’s supporting devices and mechanisms

Although learning by doing, emulation, and having the right role models are central
to policy learning, they may not be sufficient. As Hirschman highlights, the biggest
problem in underdeveloped economies is the inability to make decisions, and he
underlines that devices and mechanisms which induce investment and policy
decisions are essential. One such device might be institutionalizing a sensibility
among policymakers to linkage effects—as important to policymaking as to
production or industrialization. Some industrial policy decisions and instruments will
force government to take steps to mitigate negative outcomes or exploit the
opportunities that arise. In Ethiopia, DBE took important policy decisions to offset
the failure of loans to the floriculture sector, and this experience was applied to loans
to other sectors.
Another factor that may enhance policy learning is the latitude for performance and
Hirschman’s ‘law of the hiding hand’. According to Hirschman (1967: 28), the hiding
hand is ‘essentially a transition mechanism through which decision makers learn to
take risks; and the shorter the transition and the faster the learning, the better’.
According to this concept, the scope for creativity and energy is usually
underestimated, just as much as project risks and difficulties are systematically
underestimated. This law gives us important insights into the psyche of many
policymakers, who will not take bold decisions on large projects that may involve big
risks. The mechanism of the hiding hand—where it operates, and perhaps too little is
known about the conditions under (p.291)which it is more (or less) effective—can
help developing countries in making such decisions. Hirschman (1968: 129)
highlights this unorthodox approach as follows:
It could be argued that a country without much experience in solving
technological problems should stay away from [such] projects…But the
opposite course can also be defended: how will the country ever learn
about technology if it does not tackle technologically complex and
problem-rich tasks?…[A] certain ‘unfitness’ of the project for a country
becomes an additional and strong argument for undertaking it;…if it is
successful, [the project] will be valuable not only because of its physical
output but even more so because of the social and human changes it will
have wrought. [Emphasis added]

This approach stands in dramatic opposition to the much more common advice from
development economists and aid organizations that interventions should be tailored to
current ‘capacity’; that while South Korea or China might effectively have made bold
interventions, low-income countries should not ‘try this at home’. Again, for
Hirschman, the benefits go beyond the physical outcome in a fundamental way, in the
shifts that new projects may bring about in attitudes to development and readiness to
take risks. The Ethiopian narrative shows that government has been undertaking
extraordinarily challenging projects. The Grand Renaissance Dam project and the
integrated housing programme are typical examples. Such big projects would have
detrimental political costs were they to fail, pushing government to make
extraordinary efforts to ensure success. The hiding hand concept is also associated
with the concept of latitude for performance, as many economic activities have
narrow or wide latitude, depending on the technological and structural nature of the
industries. This concept of latitude for performance forces us to accept that
underdeveloped countries can succeed in capital intensive industries, the airline
industry being a typical example. Industries such as airlines have narrow latitude for
failure, as they need to be run right or not at all. Latitude for performance also applies
to policymaking and political latitude. According to Hirschman (1968: 139): ‘…
latitude and lack of latitude can both be valuable in facilitating learning or acquisition
of needed skills and traits…latitude is attuned to gradual learning, whereas lack of
latitude has a special affinity for the changes that take place through discontinuous
commitments to new values and types of behaviour.’ Understanding the latitude for
performance can help shape and accelerate learning and skills development.

(p.292) 8.3 Can Africa catch up?

Contemporary scholarly debate has focused on the prospects for Africa’s catching up,
and the avenues for achieving this. This is an even more critical agenda for African
policymakers and Africans in general. Most views on this subject focus on external
factors (resource booms and prices, FDI flows, cost of borrowing, etc.) or on the
unconditional convergence that is supposed to flow merely by liberalizing trade. This
book is premised on the argument that external factors are relevant only if the internal
dynamics are supportive. Despite global excitement about ‘Africa Rising’, ‘Lions on
the Move’, and so on, strong strains of pessimism persist about the prospects for
African industrialization and catch up. One strain comes from the long tradition,
evident in some UNCTAD publications, for example, that emphasizes binding
international constraints. In this, it harks back to variants of the Prebisch–Singer
hypothesis about the adverse terms of trade for exports from low-income countries.
Another strain, noted in Chapter 7, comes from economists like Dani Rodrik, who
otherwise acknowledge the role that industrial policy has played elsewhere but are
pessimistic about the chances for successful industrialization in Africa.
Although economic history does not provide all the answers, it does provide a clue as
to future trends, and may even provide (partial) answers to current challenges. The
history of industrialization shows that some late developers have indeed caught up
with advanced capitalist economies in terms of income levels, productivity, and
technical prowess. The most notable nineteenth-century examples are the US and
Germany. Both these late developers have sustained their leadership into our own
century. During the first half of the twentieth century, nobody imagined that the
greatest catch up and late industrialization would take place in East Asia. Japan was
devastated during the Second World War and Korea crumbled after the Korean War,
which in turn followed decades of deeply resented colonial oppression by Japan.
Taiwan seemed to many to be little more than a backward refuge for the corrupt
losers in China’s political upheavals. But after a few decades, Japan succeeded in
becoming a leading industrialized economy in the world. By the end of the twentieth
century, (South) Korea and Taiwan had undergone the fastest industrialization in
history to reach the upper rungs of the ladder. In three decades, China has emerged as
the second biggest economy, built the biggest sovereign (p.293) wealth fund, and
become a global manufacturing powerhouse.4 These are economic ‘miracles’ of the
not so remote past. Despite the successes of these countries, history has shown that
only a few such countries forged ahead, while most countries struggled as economic
laggards.
It is highly probable that the fate of Africa will not be different from this trend. Africa
(with more than fifty countries) is heterogeneous, with diverse political economies,
histories, cultures, and geopolitics. Many African countries (including the most
populous, such as Ethiopia and Nigeria) have great internal diversity. Most of them
have not enjoyed even moderate economic growth for many years, and have suffered
because of the colonial legacy, misguided IFI prescriptions, weak states and
fragmented political economies, and plain bad policy. Globalization has been a major
challenge, with Asian giants offering stiff competition even in labour-intensive
industries. Market fragmentation and weak infrastructure represent additional
challenges. Nonetheless, Africa has a chance to exploit trade, financing, and
investment opportunities with emerging countries. Increased labour costs in China
and other emerging economies are also an opportunity. This suggests some African
countries can catch up, while many may fail to seize the new opportunities.
The specific policies that are required are neither automatic nor the same, as contexts
differ. But even those who succeed will have to surmount the huge challenge of low
levels of industrialization and technology and the backwardness of their economies.
While East Asian countries had some industrial basis when they embarked on rapid
industrialization, most African countries have none. On the other hand, the
experience of the leather and leather goods industry discussed in this book suggests
that a manufacturing past does not automatically pave the way for success. According
to Hirschman (1958: 109), ‘The lack of interdependence and linkage is…one of the
most typical characteristics of underdeveloped economies…Agriculture in general,
and subsistence agriculture in particular, are…characterized by scarcity of linkage
effects.’ This necessitates an economic development strategy that induces sustained
forward movement. Hirschman (1958: 63, 66, 88) emphasizes that:
(p.294) The real scarcity in underdeveloped countries is not the resources
themselves but the ability to bring them into play…If the economy is to be
kept moving ahead, the task of development policy is to maintain tensions,
disproportions, and disequilibria…The investment that is induced by
complementarity effects may help to bring about a real transformation of an
underdeveloped economy…[O]ne disequilibrium calls forth a development
move which in turn leads to a similar disequilibrium and so on ad
infinitum…Once economic progress in the pioneer countries is a visible
reality, the strength of the desire to imitate and follow suit, to catch up
obviously becomes an important determinant of what will happen among
non-pioneers.

Such a possibility gives hope to latecomers from Africa, and can provide
opportunities for neighbours as well as being a source of motivation and experience.

8.3.1 The ‘soul’ of industrial policy and catching up

Successful catching up has depended on active industrial policy. What distinguishes
industrial policy is its ‘soul’, that is, the purpose and underlying pattern of industrial
policy. The basic pattern of active industrial policy is essentially the same, although
the tools change to fit the domestic and international context. Similar instruments
may be used for different purposes, and the nature of the industrial policy is what
determines the outcome of the policy. Reinert (2010: 34) highlights that ‘any policy
recommendation will depend totally on context and structural issues, and therefore on
specific knowledge’. As witnessed in many African countries, industrialization
continues to play a minor role, and the economic reforms promoted by IFIs have done
little to promote new industries (Watanabe and Hanatani 2012). To achieve catch up
(and take-off), African governments will have to successfully address the key
challenges of late development, that is, to promote institutional innovations that
create and concentrate investment and foster productivity and learning (creating
Verdoorn effects).
An active industrial policy, while initially dependent on and overlapping with a
Ricardian strategy (relying on comparative advantage in agricultural exports and low-
value light manufacturing), will eventually shift its focus to a more Kaldorian
strategy. A Ricardian strategy on its own (p.295) can neither bring structural change
to the economy nor achieve catch up.5 Ultimately, it is a Kaldorian strategy (which
partly ignores factor disadvantages or advantages, focuses on manufacturing exports,
and is investment-driven) that can address the challenges of catch up in terms of
investment concentration, learning, and innovation.
As discussed in Chapter 2, Kaldor’s growth laws emphasize manufacturing as an
engine of growth, the strategic role of exports, and the fundamental importance of the
balance of payments (Thirlwall 2002). But history repeatedly demonstrates that
realizing the benefits of these ‘laws’ and managing the structural balance of payments
constraint on low-income countries can only be achieved by an activist or
developmental state. The biggest advantage of latecomers is Gerschenkron’s ‘relative
degree of backwardness’ vis-à-vis industrial forerunners. According to Gerschenkron,
‘a point will be reached at which the advantages implied in rapid economic
development will more than offset those obstacles to economic progress, which are
inherent in the state of economic backwardness’ (Gerschenkron 1955: 13;
Hirschman 1958: 8). In contrast to neoclassical thinking, he emphatically stressed that
the degree of state intervention increases with the relative degree of backwardness
between latecomer and forerunner. The experience of the Asian forerunners is a great
example for many African countries, and has gradually inspired many African
policymakers, not least in Ethiopia.
The above discussion reinforces the argument that active industrial policies and an
activist state go hand in hand, and that an active industrial policy must essentially
focus on manufacturing and exports (because of the dynamics of increasing returns,
learning by doing, spill-over, and linkages). It is also worth mentioning that cheap
labour is no guarantee of sustained economic growth, while sustaining productivity in
line with international competition is. Continuously upgrading the technological basis
of industries and diversifying into new activities and industries is the essence of an
active industrial policy. This is what climbing the ladder means.
(p.296) Both economic theory (within a particular tradition) and, arguably more
important, economic history suggest that there is no reason for African countries not
to catch up with more advanced economies. The external environment is
unpredictable and often hostile, even when prices and foreign capital flows and so on
appear favourable. Successful catching up has to rely fundamentally on internal
changes and policies that push structural change, whatever the state of the external
environment. This book has, hopefully, shown that there is evidence of
‘developmentalism’ at work in the specific context and conditions of Ethiopian
manufacturing. In one of the poorest countries of the world where manufacturing
accounts for one of the smallest shares of total economic activity, processes of
structural change are under way, and they owe a great deal (more than is typically
understood or predicted) to ongoing learning by doing and learning by copying in
policymaking.
Furthermore, the Ethiopian experience shows that development policy is complex and
highly contested, and that industrial policy is no different. Carefully thought-through
policies are necessary but not sufficient to produce desired results. Policies often
yield intended outcomes only when driven by ‘transformative’ institutions, and where
there is strong state capacity (and adaptive capacity) to pursue goals and enough
flexibility to allow for course changes when things go wrong. These attributes have to
become embedded in institutionalized policy learning. Policymaking is a hotbed of
conflicting interests and groups jockeying to achieve narrow sectoral objectives.
Thus, the state is constantly involved in reconciling intersectoral and intra-group
competition without sacrificing the most sacred politic goal, the ‘national project’ of
structural transformation.
Will Africa forge ahead in the twenty-first century? As Hirschman always stressed, a
realistic but optimist perspective is essential to ‘prove Hamlet wrong’
(Adelman 2013). And there may be greater scope for African governments to prove
the many doubting Hamlets of development economics wrong if they are able to
carve out greater policy space, and then use it more effectively. As Thirlwall
highlights (2002: 77–8):
Ajit Singh tells how Nicholas Kaldor taught him three things. ‘First, the
only way for a country to develop is to industrialise; second, the only way
for a country to industrialise is to protect itself; and third, anyone who says
otherwise is being dishonest! The developed economies do preach double
standards.’ They preach free trade for developing countries, yet protect
their own markets.

Notes:
(1) Policy space refers to the ‘various tensions between national policy autonomy, policy
effectiveness and international economic integration’ UNCTAD (2014: vii). Globalization, market
internationalization, and legal agreements (multilateral, regional, and bilateral) create obligations
that undermine the scope of national policy.

(2) While much evidence has been presented in previous chapters, additional
evidence of possible value to researchers and policymakers may be provided by the
author. Much of this evidence would have been more difficult, even impossible, for
‘outsiders’ to assemble.
(3) Indeed, it may even be driven by something akin to the ‘mimetic envy’ that is at
the core of Girard’s work (see, for example, Girard 1977).
(4) Indeed, in mid-2014 the International Comparison Project and IMF growth
projections suggested the Chinese economy would outgrow the US economy sooner
than earlier forecasts (for instance, see Huffington Post (2014)).
(5) The availability of cheap labour is repeatedly raised, while the centrality of labour
productivity is ignored in most debates. For instance, Japan has been able to sustain
labour productivity, matching increased labour costs for almost fifty years. In
contrast, evidence suggests that in many countries (such as Vietnam) labour costs
have exceeded productivity growth, resulting in the relocation of some factories to
other countries with cheap labour. So, the only guarantee of sustained
competitiveness is a focus on productivity growth.

Bibliography
Source:
Made in Africa
Publisher:
Oxford University Press

(p.297) General bibliography

Abramovitz, Moses (1994) ‘Catch-up and convergence in the growth boom and
after’. In Convergence of productivity: Cross-national studies and historical
evidence, edited by William Baumol, Richard Nelson, and Edward Wolff. Oxford:
Oxford University Press.
Adelman, Jeremy (2013) Worldly Philosopher: The Odyssey of Albert O. Hirschman.
Princeton: Princeton University Press.
AfDB (2011) African Economic Outlook 2011: Africa and its emerging partners.
African Development Bank.
AFP (Agence France-Presse) (2003) ‘Ethiopia rejects IMF proposal to privatize loss-
making state firms’. 31 August.
Aghion, Beatriz Armendariz (1999) ‘Development banking’. Journal of Development
Economics 58(1):83–100.
Aklilu, Yakob, and Mike Wekesa (2002) Drought, livestock and livelihoods: Lessons
from the 1999–2001 emergency response in the pastoral sector in Kenya. London:
Humanitarian Practice Network, Overseas Development Institute.
Aktouf, Omar (2004) ‘The false expectations of Michael Porter’s strategic
management framework’. Revista Universidad & Empresa 6(6):9–41. Available at
<http://essay.utwente.nl/65339/1/D%C3%A4lken_BA_MB.pdf>
Aktouf, Omar (2005) ‘The False Expectations of Michael Porter’s Strategic
Management Framework’. Revista Gestao e Planejameto Ano 6, 1(11) (January–June
2005):75–94. Available at <http://www.revistas.unifacs.br/index.php/rgb/article/view/199/207>
Altenburg, Tilman (2010) Industrial Policy in Ethiopia. Discussion Paper, Bonn.
Altenburg, Tilman (2011) Industrial Policy in Developing Countries: Overview and
Lessons from Seven Country Cases. Discussion Paper 4/2011. Bonn, German
Development Institute. Available at <http://www.die-gdi.de/en/discussion-
paper/article/industrial-policy-in-developing-countries-overview-and-lessons-from-seven-country-
cases/>
Altenburg, Tilman (2013) ‘Can Industrial Policy Work under Neopatrimonial Rule?’.
In Pathways to Industrialization in the Twenty-First Century: New Challenges and
Emerging Paradigms, edited by Adam Szirmai, Wim Naudé, and Ludovico Alcorta.
Oxford: Oxford University Press.
(p.298) Amsden, Alice (n.d.), ‘Chapter Four: Getting Property Rights “Wrong”’, in
Developing from Role Models. Unpublished. Available at <http://ubuntuatwork.org/wp-
content/uploads/2010/02/Developing-From-Role-Models-Alice-Amsden.pdf> (accessed 5 May
2014).
Amsden, Alice H. (1989) Asia’s Next Giant: South Korea and Late
Industrialization. Oxford: Oxford University Press.
Amsden, Alice H. (1990) ‘Third World Industrialization: “Global Fordism” or a New
Model?’. New Left Review I(182):5–31. Available at <http://newleftreview.org/I/182/alice-
h-amsden-third-world-industrialization-global-fordism-or-a-new-model>
Amsden, Alice H. (2001) The ‘Rise of the Rest’: Challenges to the West from Late-
Industrializing Economies. Oxford: Oxford University Press.
Amsden, Alice H. (2007a) Escape from Empire: The Developing World’s Journey
through Heaven and Hell. Cambridge MA: MIT Press.
Amsden, Alice H. (2007b) Nationality of Firm Ownership in Developing Countries:
Who Should ‘Crowd Out’ Whom in Imperfect Markets? Momigliano Lecture 2007.
Amsden, Alice H. (2008) ‘The Wild Ones: Industrial Policies in the Developing
World’. In The Washington Consensus Reconsidered, edited by Narcis Serra and
Joseph E. Stiglitz. Oxford: Oxford University Press.
Amsden, Alice H. (2009) Grass Roots War on Poverty. Available at
<http://werdiscussion.worldeconomicsassociation.org/wp-content/uploads/Amsden-Poverty-
reformatted.pdf>
Amsden, Alice H., and Wan-Wen Chu (2003) Beyond Late Development: Taiwan’s
Upgrading Policies. Cambridge MA: MIT Press.
Amsden, Alice H., and Takashi Hikino (1994) ‘Staying Behind, Stumbling Back,
Sneaking Up and Soaring Ahead: Late Industrialization in Historical Perspective’.
In Convergence of productivity: Cross-national studies and historical evidence,
edited by William Baumol, Richard Nelson, and Edward Wolff. Oxford: Oxford
University Press.
Ancharaz, Vinaye D. (2003) Trade Policy, Trade Liberalization and Industrialization
in Mauritius: Lessons for Sub-Saharan Africa. November. Department of Economics
and Statistics, University of Mauritius. Available at <http://www.docin.com/p-
67824501.html>
Anderson, Benedict (1991) Imagined Communities: Reflections on the Origins and
Spread of Nationalism. London: Verso.
Ash, A.J., J.A. Bellamy, and T.G.H Stockwell (1994) ‘State and transition models for
rangelands: Application of state and transition models to rangelands in northern
Australia’. Tropical Grasslands 28:223–8.
Austin, Ian Partick (2009) Common Foundations of American and East Asian
Modernisation: From Alexander Hamilton to Junichero Koizumi. Singapore: Select
Publishing.
Balema, Adis Alem (2014) Democracy and Economic Development in
Ethiopia. Trenton NJ: Red Sea Press.
Bayliss, Kate (2006) ‘Privatization theory and practice: A critical analysis of policy
evolution in the development context’. In The New Development Economics: After
the Washington Consensus, edited by K.S. Jomo and Ben Fine. New Delhi: Tulika
Books.
(p.299) Bayliss, Kate, and Ben Fine (2008) Privatization and alternative public
sector reform in Sub-Saharan Africa: Delivering on Electricity and Water. New York:
Palgrave Macmillan.
BBC (2003) ‘Ethiopia Hits Out at IMF’. BBC News, 1 September 2003.
Behnke, Roy (2010) The Contribution of Livestock to the Economies of IGAD
Member States. Study Findings, Application of the Methodology in Ethiopia and
Recommendations for Further Work. IGAD LPI Working Paper No. 02. United
Kingdom: Odessa Centre. Available at
<http://www.fao.org/fileadmin/user_upload/drought/docs/IGAD%20LPI%20WP%2002-10.pdf>
Behnke, Roy, and Ian Scoones (1992) Rethinking Range Ecology: Implications for
Rangeland Management in Africa. Dryland Networks Programme, Overseas
Development Institute, Issues Paper No. 33, March.
Bhagwati, Jagdish (1989a) ‘US Trade Policy at Crossroads’. World Economy 12(4):
439–80.
Bhagwati, Jagdish (1989b) Protectionism. Cambridge MA: MIT Press.
Bhowon, Veeping, Narainduth Boodhoo, and Pynee A. Chellapermal (2004)
‘Mauritius: Policy making in Africa’. In Politics of Trade and Industrial Policy in
Africa: Forced Consensus?, edited by Charles Soludo, Osita Ogbu, and Ha-Joon
Chang. Trenton NJ: Africa World Press.
Brautigam, Deborah (2011) The Dragon’s Gift: The Real Story of China in Africa.
Oxford: Oxford University Press.
Brautigam, Deborah, and Tania Diolle (2009) ‘Coalitions, Capitalists and Credibility:
Overcoming the crisis of confidence at independence in Mauritius’. April 2009.
Available at <http://publications.dlprog.org/Coalitions,%20Capitalists%20and
%20Credibility.pdf>
Brian.Carnell.com (2003) ‘Ethiopia and the International Monetary Fund at
Loggerheads Over Privatization’. Available at
<https://brian.carnell.com/articles/2003/ethiopia-and-the-international-monetary-fund-at-
loggerheads-over-privatization/> (accessed 22 February 2015).
Buur, Lars (2014) The Development of Natural Resource Linkages in Mozambique:
The Ruling Elite Capture of New Economic Opportunities. DIIS Working Paper No.
03. Available at
<http://rucforsk.ruc.dk/site/files/48177968/Buur_The_development_of_natural_resource_linkages_
in_Mozambique_December_27.docx>
Buur, Lars, Obede Baloi, and Carlota Mondlane Tembe (2012) Mozambique
Synthesis Analysis: Between Pockets of Efficiency and Elite Capture. DIIS Working
Paper No. 01. Available at <http://transparentsea.co/images/f/fe/WP2012-01-
Mozambique_web.pdf>
Cartwright, Nancy, and Jeremy Hardie (2012) Evidence Based Policy: A Practical
Guide to Doing it Better. Oxford: Oxford University Press.
CBI (2002) ‘Cut flower and Foliage’, EU Strategic Marketing Guide. Centre for
Promotion of Imports from Developing Countries, Ministry of Foreign Affairs, The
Netherlands. Available at <http://www.cbi.eu>
CBI (2013) Cut flower market information. Centre for the Promotion of Imports,
Ministry of Foreign Affairs of the Netherlands. Available at <http://www.cbi.eu/market-
information/cut-flowers-foliage>
(p.300) Chabal, Patrick, and Jean-Pascal Daloz (1999) Africa Works: Disorder as
political instrument. Oxford: James Currey.
Chandler, Alfred (2004) Scale and Scope: The dynamics of industrial
capitalism. Cambridge MA: Harvard University Press.
Chandler, Alfred, Franco Amatori, and Taksashi Hikino (eds.) (1997) Big Business
and the Wealth of Nations. Historical and Comparative Contours of Big Business.
Cambridge: Cambridge University Press.
Chang, Ha-Joon (1994) The Political Economy of Industrial Policy. London:
Macmillan.
Chang, Ha-Joon (1999) ‘The Economic Theory of the Developmental State’. In The
Developmental State, edited by Meredith Woo-Cumings. Ithaca: Cornell University
Press.
Chang, Ha-Joon (2003a) Globalisation, Economic Development and the Role of the
State. London: Zed Books.
Chang, Ha-Joon (2003b) Kicking Away the Ladder: Development Strategy in
Historical Perspective. London: Anthem Press.
Chang, Ha-Joon (ed.) (2006) Rethinking development economics. London: Anthem
Press.
Chang, Ha-Joon, and Ajit Singh (1997) ‘Can large firms be run efficiently without
being bureaucratic?’. Journal of International Development 9(6):865–75.
Chenery, Hollis (1960) ‘Patterns of Industrial Growth’. American Economic Review,
vol. L, No. 4, September:624–54.
Chenery, Hollis, Sherman Robinson, and Moshe Syrquin (eds.)
(1986) Industrialization and growth: A comparative study. Oxford: Oxford University
Press.
China Cement Association (2013) ‘The development of China cement industry’.
Presentation of Lei QIanzhi, President of China Cement Association in Izmir, Turkey,
in October 2011. <http://www.tcma.org.tr/images/file/Cin%20Cimento%20Birligi%20Baskani
%20Lei%20QIANZHI%20.pdf>
China Leather Industry Association (2012). Available at
<http://www.chinaleather.org/eng/>
Clapham, Christopher ([1985] 2004) Third World Politics: An Introduction. Taylor
and Francis e-Library.
Clapham, Christopher (1996) ‘Governmentality and economic policy in sub-Saharan
Africa’. Third World Quarterly 17(4):809–24.
Clapham, Christopher (2009) ‘Post-war Ethiopia: The trajectories of crises’. Review
of African Political Economy 36:120, 181–92. Available at
<http://www.tandfonline.com/doi/abs/10.1080/03056240903064953#.Ui4wKMaTiSo>
Collier, Paul (2006) Africa: Geography and growth. Centre for the Study of African
Economies, Oxford University, July 2006. Available at
<http://users.ox.ac.uk/~econpco/research/pdfs/AfricaGeographyandGrowth.pdf>
Collier, Paul (2007) ‘Growth Strategies for Africa’. Paper prepared for the Spence
Commission on Economic Growth, Centre for the Study of African Economies,
Oxford.
Collier, Paul (2009) The political economy of state failure. Revised March 2009.
Available at <http://www.iig.ox.ac.uk/output/articles/OxREP/iiG-OxREP-Collier.pdf>
Collier, Paul, and Willem Gunning (1999) ‘Why has Africa grown slowly?’ Journal
of Economic Perspectives 13(3):3–22.
(p.301) Commission for Africa (2005) Our Common Interest: Report of the
Commission for Africa. Available at <http://www.commissionforafrica.info/wp-
content/uploads/2005-report/11-03-05_cr_report.pdf>
Cramer, Christopher (1999a) ‘Can Africa Industrialize by Processing Primary
Commodities? The Case of Mozambican Cashew Nuts’. World
Development 27(7):1247–66.
Cramer, Christopher (1999b) The Economics and Political Economy of Conflict in
Sub-Saharan Africa. Centre for Development Policy and Research, Discussion Paper
1099. Available at <http://www.soas.ac.uk/cdpr/publications/papers/file24327.pdf>
Cramer, Christopher (1999c) Privatisation and the Post-Washington Consensus:
Between the lab and the real world? Centre for Development Policy and Research,
Discussion Paper 0799.
Cramer, Christopher (1999d) Consultancy study on Ethiopian leather sector,
conducted for the Government of Ethiopia, 1999. Unpublished.
Cramer, Christopher (2000) Privatisation and Adjustment in Mozambique: A Hospital
Pass? SOAS Department of Economics Working Paper Number 111.
Cramer, Christopher (2006) Civil War is Not a Stupid Thing: Accounting for Violence
in Developing Countries. London: Hurst.
Cramer, Christopher (2012) ‘East Africa between tragedy and hyperbole’.
Unpublished presentation, SOAS.
CSTEP (2012) A study of energy efficiency in the Indian cement industry. Bengalaru:
Center for Study of Science, Technology and Policy.
De Waal, Alex (2012) ‘The theory and practice of Meles Zenawi’. African
Affairs 112(446):148–55.
Di John, Jonathan (2006) The Political Economy of Taxation and Tax Reform in
Developing Countries. WIDER Research Paper 2006/74. Helsinki: UNU-WIDER.
Di John, Jonathan (2008) ‘Why is the Tax System so Ineffective and Regressive in
Latin America?’ Development Viewpoint, No. 5, June.
Di John, Jonathan (2009) From Windfall to Curse? Oil and Industrialization in
Venezuela, 1920 to the Present. University Park: Penn State University Press.
Di John, Jonathan (forthcoming) Manufacturing Brazil: Social Order and
Industrialization, 1990 to the Present.
Diamond, William (1957) Development Banks. Washington DC: Mimeo: AFD.
Economic Development Institute, International Bank for Reconstruction and
Development.
Diamond, William, and V.S. Raghavan (eds.) (1982) Aspects of development bank
management. London: Johns Hopkins University Press.
Dinh, Hinh T., Vincent Palmade, Vandana Chandra, and Frances Cossar (2012) Light
Manufacturing in Africa: Targeted policies to enhance private investment and create
jobs. Washington DC: World Bank.
Doner, Richard, Bryan Ritchie, and Dan Slater (2005) ‘Systemic vulnerability and the
origins of developmental states: Northeast and Southeast Asia in comparative
perspective’. International Organization 59(2):327–61.
Dos Santos, Paulo (2011) ‘A Policy Wrapped in Analysis: The World Bank’s Case for
Foreign Banks’. In The Political Economy of Development: The World Bank,
Neoliberalism (p.302) and Development Research, edited by Kate Bayliss, Ben Fine,
and Elisa Van Waeyenberge. London: Pluto.
Drucker, Peter (1999) Management: Tasks, Responsibilities, Practices. Oxford:
Butterworth-Heinemann.
Earle, Edward Mead (1986) ‘Adam Smith, Alexander Hamilton, Friedrich List: The
Economic Foundations of Military Power’. In Makers of Modern Strategy from
Machiavelli to the Nuclear Age, edited by Peter Paret. Oxford: Clarendon Press.
Easterly, William (2002) The Elusive Quest for Growth: Economist’s Adventures and
Misadventures in the Tropics. Cambridge MA: MIT Press.
Easterly, William, and Ross Levine (1997) ‘Africa’s growth tragedy: Policies and
ethnic divisions’. Quarterly Journal of Economics 112(4):1203–50.
EEA (2013) Ethiopia financial sector performance review. Proceedings of the Third
International Conference on the Ethiopian Economy. Ethiopian Economic
Association. Addis Ababa: Ethiopian Economic Association. Available at
<http://www.eeaecon.org/node/4944>
Embassy of Japan (2008) A Series of Studies on Industries in Ethiopia. Addis Ababa:
Embassy of Japan. Available at <http://www.et.emb-japan.go.jp/Eco_Research_E.pdf>
Evans, Peter (1995) Embedded Autonomy: States and Industrial
Transformation. Princeton: Princeton University Press.
Evans, Peter (1997) ‘State structures, government-business relations and economic
transformation’. In Business and the state in developing countries, edited by Sylvia
Maxfield and Ben Schneider. London: Cornell University Press.
FAO (2008) Global hides and skins market: Review of 2004–2007 and prospects for
2008. Food and Agriculture Organization of the United Nations. Available at
<www.fao.org>
FAO (2009) Global hides and skins markets: A review and short-term outlook (As of
November 2009). Food and Agriculture Organization of the United Nations.
Available at <http://www.fao.org/economic/est/est-commodities/hides-skins/en/>
FAO (2011) World statistical compendium for raw hides and skins, leather and
leather footwear 1992–2011. Food and Agriculture Organization of the United
Nations. Available at
<http://www.fao.org/fileadmin/templates/est/COMM_MARKETS_MONITORING/Hides_Skins/D
ocuments/AA_COMPENDIUM_2011.pdf>
FAO (2013) World Statistical Compendium for raw hides and skins, leather and
leather footwear 1993–2012. Food and Agriculture Organization of the United
Nations. Available at
<http://www.fao.org/fileadmin/templates/est/COMM_MARKETS_MONITORING/Hides_Skins/D
ocuments/COMPENDIUM2013.pdf>
FAO/GOB (2013) Botswana agrifood value chain project: Beef value chain
study. Food and Agriculture Organization of the United Nations and the Government
of Botswana.
FDDI (2012) Presentation on leather products industry. Unpublished Notes, Indian
Footwear Design and Development Institute, Addis Ababa, Ethiopia.
Fine, Ben (1996) ‘Introduction: State and development’. Journal of Southern African
Studies 22(1):5–26.
(p.303) Fine, Ben (2006) ‘The developmental state and the political economy of
development’. In The New Development Economics: After the Washington
Consensus, edited by K.S. Jomo and Ben Fine. London: Zed Books.
FloraHolland (2010) Abridged 2010 Annual Report. Available at
<http://www.floraholland.com/media/168711/Jaarverslag%202010%20Eng.pdf>
FloraHolland (2012) Annual Report 2011.
<http://www.floraholland.com/media/477979/jaarverslag_floraholland_2011_engels.pdf>
FloraHolland (2014) ‘FloraHolland books small profit in turbulent year for the
floriculture industry’. Press Release. Available at
<https://www.floraholland.com/en/speciale-paginas/search-in-news/v20291/floraholland-books-
small-profit-in-turbulent-year-for-the-floriculture-industry/>
Fosu, Augustin K. (1992) ‘Political instability and economic growth: Evidence from
sub-Saharan Africa’. Economic Development and Cultural Change 40(4):829–41.
Fosu, Augustin K. (2012) ‘The African economic growth record, and the roles of
policy syndromes and governance’. In Good growth and governance in Africa:
Rethinking development strategies, edited by Akbar Noman, Kwesi Botchwey,
Howard Stein, and Joseph E. Stiglitz. Oxford: Oxford University Press.
Frank, André Gunder (1967) Capitalism and underdevelopment in Latin America:
Historical studies of Chile and Brazil. New York: Monthly Review Press.
Gebreeyesus, Mulu, and Michiko Iizuka (2010) Discovery of the flower industry in
Ethiopia: Experimentation and coordination. UNU-MERIT Working Paper Series No.
25.
Gereffi, Gary (1994) ‘The organization of buyer driven global commodity chain:
How US retailers shape overseas production networks’. In Commodity Chains and
Global Capitalism, edited by G. Gereffi and M. Korzeniewicz. London: Praeger.
Gereffi, Gray, and Karina Fernandez-Stark (2011) Global Value Chain Analysis: A
Primer Centre on Globalization, Governance and Competitiveness. 31 May. Available
at <http://www.cggc.duke.edu/pdfs/2011-05-31_GVC_analysis_a_primer.pdf>
Gereffi, Gary, John Humphrey, and Timothy Sturgeon (2005) ‘The governance of
global value chains’. Review of International Political Economy 12(1):78–104.
Gerschenkron, Alexander (1962) Economic Backwardness in Historical Perspective.
Cambridge MA: Harvard University Press.
Gerschenkron, Alexander (1955) ‘The Problem of Economic Development in Russian
Intellectual History.’ In Continuity and Change in Russian and Soviet Thought, ed. E.
J. Simmons. Cambridge, MA: Harvard University Press.
Giezen, Mendel, Luca Bertolini, and Willem Salet (2014) ‘Adaptive Capacity within
a Mega Project: A Case Study on Planning and Decision-Making in the Face of
Complexity’. European Planning Studies. Available
at <http://www.tandfonline.com/doi/pdf/10.1080/09654313.2014.916254#.VFK6jFdBeE8>
Girard, Rene (1977) Violence and the Sacred. Baltimore: Johns Hopkins University
Press.
GIZ (2009) Leather and footwear industry in Vietnam: The labour markets and
gender impact of the global economic slowdown on value chains. Eschborn:
Deutsche Gesellschaft für Internationale Zusammenarbeit.
(p.304) Global Cement (2012) Top global cement companies. 17 December. Available
at <www.globalcement.com>
Global Development Solutions (2011) The Value Chain and Feasibility Analysis;
Domestic Resource Cost Analysis. Background paper, World Bank, Washington DC.
(Light Manufacturing in Africa Study). Available at
<http://econ.worldbank.org/africamanufacturing>
Goodrich, Carter (1965) The Government and the economy, 1783–1861. New York:
Bobbs-Merrill.
Grabel, Ilene (ed.) (2003) ‘International Private Capital Flows and Developing
Countries’. In Rethinking development economics, edited by Ha-Joon Chang.
London: Anthem Press.
de Haan, Cornelis, and Dina Umali (1994) ‘Public and private sector roles in the
supply of veterinary services’. World Bank Research Observer 9(1):71–96.
Hagmann, Tobias, and Jon Abbink (2012) ‘Twenty years of revolutionary democratic
Ethiopia, 1991–2011’. Journal of Eastern African Studies 5(4):579–95.
Hagmann, Tobias, and Jon Abbink (eds.) (2013) Reconfiguring Ethiopia: The politics
of authoritarian reform. London: Routledge.
Hall, Peter (1987) Governing the Economy: The politics of state intervention in
Britain and France. Cambridge: Polity Press.
Hamilton, Alexander (1934) Papers on Public Credit, Commerce and Finance, ed.
Samuel McKee. New York: Columbia University Press.
Hardin, Garrett (1968) ‘The Tragedy of the Commons’. Science, new series,
162(3859):1243–48.
Henderson, Vernon (1974) ‘The sizes and types of cities’. American Economic
Review 64(4):640–56.
Henderson, Vernon (2003) ‘Marshall’s scale economies’. Journal of Urban
Economics 53:1–28.
Herring, Roland J. (1999) ‘Embedded particularism: India’s failed developmental
state’. In The Developmental State, edited by Meredith Woo-Cumings. Ithaca:
Cornell University Press.
Hirschman, Albert O. (1958) Strategy of Economic Development. New Haven and
London: Yale University Press.
Hirschman, Albert O. (1963) Journeys toward Progress: Studies of economic policy-
making in Latin America. New York: Twenty Century Fund.
Hirschman, Albert O. (1967) Development Projects Observed. Washington DC:
Brookings Institution.
Hirschman, Albert O. (1968) ‘The Political Economy of Import-Substituting
Industrialization in Latin America’. Quarterly Journal of Economics 82(1):1–32.
Hirschman, Albert O. (1971) A Bias for Hope: Essays on development and Latin
America. New Haven: Yale University Press.
Hirschman, Albert O. ([1981] 2008) Essays in Trespassing: Economics to politics
and beyond. Cambridge: Cambridge University Press.
Hirschman, Albert O. ([1986] 1992) Rival views of market society and other recent
essays. Cambridge MA: Harvard University Press.
(p.305) Hirschman, Albert O. (1991) The Rhetoric of Reaction: Perversity, Futility,
Jeopardy. London: Harvard University Press.
Hirschman, Albert O. (2013) The Essential HIRSCHMAN, edited and with an
introduction by Jeremy Adelman. Princeton and Oxford: Princeton University Press.
Hobday, Michael (1995) ‘East Asian latecomer firms: Learning the technology of
electronics’. World Development 23(7): 1171–93.
Hobday, Mike (2013) ‘Learning from Asia’s success: Beyond simplistic “lesson-
making”’. In Pathways to Industrialization in the Twenty-First Century: New
Challenges and Emerging Paradigms, edited by Adam Szirmai, Wim Naudé, and
Ludovico Alcorta. Oxford: Oxford University Press.
HSRG (2012) Human security report 2012. Human Security Research Group.
Available at <http://www.hsrgroup.org/human-security-reports/2012/overview.aspx>
Huffington Post (2014) ‘China To Have World’s Largest Economy This Year: World
Bank’, 5 March. Available at <http://www.huffingtonpost.ca/2014/05/03/china-worlds-largest-
economy_n_5255825.html>
Humphrey, John, and Hubert Schmitz (2004) ‘Chain governance and upgrading:
Taking stock’. In Local Enterprises in the Global Economy, edited by Hubert
Schmitz. Cheltenham: Edward Elgar.
IDS (1997) Collective Efficiency: A Way Forward For Small Firms. IDS Policy
Briefing. Issue 10, April. Available at <http://www.ids.ac.uk/files/dmfile/PB10.pdf>
Imara (2011) Imara African cement report: Africa, the last cement frontier, February.
Available at
<http://www.cgmplc.com/content/assets/pdf/imara_201102_african_cement_report.pd>
International Cement Review (2012) Cemnet.com (South Africa’s PPC invests in
Ethiopian cement firm, By ICR Newsroom, Published 26 July 2012
<http://www.cemnet.com/News/story/150263/south-africa-s-ppc-invests-in-ethiopian-cement-
firm.html>
ITC (2012) Market news service. International Trade Centre. Available from
<http://www.intracen.org>
Jackson, Patrick (2006) Civilizing the Enemy: German Reconstruction and the
Invention of the West. Ann Arbor: University of Michigan Press.
Jalilian, Hossein, and John Weiss (1997) ‘Bureaucrats, business and economic
growth’. Journal of International Development 9:877–85.
Jalilian, Hossein, and John Weiss (2000) ‘De-industrialisation in Sub-Saharan Africa:
Myth or Crisis?’. Journal of African Economics 9(1):24–43.
Jerven, Morten (2010a) Accounting for the African Growth Miracle: The official
evidence—Botswana 1965–1995. Journal of Southern African Studies, 36(1):73–94.
Available at <http://www.tandfonline.com/doi/abs/10.1080/03057071003607337>
Jerven, Morten (2010b) African Growth Recurring: An Economic History Perspective
on African Growth Episodes, 1690–2010. Simons Papers in Security and
Development, No. 4, June.
Jerven, Morten (2011) ‘The quest for the African Dummy: Explaining African Post-
colonial economic performance revisited’. Journal of International
Development 23(2):288–307.
(p.306) Jianhua, Zhong (2013) China’s special representative on African affairs on
Trade, Aid and Jobs. Africa Research Institute, 6 August. Available at
<www.africaresearchinstitute.org/publications/ambassador-zhong-jianhua-on-trade-aid-and-jobs/>
Johnson, Chalmers (1982) MITI and the Japanese miracle: The growth of industrial
policy, 1925–1975. Stanford: Stanford University Press.
Johnson, Chalmers (ed.) (1984) The Industrial Policy Debate. San Francisco:
Institute for Contemporary Studies.
Jomo, K.S., and Rudiger von Arnim (2012) ‘Economic liberalization and constraints
to development in Sub-Saharan Africa’. In Good growth and governance in Africa:
Rethinking development strategies, edited by Akbar Noman, Kwesi Botchwey,
Howard Stein, and Joseph E. Stiglitz. Oxford: Oxford University Press.
Jourdan, Paul, Gibson Chigumira, Isaac Kwesu, and Erina Chipumho (2012) Mining
Sector Policy Study. Zimbabwe Economic Policy Analysis and Research Unit.
December.
Kaldor, Nicholas (1966) Causes of the Slow Rate of Economic Growth of the United
Kingdom. Cambridge: Cambridge University Press.
Kaldor, Nicholas (1967) Strategic Factors in Economic Development. Ithaca: New
York State School of Industrial and Labour Relations, Cornell University.
Kaplinsky, Raphael (2005) Globalization, poverty and equality: Between a rock and
a hard place. Cambridge: Polity Press.
Kaplinsky, Raphael, and Mike Morris (2000) A handbook for value chain
research. Prepared for the IDRC. <http://www.ids.ac.uk/ids/global/pdfs/VchNov01.pdf>
Kay, Cristobal (2002) ‘Why East Asia Overtook Latin America: Agrarian Reform,
Industrialisation and Development’. Third World Quarterly 23(6):1073–102.
Kelsall, Tim (2013) Business, politics and the state in Africa: Challenging the
orthodoxies on growth and transformation. London: Zed Books.
Kelsall, Tim, David Booth, Diana Cammack, and Frederick Golooba-Mutebi (2010)
Developmental patrimonialism? Questioning the orthodoxy on political governance
and economic progress in Africa. Working Paper No. 9, July. Africa Power and
Politics. Available at <http://r4d.dfid.gov.uk/PDF/Outputs/APPP/20100708-appp-working-
paper-9-kelsall-and-booth-developmental-patrimonialism-july-2010.pdf>
Khan, Azizur Rahman (2012) ‘Employment in Sub-Saharan Africa: Lessons to be
Learnt from the East Asian Experience’. In Good growth and governance in Africa:
Rethinking development strategies, edited by Akbar Noman, Kwesi Botchwey,
Howard Stein, and Joseph E. Stiglitz. Oxford: Oxford University Press.
Khan, Mushtaq H. (2000a) ‘Rents, efficiency and growth’. In Rents, rent-seeking and
economic development: Theory and evidence in Asia, edited by Mushtaq Khan and
K.S. Jomo. Cambridge: Cambridge University Press.
Khan, Mushtaq H. (2000b) ‘Rent-seeking as process’. In Rents, rent-seeking and
economic development: Theory and evidence in Asia, edited by Mushtaq Khan and
K.S. Jomo. Cambridge: Cambridge University Press.
Khan, Mustaq H. (2006) ‘Corruption and governance’. In The New Development
Economics: After the Washington Consensus, edited by K.S Jomo and Ben Fine.
London: Zed Books.
(p.307) Khan, Mushtaq H., and Stephanie Blankenburg (2009) ‘The Political
Economy of Industrial Policy in Asia and Latin America’. In Industrial Policy and
Development: The Political Economy of Capabilities Accumulation, edited by Mario
Cimoli, Giovanni Dosi, and Joseph E. Stiglitz. Oxford: Oxford University Press.
Khan, Mushtaq H., and K.S. Jomo (eds.) (2000) Rents, rent-seeking and economic
development: Theory and evidence in Asia. Cambridge: Cambridge University Press.
Kitching, Gavin (1982) Development and underdevelopment in historical
perspective: Populism, nationalism and Industrialization. London: Methuen.
Kohli, Atul (2004) State-Directed Development: Political Power and
Industrialization in the Global Periphery. Cambridge: Cambridge University Press.
KOICA (2013) Project for Establishment of the Detailed Action Plans Strategy to
Implement Ethiopian National Economic Development Plan Strategy (Growth and
Transformation Plan, GTP). Korea International Cooperation Agency, Korea
International Development Institute, and Dalberg. August. Addis Ababa, Ethiopia.
Kolli, Ramesh (2010) A Study on the determination of share of the private sector in
Ethiopian Gross Domestic Product. Addis Ababa: Addis Ababa Chamber of
Commerce and Sectoral Association.
Kornai, Janos (1980) Economics of shortage. 2 vols. Amsterdam: North Holland.
Kornai, Janos (1990) The road to free economy: Shifting from a socialist system. The
Example of Hungary. New York: Norton.
Krueger, Anne (1974) ‘The political economy of rent-seeking society’. American
Economic Review 64(3):291–303.
Krueger, Anne (1980) ‘Trade policy as an input to development’. American
Economic Review 70(2):288–92.
Krueger, Anne (1990) ‘Government failure in economic development’. Journal of
Economic Perspectives 4(3):9–23.
Lall, Sanjaya (1992) ‘Technological Capabilities and Industrialization’. World
Development 20(2):165–86.
Lall, Sanjaya (1996) Learning from the Asian Tigers: Studies in Technology and
Industrial Policy. London: Macmillan.
Lall, Sanjaya (1999) Promoting Industrial Competitiveness in Developing Countries:
Lessons from Asia. Commonwealth Secretariat, Commonwealth Economics Paper
Series, Economic Paper 39. October.
Lall, Sanjaya (2000a) Selective Industrialization Trade Policies in Developing
Countries: Theoretical and Empirical Issues. QEH Working Paper No. 48.
Lall, Sanjaya (2000b) The Technological Structure and Performance of Developing
Country Manufactured Exports, 1985–1998. QEH Working Paper No. 44.
Lall, Sanjaya (2003) ‘Technology and industrial development in an era of
globalization’. In Rethinking development economics, edited by Ha-Joon Chang.
London: Anthem Press.
Lall, Sanjaya (2004) Reinventing Industrial Strategy: The Role of Government Policy
in Building Industrial Competitiveness. UNCTAD, G24 Discussion Paper Series No.
28, April. Available at <http://unctad.org/en/Docs/gdsmdpbg2420044_en.pdf>
(p.308) Lall, Sanjaya (2005) Is African Industry Competing? QEH Working Paper
Series Paper No. 121. Available at
<http://213.154.74.164/invenio/record/18469/files/lall.pdf>
Lavopa, Alejandro, and Adam Szirmai (2012) Industrialization, employment and
poverty. 2012-081. UNU-MERIT Working Paper Series, 2 November 2012. Available
at <www.merit.unu.edu/publications/wppdf/2012/wp2012-081.pdf>
LEFASO (2004) Vietnam leather and footwear industry: Opportunities and
challenges on the way Vietnam entering WTO. Presentation by Secretary General of
the Vietnam Leather and Footwear Association, Nguyen Thi Tong. Available at
<http://siteresources.worldbank.org/INTRANETTRADE/Resources/WBI-Training/288464-
1139428366112/Session6-NguyenThiTong-Footwear_EN.pdf>
LEFASO (2012) Decision on permitting to rename Vietnam Leather - Footwear
Association to Vietnam Leather, Footwear and Handbag Association. Available at
<http://www.lefaso.org.vn/default.aspx?ZID1=240&ID1=2>
Lefort, René (2007) ‘Powers—mengist —and peasants in rural Ethiopia: The May
2005 elections’. Journal of Modern African Studies 45(2):435–60.
Lefort, René (2013) ‘The theory and practice of Meles Zenawi: A response to Alex de
Waal’. African Affairs. Journal of the Royal African Society 112(448):460–70.
Lin, Justin Y. (2009) Economic Development and Transition: Thought, Strategy and
Viability. Cambridge: Cambridge University Press.
Lin, Justin and Ha-Joon Chang (2009) ‘Should Industrial Policy in Developing
Countries Conform to Comparative Advantage or Defy it? A Debate between Justin
Lin and Ha-Joon Chang’. Development Policy Review 27(5):483–502.
Lindbeck, Assar (2007) Janos Kornai’s contributions to economic analysis. IFN
Working Paper No. 724.
List, Friedrich (1827) Outlines of American Political Economy, in Series of Letters.
Letter VII. Reading. 22 July 1827. Philadelphia: Samuel Parker.
List, Friedrich ([1841] 2005) National System of Political Economy. Volume 1: The
History. New York: Cosimo.
List, Friedrich (1856) National System of Political Economy. Vol. I–IV. Memphis:
Lippincott.
Little, Peter, Roy Behnke, John McPeak, and Getachew Gebru (2010) Future
Scenarios for Pastoral Development in Ethiopia, 2010–2025. Report No. 2. Pastoral
Economic Growth and Development Policy Assessment, Ethiopia.
Mahmoud, Hussein (2010) Livestock Trade in the Kenyan, Somali and Ethiopian
Borderlands. Africa Programme. September. AFP BP 2010/02. Chatham House,
London. Available at
<https://www.chathamhouse.org/sites/files/chathamhouse/public/Research/Africa/0910mahmoud.p
df>
Mahoney, James (2000) ‘Path dependence in historical sociology’. Theory and
Society 29(4):507–48.
Mano, Yukichi and Aya Suzuki (2011) The case of the Ethiopian cut flower industry.
Hermes Ir, Japan. Technical report. Hitotsubashi University Repository Available at
<http://hermes-ir.lib.hit-u.ac.jp/rs/bitstream/10086/25583/1/070econDP13-04.pdf>
Markakis, John (2011) Ethiopia: The last two frontiers. New York: James Currey.
Marshall, Alfred (1920) Principles of Economics. London: Palgrave Macmillan.
(p.309) Mazzucato, Mariana (2011) The Entrepreneurial State. DEMOS, Open
Research Online, London. Available at
<www.demos.co.uk/publications/theentrepreneurialstate>
Mazzucato, Mariana (2013a) The entrepreneurial state: Debunking public vs. private
sector myths. London: Anthem Press.
Mazzucato, Mariana (2013b) ‘Lighting the innovation spark’. In The Great
Rebalancing: How to Fix the Economy, edited by Andrew Harrop. Fabian Society,
London. Available at <http://www.fabians.org.uk/wp-
content/uploads/2013/01/FABJ320_web.pdf>
McMillan, Margaret and Dani Rodrik (2011) Globalization, structural change and
productivity growth. NBER Working Paper No. 17143, June. Available at
<http://www.nber.org/papers/w17143>
Meek, James (2014) Private Island: Why Britain Now Belongs to Someone
Else. London: Verso.
Meisel, Nicolas (2004) Governance and development: A different perspective on
corporate governance. OECD Development Centre Studies, May.
Melese, Ayelech Tiruwha, and A.H. Helmsing, (2010) ‘Endogenisation or enclave
formation? The development of the Ethiopian cut flower industry’. Journal of
Modern African Studies48(1):35–66.
Milhaupt, Curtis, and Wentong Zheng (2014) Beyond Ownership: State Capitalism
and the Chinese Firm. Available at <http://papers.ssrn.com/sol3/papers.cfm?
abstract_id=2413019>
Mkandawire, Thandika (2001) ‘Thinking about Developmental States in
Africa’. Cambridge Journal of Economics 25(3):289–313.
Mkandawire, Thandika (2010) ‘From maladjusted states to democratic developmental
states in Africa’. In Constructing a Developmental State in South Africa: Potentials
and challenges, edited by Omano Edigheji. Pretoria: HSRC Press.
Mkandawire, Thandika (2013) Neopatrimonialism and the political economy of
economic permormance in Africa: Critical reflections. Institute for Future Studies. 22
April. Available at <http://www.iffs.se/wp-
content/uploads/2013/05/2013_1_thandika_mkandawire.pdf>
Mühlenkamp, Holger (2013) From state to market revisited: More empirical
evidence on the efficiency of public (and privately-owned) enterprises. Munich
University Library, Munich. Available at <http://mpra.ub.uni-muenchen.de/47570/>
Musacchio, Aldo, and Sergio, Lazzarini (2014a) Reinventing State Capitalism:
Leviathan in Business, Brazil and Beyond. Cambridge MA: Harvard University Press.
Musacchio, Aldo, and Sergio, Lazzarini (2014b) ‘State-owned Enterprises in Brazil:
History and Lessons’. Paper prepared for OECD Workshop on State-owned
Enterprises in the Development Process, Paris, 4 April. Revised version, 28 February.
Mwangi, Esther (2007) Socioeconomic Change and Land Use in Africa: The
Transformation of Property Rights in Maasailand. London: Palgrave Macmillan.
Nico, de Froot (1998) ‘Floriculture World Trade: Trade and Consumption Patterns’.
Agricultural Economics Research Institute (LEIDCO), The Netherlands.
Nolan, Peter (2003) ‘Industrial policy in the early 21st century: The challenge of the
Global Business Revolution’. In Rethinking development economics, edited by Ha-
Joon Chang. London: Anthem Press.
Nolan, Peter (2012) Is China Buying the World? Cambridge: Polity Press.
(p.310) Noland, Marcus, and Howard Pack (2003) Industrial policy in an era of
globalization: Lessons from Asia. Washington DC: Institute for International
Economics.
North, Douglass C. (1990) Institutional changes and economic performance.
Cambridge: Cambridge University Press.
O’Brien, Patrick K. (1991) Power with Profit: The State and the Economy 1688–
1815: An Inaugural Lecture Delivered in the University of London. London:
University of London.
Ocampo, José A. (2005) ‘The quest for Dynamic Efficiency: Structural Dynamics and
Economic Growth in Developing Countries’. In Beyond Reforms, Structural
Dynamics and Macroeconomic Vulnerability, edited by José A. Ocampo. Palo Alto:
Stanford University Press and ECLAC.
Ocampo, José A. (2007) ‘Introduction’. In UN-DESA (2007) Industrial Development
for the 21st Century: Sustainable Development Perspectives. United Nations
Department of Economic and Social Affairs, New York. Available at
<http://www.un.org/esa/sustdev/publications/industrial_development/full_report.pdf>
Ocampo, José A. (2008) ‘Introduction’. In Industrial Development for the 21st
century, edited by David O’Connor and Monica Kjollerstrom. London: Zed Books.
Ocampo, José A., Codrina Rada, and Lance Taylor (2009) Growth and policy in
developing countries: A structuralist approach. New York: Columbia University
Press.
OECD (1975) The Aims and Instruments of Industrial Policy: A Comparative
Study. Paris: Organisation for Economic Co-operation and Development.
OECD (1976) The footwear industry: Structure and governmental policies. Paris:
Organisation for Economic Co-operation and Development.
Ottaway, Marina (1999) Africa’s new leaders: Democracy or state
reconstruction? Washington DC: Carnegie Endowment for International Peace.
Ottaway, Marina (2003) ‘African priorities: Democracy isn't the place to start’. New
York Times, 23 May. Available at <http://www.nytimes.com/2003/05/23/opinion/23iht-
edmarina_ed3_.html>
Owens, Trudy, and Adrian Wood (1997) ‘Export-oriented industrialization through
primary processing?’. World Development 25(9):1453–70.
Oya, Carlos (2010) Rural inequality, wage employment and labour market formation
in Africa: Historical and micro level evidence. International Labour Organization,
Working Paper 197, August.
Oya, Carlos (2011) ‘Agriculture in the World Bank: Blighted Harvest Persists’.
In The political economy of development: The World Bank, neoliberalism and
development research, edited by Kate Bayliss, Ben Fine, and Elisa Waeyenberge.
London: Pluto.
Oya, Carlos (2012) ‘Combining Qualitative and Quantitative Methods: Issues, Tools
and Challenges’. Unpublished notes.
Padayachee, Vishnu (ed.) (2010) The Political Economy of Africa. New York:
Routledge.
Page, John (2011) ‘Can Africa industrialise?’ Journal of African Economies 21,
Supplement 2, AERC Plenary Session. Available at
<http://jae.oxfordjournals.org/content/21/suppl_2/ii86.full>
(p.311) Page, John (2013) ‘Should Africa Industrialize?’ In Pathways to
Industrialization in the Twenty-First Century: New Challenges and Emerging
Paradigms, edited by Adam Szirmai, Wim Naudé, and Ludovico Alcorta. Oxford:
Oxford University Press.
Page, John, and Mans Söderbom (2012) Is small beautiful? Small enterprise, aid and
employment in Africa. United Nations University World Institute for Development
Economics Research, Working Paper No. 2012/94.
Palma, Gabriel J. (2003) ‘The “Three Routes” to Financial Crises: Chile, Mexico and
Argentina [1]; Brazil [2]; and Korea, Malaysia and Thailand [3]’. In Rethinking
Development Economics, edited by Ha-Joon Chang. London: Anthem Press.
Palma, Gabriel J. (2009) ‘Flying geese and waddling ducks: The different capabilities
of East Asia and Latin America to “demand-adapt” and “supply-upgrade” their export
productive capacity’. In Industrial Policy and Development: The Political Economy
of Capabilities Accumulation, edited by Mario Cimoli, Giovanni Dosi, and Joseph E.
Stiglitz. Oxford: Oxford University Press.
Palma, Gabriel J. (2011) How to create a financial crisis by trying to avoid one: The
Brazilian 1999-financial collapse as ‘Macho-Moneterism’ can’t handle ‘Bubble Thy
Neighbour’ level of inflows. Cambridge Working Papers in Economics (CWPE)
1301. Available at <http://www.econ.cam.ac.uk/research/repec/cam/pdf/cwpe1301.pdf>
Palma, Gabriel J. (2012) How the full opening of the capital account to highly liquid
financial markets led Latin America to two and half cycles of ‘mania, panic and
crash’. Cambridge Working Papers in Economics (CWPE) 1201. Available at
<http://www.econ.cam.ac.uk/research/repec/cam/pdf/cwpe1201.pdf>
Pankhurst, Richard (1996) ‘Post-World War II Ethiopia: British military policy and
action for the dismantling and acquisition of Italian factories and other assets, 1941–
2’, Journal of Ethiopian Studies 29(1):35–77.
Pasinetti, Luigi (1981) Structural Change and Economic Growth: A theoretical essay
on the dynamics of the wealth of nations. Cambridge: Cambridge University Press.
Pasinetti, Luigi (1993) Structural economic dynamics: A theory of the economic
consequences of human learning. Cambridge: Cambridge University Press.
Peres, Wilson (2013) ‘Industrial policies in Latin America’. In Pathways to
Industrialization in the Twenty-First Century: New Challenges and Emerging
Paradigms, edited by Adam Szirmai, Wim Naudé and Ludovico Alcorta. Oxford:
Oxford University Press.
Polanyi, Karl (1944) The Great Transformation: The Political and Economic Origins
of our time. Boston MA: Beacon Press.
Porter, Michael E. (1998) The Competitive Advantage of Nations. New York:
Palgrave Macmillan.
Porter, Michael E. (2008) The five competitive forces that shape strategy. Harvard
Business Review. January. Reprint No. R0801E, Harvard Business Review. Available
at <http://www.exed.hbs.edu/assets/documents/hbr-shape-strategy.pdf/>
Possas, Mario, and Heloisa Borges (2009) ‘Competition Policy and Industrial
Development’. In Industrial Policy and Development: The Political Economy of
Capabilities Accumulation, edited by Mario Cimoli, Giovanni Dosi, and Joseph E.
Stiglitz. Oxford: Oxford University Press.
(p.312) Prebisch, Raúl (1950) The economic development of Latin America and its
principal problems. New York: UN Department of Economic Affairs.
Reinert, Erik S. (2009) ‘Emulation versus comparative advantage: Competing and
complementary principles in the history of economic policy’. In Industrial Policy
and Development: The Political Economy of Capabilities Accumulation, edited by
Mario Cimoli, Giovanni Dosi, and Joseph E. Stiglitz. Oxford: Oxford University
Press.
Reinert, Erik S. (2010) How Rich Countries Got Rich and How Poor Countries Stay
Poor. London: Constable.
Reuters (2009) ‘Ethiopia says recession hits Dutch flower sales’. Available at
<http://af.reuters.com/article/investingNews/idAFJOE5210BX20090302>
Rhee, Yung Whee, Bruce Ross-Larson, and Garry Pursell (2010 [1984]) How Korea
did it? From economic basket case to successful development strategy in the 1960s
and 1970s. Seoul: Random House Korea.
Ricardo, David ([1817] 2004) The Principles of Political Economy and Taxation.
New York: Dover Publications.
Riddell, Roger (1990) Manufacturing Africa: Performance and Prospects of Seven
Countries in Sub-Saharan Africa. London: James Currey.
Rikken, Milco (2011) Global Competitiveness of the Kenyan Flower Industry.
Technical Paper for Kenya Flower Council and the World Bank Group. Available at
<http://www.euacpcommodities.eu/files/2.ESA_.D02_KFC_Seminar_Issue_-
_Global_Competitiveness.pdf>
Roberts, Simon (2004) The role for competition policy in economic development:
The South African experience. Trade and Industrial Policy Strategies (TIPS) Working
Paper 8. Available at <http://crdi.ca/EN/Documents/The-Role-of-Competition-Policy-in-
Economic-Development.pdf>
Rodrik, Dani (1997) Trade Policy and Economic Performance in Sub-Saharan Africa.
Harvard University-Harvard Kennedy School, Centre for Policy Research, NBER
Working Paper No. 6562.
Rodrik, Dani (2004) ‘Industrial Policy for the 21st Century’. September. Available at
<http://www.hks.harvard.edu/fs/drodrik/Research%20papers/UNIDOSep.pdf>
Rodrik, Dani (2008a) ‘A practical approach to formulating growth strategies’. In The
Washington Consensus Reconsidered: Towards a New Global Governance, edited by
Narcis Sera and Joseph E. Stiglitz. Oxford: Oxford University Press.
Rodrik, Dani (2008b) ‘Notes on the Ethiopian economic situation: Refining Ethiopian
Industrial Policy Strategy’. Unpublished.
Rodrik, Dani (2008c) Normalizing Industrial Policy. Commission on Growth and
Development Working Paper No. 3, Washington DC.
Rodrik, Dani (2011) The Future of Economic Convergence. National Bureau of
Economic Research Working Paper 17400. Available at
<http://www.nber.org/papers/w17400.pdf>
Rodrik, Dani (2012) The Globalization Paradox: Why Global Markets, States, and
Democracy Can’t Coexist. Oxford: Oxford University Press.
Rodrik, Dani (2014) An African Growth Miracle? April. Available at
<https://www.sss.ias.edu/files/pdfs/Rodrik/Research/An_African_growth_miracle.pdf>
(p.313) Rodriguez, Franscisco, and Dani Rodrik (2001) Trade Policy and Economic
Growth: A Sketpic’s Guide to the Cross-National Evidence. Available at
<http://www.nber.org/chapters/c11058.pdf>
SaKong, Il, and Youngsun Koh (eds.) (2010) The Korean Economy: Six Decades of
Growth and Development. The Committee for the Sixty-Year History of the Korean
Economy. Seoul: The Korean Development Institute (KDI).
Schmitz, Hubert (1995a) ‘Collective Efficiency: growth path for small-scale
industry’. Journal of Development studies 31(4):529–66.
Schmitz, Hubert (1995b) ‘Small shoemakers and Fordist giants: Tale of a super
cluster’. World Development 23(1):9–28.
Schmitz, Hubert (1998) Responding to Global Competitive Pressure: Local Co-
operation and Upgrading in the Sino’s Valley, Brazil. IDS Working Paper 82.
Available at <https://www.ids.ac.uk/files/Wp82.pdf>
Schmitz, Hubert (1999) ‘Global competition and local cooperation: Success and
failure in the Sinos valley, Brazil’. World Development 27(9):1627–50.
Schmitz, Hubert (2004) ‘Local Upgrading in Global Chains: Recent Findings’. Paper
presented to DRUID Summer Conference 2004 on Industrial Dynamics, Innovation
and Development. Elsinore, Denmark, 14–16 June. Available at
<http://www.druid.dk/uploads/tx_picturedb/ds2004-1422>
Schmitz, Hubert (2007) ‘Reducing complexity in the industrial policy
debate’. Development Policy Review 25(4):417–28.
Schmitz, Hubert, and Peter Knorringa (2000) ‘Learning from global buyers.’ Journal
of Development Studies 37(2):177–205.
Schwartz, Herman (2010) States versus markets: The emergence of a global
economy. New York: Palgrave.
Seguino, Stephanie (2000) ‘The effects of structural change and economic
liberalisation on gender wage differentials in South Korea and Taiwan’. Cambridge
Journal of Economics 24(4):437–59.
Selim, Tarek, and Ahmed Salem (2010) Global Cement Industry: Competitive and
Institutional Dimensions. MPRA_paper_24464, Munich, Munich University Library.
Available at <http://mpra.ub.uni-muenchen.de/24464/2/MPRA_paper_24464.pdf>
Selwyn, Ben (2012) Workers, state and development in Brazil: Powers of labour,
chains of value. Manchester: Manchester University Press.
Sender, John (1999) ‘Africa’s Economic Performance: Limitations of the current
consensus’. Journal of Economic Perspectives 13(3):89–114.
Sender, John (2003) ‘Rural Poverty and Gender: Analytical Frameworks and Policy
Proposals’. In Rethinking development economics, edited by Ha-Joon Chang.
London: Anthem Press.
Sender, John, Christopher Cramer, and Carlos Oya (2005) Unequal prospects:
disparities in the quantity and quality of labour supply in sub-Saharan Africa. World
Bank, Washington DC, World Bank Social Protection Discussion Paper Series 0525.
<http://siteresources.worldbank.org/SOCIALPROTECTION/Resources/0525.pdf>
(p.314) Sender, John, Carlos Oya, and Christopher Cramer (2006) ‘Women working
for wages: Putting flesh on the bones of a rural labour market survey in
Mozambique’. Journal of Southern African Studies 32(2):313–33.
Sender, John, and Sheila Smith (1986) The Development of Capitalism in Africa.
London: Methuen.
Shin, Jang-Sup (1996) The Economics of the Latecomers: Catching-up, technology
transfer and institutions in Germany, Japan and South Korea. London: Routledge.
Shirley, Mary (1997) ‘The economics and politics of government of
ownership’. Journal of International Development 9(6):849–64.
Simon, Bereket (2011) The tale of two elections (in Amharic). Addis Ababa: Mega
Printing Enterprise.
Singh, Ajit (2011) Comparative advantage, industrial policy and the World Bank:
Back to first principles. March. Available at <http://www.cbr.cam.ac.uk/pdf/WP418.pdf>
Söderbom, Mans (2011) ‘Firm size and structural change: A case study of
Ethiopia’. Journal of African Economies 21, Supplement 2, AERC Plenary Session,
May. Oxford Journals, Oxford: Oxford University. Available at
<http://jae.oxfordjournals.org/content/21/suppl_2/ii126.full.pdf+html>
Soludo, Charles, Osita Ogbu, and Ha-Joon Chang (eds.) (2004) The politics of trade
and industrial policy in Africa: Forced Consensus? Trenton NJ: Africa World Press.
Sonobe, Tetushi, and Keijiro Otsuka (2006) Cluster-based industrial development:
An East Asian Model. New York: Palgrave Macmillan.
Stein, Howard (2006) ‘Rethinking African Development’. In Rethinking development
economics, edited by Ha-Joon Chang. London: Anthem Press.
Stein, Howard (2012) ‘Africa, industrial policy, and export processing zones: Lessons
from Asia’. In Good growth and governance in Africa: Rethinking development
strategies, edited by Akbar Noman, Kwesi Botchwey, Howard Stein, and Joseph E.
Stiglitz. Oxford: Oxford University Press.
Stewart, Frances (2002) Horizontal Inequalities: A Neglected Dimension of
Development. QEH Working Paper Series. Working Paper No. 8. Available at
<http://www3.qeh.ox.ac.uk/pdf/qehwp/qehwps81.pdf>
Stewart, Frances (2008) Horizontal inequalities and conflict: Understanding group
violence in multiethnic societies. New York: Palgrave MacMillan.
Stiglitz, Joseph (1998) ‘More Instruments and Broader Goals: Moving Towards the
Post-Washington Consensus’. WIDER Annual Lecture, Helsinki, 7 January 1998.
Stiglitz, Joseph E. (2002) Globalization and its discontents. London: Penguin.
Stiglitz, Joseph E. (2011) The Mauritius Miracle. Project Syndicate. Available at
<http://www.project-syndicate.org/commentary/the-mauritius-miracle>
Studwell, Joe (2013) How Asia Works: Success and Failure in the World’s Most
Dynamic Region. London: Profile Books.
Sutton, John, and Nebil Kellow (2010) An Enterprise Map of Ethiopia. International
Growth Centre.
Szirmai, Adam, Wim Naudé, and Ludovico Alcorta (2013) Pathways to
Industrialization in the 21st century: New Challenges and Emerging Paradigms.
Oxford: Oxford University Press.
(p.315) Taleb, Nassim N. (2012) Antifragile: Things that gain from disorder. London:
Penguin.
Tareke, Gebru (1990) ‘Continuity and discontinuity in peasant mobilization: The
cases of Bale and Tigray’. In The Political Economy of Ethiopia, edited by Marina
Ottaway. New York: Praeger.
Tareke, Gebru (1991) Ethiopia: Power and Protest: Peasant Revolts in the Twentieth
Century. Cambridge: Cambridge University Press.
Tekeste, Abarham (2014) Trade Policy and Performance of Manufacturing Firms in
Ethiopia. Saarbrücken: Scholars’ Press.
The Economist (2008) ‘Kenya’s flower industry’. 7 February. Available at
<http://www.economist.com/node/10657231>
The Economist (2009) ‘Dutch flower auctions’. 8 April.
The Economist (2013) ‘The entrepreneurial state: A new book points out the big role
governments play in creating innovative businesses’. 31 August. Available at
<http://www.economist.com/news/business/21584307-new-book-points-out-big-role-governments-
play-creating-innovative-businesses>
The Economist Intelligence Unit (2012) Ethiopia: At an economic crossroads. 21
August. Available at <http://country.eiu.com/article.aspx?
articleid=659462850&Country=Ethiopia&topic=Economy>
The Guardian (2011) ‘Ethiopia's partnership with China: China sees Ethiopia as a
land of business opportunities, but the African country remains in charge of any
deals’. 30 December. Available at <www.theguardian.com/global-development/poverty-
matters/2011/dec/30/china-ethiopia-business-opportunities>
Thirlwall, Anthony (1980) Balance-of-Payments Theory and the United Kingdom
Experience. London: Macmillan.
Thirlwall, Anthony (2002) The nature of economic growth: An alternative framework
for understanding the performance of nations. Cheltenham: Edward Elgar.
Thirlwall, Anthony (2011) Economics of Development, 9th edn. London: Palgrave
Macmillan.
Thirlwall, Anthony, and Stephen Bazen (1989) Deindustrialization. Oxford:
Heinemann.
Tregenna, Fiona (2008a) ‘The contributions of manufacturing and services to
employment creation and growth in South Africa’. South African Journal of
Economics 76(S2):s175–s204.
Tregenna, Fiona (2008b) ‘Sectoral engine of growth in South Africa: An analysis of
services and manufacturing’. UNU-WIDER Research Paper No. 98. November.
Tregenna, Fiona (2012) ‘Sources of subsectoral growth in South Africa’. Oxford
Development Studies 40(2):162–89.
Tregenna, Fiona (2013) ‘Deindustrialization and Reindustrialization’. In Pathways to
Industrialization in the Twenty-First Century: New Challenges and Emerging
Paradigms, edited by Adam Szirmai, Wim Naudé, and Ludovico Alcorta. Oxford:
Oxford University Press.
UN COMTRADE (2012) United Nations Commodity Trade Statistics, Statistics
Division.
UNCTAD (2000) Tax Incentives and Foreign Direct Investment: A global survey.
ASIT Advisory Studies No. 16. Geneva: United Nations Conference on Trade and
Development.
(p.316) UNCTAD (2006) LDC Report. United Nations Conference on Trade and
Development. Available at <www.unctad.org>
UNCTAD (2008) Trade and Development Report, 2008. Domestic Sources of
Finance and Investment in Productive Capacity, Chapter IV. Available at
<http://unctad.org/en/docs/tdr2008ch4_en.pdf>
UNCTAD (2011) UNCTAD Least Developed Countries Report 2011. United Nations
Conference on Trade and Development.
UNCTAD (2014) Global governance and policy space for development, Trade and
Development Report, 2014. New York and Geneva.
UNCTAD-UNIDO (2011) Economic Development in Africa Report 2011: Fostering
Industrial Development in Africa in the New Global Environment. United Nations
Industrial Development Organization, New York and United Nations Conference on
Trade and Development, Geneva.
UN-DESA (2009) World Population Policies 2009. United Nations Department of
Economic and Social Affairs, New York.
UN-DESA (2013) World Population Prospects: The 2012 Revision, New York.
UNECA and AU (2011) Economic Report on Africa 2011. Governing Development in
Africa: The Role of the State in Economic Transformation. Economic Commission
for Africa and Africa Union. Available at
<http://www.uneca.org/sites/default/files/publications/era2011_eng-fin.pdf>
UNECA (2012) Report on livestock value chains in Eastern and Southern Africa: A
regional perspective. United Nations Economic Commission for Africa, Addis Ababa.
November.
UNECA and AU (2013) Making the Most of Africa’s Commodities: Industrializing
for Growth, Jobs and Economic Transformation. Economic Report on Africa 2013.
Available at <http://www.uneca.org/publications/economic-report-africa-2013>
UNECA and AU (2014) Dynamic Industrial Policy in Africa: Innovative Institutions,
Effective Processes and Flexible Mechanisms. Economic Report on Africa 2014.
<http://www.uneca.org/publications/economic-report-africa-2014>
UN-HABITAT (2010) The Ethiopia Case of Condominium Housing: The Integrated
Housing Development Programme. United Nations Human Settlements Programme,
Nairobi.
UNIDO (2003) A blueprint for the African leather industry: A development,
investment and trade guide for the leather industry in Africa. United Nations
Industrial Development Organization, Vienna.
UNIDO (2005) A Strategic Plan for the Development of Ethiopian Leather and
Leather Products Industry. Volume I and II. United Nations Industrial Development
Organization, Vienna.
UNIDO (2012) Ethiopia: Technical assistance project for the upgrading of the
Ethiopian leather and leather products industry. United Nations Industrial
Development Organization, Vienna.
USAID (2008) Success Story: Ethiopians Learning to Fight Ectoparasites. United
States Agency for International Development, Washington, DC.
(p.317) USGS (2010a) 2010 Minerals Yearbook, Ethiopia (Advance Release). US
Geological Survey and US Department of the Interior. June.
USGS (2010b) Hydraulic Cement African Production, by Country. US Geological
Survey.
USGS (2011a) Mineral Commodity Summaries: Cement. US Geological Survey.
January.
USGS (2011b) USGS Mineral Program Cement Report. Available at
<http://minerals.usgs.gov/minerals/pubs/commodity/cement/mcs-2011-cemen.pdf>
USGS (2013) Mineral Commodity Summaries: Cement. US Geological Survey.
January. Available at <www.minerals.usgs.gov/minerals/pubs/mcs/2013/mcs2013.pdf>
Vaughan, Sarah, and Mesfin Gebremichael (2011) Rethinking business and politics in
Ethiopia: The role of EFFORT, The Endowment Fund for the Rehabilitation of
Tigray. APPP Research Report 2.
Vergne, Jean-Philippe, and Rodolphe Durand (2010) ‘The Missing Link between the
Theory and Empirics of Path Dependence: Conceptual Clarification, Testability Issue,
and Methodological Implications’. Journal of Management Studies 47(4):736–59.
Wade, Robert ([1990] 2004) Governing the Market: Economic Theory and the Role
of the Government in East Asian Industrialization. Princeton: Princeton University
Press.
Warren-Rodriguez, Alex (2008) ‘Uncovering dynamics in the accumulation of
technological capabilities and skills in the Mozambican manufacturing sector’.
Available at <http://www.soas.ac.uk/economics/research/workingpapers/file43193.pdf/>
Warren-Rodriguez, Alex (2010) ‘Industrialisation, state intervention and the demise
of manufacturing development in Mozambique’. In African Political Economy, edited
by Vishu Padayachee. New York: Routledge.
Warwick, Ken (2013) Beyond Industrial Policy: Emerging Issues and New Trends.
OECD Science, Technology and Industry Policy Papers No. 2.
Washington Council for International Trade (2013) <http://wcit.org>
Watanabe, Matsuo, and Atsushi Hanatani (2012) ‘Issues in Africa’s Industrial Policy
Process’. In Good Growth and Governance in Africa: Rethinking Development
Strategies, edited by Adam Szirmai, Wim Naudé, and Ludovico Alcorta. Oxford:
Oxford University Press.
Weber, Max (1947) The Theory of Social and Economic Organization. New York:
Free Press.
Weiss, John (2013) ‘Industrial Policy in the Twenty-First Century: Challenges for the
Future’. In Pathways to Industrialization in the Twenty-First Century: New
Challenges and Emerging Paradigms, edited by Adam Szirmai, Wim Naudé, and
Ludovico Alcorta. Oxford: Oxford University Press.
Whitfield (2011) Growth without Economic Transformation: Economic Impacts of
Ghana’s Political Settlement. DIIS Working Paper No. 28. Available at
<http://subweb.diis.dk/graphics/Publications/WP2011/WP2011-28-Growth-without-Economic-
Transformation-Ghana_web.pdf>
Williamson, John (2008) ‘A short history of the Washington Consensus’. In The
Washington Consensus Reconsidered: Towards a New Global Governance, edited by
Narcis Sera and Joseph E. Stiglitz. Oxford: Oxford University Press.
(p.318) Woo-Cumings, Meredith (ed.) (1999) The Developmental State. Ithaca:
Cornell University Press.
Woo-Cumings, Meredith (1999) ‘Introduction: Chalmers Johnson and the Politics of
Nationalism and Development’. The Developmental State, edited by Meredith Woo-
Cumings. Ithaca: Cornell University Press.
World Bank (1995) Bureaucrats in Business: The Economics and Politics of
Government Ownership. Policy Research Report. World Bank, Washington DC.
World Bank (2008) Reshaping Economic Geography, 2009 World Development
Report, Washington DC.
World Bank (2009a) Cement sector in Africa and CDM: Investing in clean
technologies and energy savings. Key sheet. May.
World Bank (2009b) Cement Sector Program in Sub-Saharan Africa: Barriers
analysis to CDM and solutions. World Bank/CF Assist. April.
World Bank (2009c) Ethiopia: The Employment Creation Effects of the Addis Ababa
Integrated Housing Program. Report No. 47648-ET. March 2009.
World Bank (2010) Africa’s Infrastructure: A time for transformation. African
Development Forum.
World Bank (2012a) ‘Chinese FDI in Ethiopia: A World Bank Survey’. Draft,
unpublished. World Bank, Poverty Reduction and Economic Management, Africa
Region, AFTP2, October, Addis Ababa.
World Bank (2012b) Diagnosing corruption in Ethiopia: Perceptions, Realities, and
the way forward for key sectors, edited by Jannelle Plummer. Washington DC: World
Bank.
World Bank (2013) Life expectancy at birth, total (years). Available at
<http://data.worldbank.org/indicator/SP.DYN.LE00.IN>
World Bank and International Finance Corporation (2011) Doing Business 2011:
Making a difference for Entrepreneurs. Washington DC: World Bank.
World Bank and International Finance Corporation (ed.) (2013) Doing Business
2013: Smarter Regulation for Small and Medium Enterprises. Washington DC: World
Bank.
Wrong, Michela (2005) I didn’t do it for you: How the world used and abused a small
African nation. London: Harper Perennial.
Wu, John C. (2004) The mineral industry of the Republic of Korea. Reston, VA:
USGS. Available at <http://minerals.usgs.gov/minerals/pubs/country/2004/ksmyb04.pdf>
Young, Allyn (1928) ‘Increasing returns and economic progress’. Economic
Journal 38(152):527–42.
Young, John ([1997] 2006) Peasant Revolution in Ethiopia: The Tigray People’s
Liberation Front 1975–1991. Cambridge: Cambridge University Press.
Zenawi, Meles (2012) ‘States and Markets: Neoliberal Limitations and the Case for a
Developmental State’. In Good growth and governance in Africa: Rethinking
development strategies, edited by Akbar Noman, Kwesi Botchwey, Howard Stein,
and Joseph E. Stiglitz. Oxford: Oxford University Press.
Zewde, Bahru (ed.) (2002a) The History of Modern Ethiopia 1955–1991. Athens:
Ohio University Press.
Zewde, Bahru (2002b) Pioneers of Change in Ethiopia: The Reformist Intellectuals
of the Early Twentieth Century. Oxford: James Currey.

(p.319) Selected local references

AACG (2003) Housing stock and demand in Addis Ababa. Unpublished. Housing
Development Agency.
CBE (2011) Special truck loan and pre-shipment export credit facility guideline.
Letter of Commercial Bank of Ethiopia’s president to all branches, district, and head
office organs. P/BZ/08/11. Addis Ababa, Ethiopia.
CBE (2012) Summary of performance progress on grand initiatives, priority sector
financing, and FCY inflow. Commercial Bank of Ethiopia. Unpublished. Addis
Ababa, Ethiopia.
Chemical Industry Development Directorate, MOI (2012) Letters and minutes (Nos.
1–80) compiled by Ministry of Industry from 2009–12. Addis Ababa, Ethiopia.
CSA (1988) Results of the survey of manufacturing and electricity
industries (1984/85). Central Statistical Authority, February 1988, Addis Ababa,
statistical bulletin 58.
CSA (1990) Results of the survey of manufacturing and electricity
industries (1986/87). Central Statistical Authority, January 1990, Addis Ababa,
statistical bulletin 75.
CSA (1991) Results of the survey of manufacturing and electricity
industries (1987/88). Central Statistical Authority, January 1991, Addis Ababa,
statistical bulletin 87.
CSA (1992) Results of the survey of manufacturing and electricity
industries (1988/89). Central Statistical Authority, April 1992, Addis Ababa,
statistical bulletin 97.
CSA (1993) Results of the survey of manufacturing and electricity
industries (1989/90). Central Statistical Authority, April 1993, Addis Ababa,
statistical bulletin 112.
CSA (1994) Results of the survey of manufacturing and electricity
industries (1992/93). Central Statistical Authority, August 1994, Addis Ababa,
statistical bulletin 126.
CSA (1995) Results of the survey of manufacturing and electricity
industries (1993/94). Central Statistical Authority, August 1995, Addis Ababa,
statistical bulletin 136.
CSA (1996) Results of the survey of manufacturing and electricity
industries (1994/95).Central Statistical Authority, September 1996, Addis Ababa,
statistical bulletin 158.
CSA (1997) Report on large scale manufacturing and electricity industries survey.
Central Statistical Authority, October 1997, Addis Ababa, statistical bulletin 178.
CSA (1998) Report on large scale manufacturing and electricity industries survey.
Central Statistical Authority, October 1998, Addis Ababa, statistical bulletin 191.
CSA (1999) Report on large scale manufacturing and electricity industries survey.
Central Statistical Authority, September 1999, Addis Ababa, statistical bulletin 210.
CSA (2000) Report on large scale manufacturing and electricity industries survey.
Central Statistical Authority, October 2000, Addis Ababa, statistical bulletin 231.
CSA (2001) Report on large scale manufacturing and electricity industries survey.
Central Statistical Agency, February 2001, Addis Ababa, statistical bulletin 248.
CSA (2001, 2002, 2007, 2008, 2010, and 2011b) Statistical Abstracts of 2001, 2002,
2007, 2008, 2010, and 2011. Central Statistical Agency, Addis Ababa, Ethiopia.
CSA (2002) Report on large scale manufacturing and electricity industries survey.
Central Statistical Agency, October 2002, Addis Ababa, statistical bulletin 265.
CSA (2003) Report on large scale manufacturing and electricity industries survey.
Central Statistical Agency, October 2003, Addis Ababa, statistical bulletin 281.
(p.320) CSA (2004) Report on large scale manufacturing and electricity industries
survey. Central Statistical Agency, November 2004, Addis Ababa, statistical bulletin
321.
CSA (2005) Report on large scale manufacturing and electricity industries survey.
Central Statistical Authority.
CSA (2006) Report on large scale manufacturing and electricity industries survey.
Central Statistical Agency, November 2006, Addis Ababa, statistical bulletin 380.
CSA (2007) Report on large scale manufacturing and electricity industries survey.
Central Statistical Agency, October 2007, Addis Ababa, statistical bulletin 403.
CSA (2008) Report on large scale manufacturing and electricity industries survey.
Central Statistical Agency, October 2008, Addis Ababa, statistical bulletin 431.
CSA (2009) Report on large scale manufacturing and electricity industries survey.
Central Statistical Agency, December 2009, Addis Ababa, statistical bulletin 472.
CSA (2010) Report on large scale manufacturing and electricity industries
survey. Central Statistical Agency, November 2010, Addis Ababa, statistical bulletin
500.
CSA (2011) Report on large scale manufacturing and electricity industries survey.
Central Statistical Agency, August 2011, Addis Ababa, statistical bulletin 531.
CSA (2012a) Agricultural sample surveys (from 2005 to 2012), Vol. II, Statistical
Bulletins 331, 361, 388, 417, 446, 468, 505, and 532. Central Statistical Agency,
Addis Ababa, Ethiopia.
CSA (2012b) List of firms involved in the 2012 survey of manufacturing sector,
leather and leather products sector. Central Statistical Agency, Addis Ababa, Ethiopia.
CSA (2012c) Report on large scale manufacturing and electricity industries survey.
Central Statistical Agency, August 2012, Addis Ababa, statistical bulletin 532.
CSA (2013) Population size of towns by sex, region, zone, and wereda. July. Central
Statistical Agency, Addis Ababa, Ethiopia.
CSO (1977) Results of the survey of manufacturing industries in Ethiopia (1976).
Statistical Bulletin 17. June. Provisional Military Government of Socialist Ethiopia,
Central Statistical Office, Addis Ababa, Ethiopia.
CSO (1983) Ethiopia Statistical Abstracts of 1968, 1980, and 1982. Central Statistical
Office, Addis Ababa, Ethiopia.
CSO (1987) Results of the surveys of manufacturing industries (1976/77 to 1983/84).
Statistical Bulletins 23, 29, 30, 38, 48, and 53. Central Statistical Office, Addis
Ababa, Ethiopia.
CSO and MCIT (1969) Survey of manufacturing and electricity industry (1964/5–
1966/7). Statistical Bulletin 2. June. Central Statistical Office and Ministry of
Commerce, Industry and Tourism, Addis Ababa, Ethiopia.
DBE (2009) The priority area projects of the DBE. As approved by the Board of
Directors of the Development Bank of Ethiopia, Addis Ababa, Ethiopia.
DBE (2012a) A Guide to Development Bank of Ethiopia. Development Bank of
Ethiopia, Addis Ababa, Ethiopia.
DBE (2012b) Loan portfolio concentration report as of December 2011. Unpublished.
Development Bank of Ethiopia, Addis Ababa, Ethiopia.
DBE (2012c) Monthly Report on Status of Flowers Projects. Assistant President,
PO/241/12, Development Bank of Ethiopia, Addis Ababa, Ethiopia.
(p.321) DBE (2012d) Third Quarter Performance Report 2011/12. Unpublished.
Development Bank of Ethiopia, Addis Ababa, Ethiopia.
DLV (2012) Quantitative unified information. DLV consultancy to Ethiopian
horticulture sector <http://www.dlvplant.nl/uk/content/projects-africa.html>. Addis Ababa,
Ethiopia.
EAL (2012) Freight data. Unpublished. Ethiopian Airlines. Addis Ababa, Ethiopia.
EEPCO (2012) Detailed Information of EEPCO. Supplementary data for interview
with CEO of Ethiopian Electric Power Corporation, Addis Ababa, Ethiopia.
EFFORT (1995) Memorandum of Foundation. Endowment Fund for the
Rehabilitation of Tigray.
EFFORT (2010) EFFORT Corporate Strategy. Endowment Fund for the
Rehabilitation of Tigray.
EFFORT (2011) EFFORT Performance Report. Endowment Fund for the
Rehabilitation of Tigray.
EHDA (2011a) Export performance report of 2011. Amharic version. Ethiopian
Horticulture Development Agency, Addis Ababa, Ethiopia.
EHDA (2011b) Guideline on the value chain of floriculture export. Ethiopian
Horticulture Development Agency, Addis Ababa, Ethiopia.
EHDA (2011c) The 2011 horticulture sector report. Ethiopian Horticulture
Development Agency, Addis Ababa, Ethiopia.
EHDA (2012a) EHDA Statistical Bulletin 01, October. Ethiopian Horticulture
Development Agency, Addis Ababa, Ethiopia.
EHDA (2012b) Report to National Export Coordinating Committee. Unpublished.
Ethiopian Horticulture Development Agency, Addis Ababa, Ethiopia.
EHDA (2012c) New Approach to Market Ethiopian Flowers in the Russian
Federation. Ethiopian Horticulture Development Agency, Addis Ababa, Ethiopia.
EHDA (2012d) Minutes of meetings (Numbers 1–25 of EHDA from 2009–12).
Unpublished. Archives of Ethiopian Horticulture Development Agency. Addis Ababa,
Ethiopia.
EHDA (2012e) Correspondence from firms and other stakeholders (Numbers 1–35),
2009–12. Archives of Ethiopian Horticulture Development Agency, Addis Ababa,
Ethiopia.
EHDA (2013) Labour records of firms and summary review. May. Addis Ababa,
Ethiopia.
EHPEA (2007) A Guide for new investors. Ethiopian Horticulture Producers and
Exporters Association. Available at <http://www.EHPEA.org.et>
ELIA (2012) Letter submitted to the Prime Minister at the government-private sector
dialogue on 24 October 2012. Unpublished.
EPRDF (2011a) Five Years Growth and Transformation Plan and Ethiopian
Renaissance. Addis Ababa: Ethiopian People’s Revolutionary Democratic Front and
Mega Printing.
EPRDF (2011b) The EPRDF from founding to the present. Amharic version. Addis
Ababa: Ethiopian People’s Revolutionary Democratic Front and Mega Printing.
EPRDF (2011c) Ethiopia’s Renaissance and EPRDF Lines. In Amharic. Addis
Ababa: Mega Printing.
(p.322) EPRDF (2013a) EPRDF reaffirmed its commitment to the renaissance of
Ethiopia and the construction of the Renaissance Dam. Addis Ababa: Ethiopian
People’s Revolutionary Democratic Front.
EPRDF (2013b) Meles’s Portrait: Sketch of who he is. Ethiopian People’s
Revolutionary Democratic Front, Addis Ababa. Available at <www.eprdf.org.et>
EPRDF (2013c) New Vision (bimonthly editions). Ethiopian People’s Revolutionary
Democratic Front, Addis Ababa <www.eprdf.org.et>
EPRDF (2013d) Report of 9th EPRDF Congress, Ethiopian People’s Revolutionary
Democratic Front. Bahirdar, Ethiopia.
EPRDF (2013e) EPRDF Renaissance Path and Ethiopia’s Future. In Amharic, March
2013. Office of EPRDF, Addis Ababa, Ethiopia.
ERCA (2012a) Export revenues and volume. Unpublished data. Ethiopian Revenues
and Customs Authority, Addis Ababa, Ethiopia.
ERCA (2012b) Data on the imports of textile and apparel imports 1998–2012.
Ethiopian Revenue and Customs Authority, Addis Ababa, Ethiopia.
FDRE (1992) Public Enterprises Proclamation No. 25/1992. House of Peoples’
Representatives, Addis Ababa, Ethiopia.
FDRE (1994) An Economic Development Strategy for Ethiopia. Federal Democratic
Republic of Ethiopia, Addis Ababa, Ethiopia.
FDRE (1995) Constitution of the Federal Democratic Republic of Ethiopia. House of
Representatives, Addis Ababa, Ethiopia.
FDRE (1996) Ethiopia’s Development Strategy (ADLI). Federal Democratic Republic
of Ethiopia, Addis Ababa, Ethiopia.
FDRE (1998a) Ethiopian Export Promotion Agency Establishment Proclamation.
Federal Democratic Republic of Ethiopia, Addis Ababa, Ethiopia.
FDRE (1998b) Export Promotion Strategies (Export Development Strategies),
Number 5/Decision 159/1998. Amharic version. Federal Democratic Republic of
Ethiopia, Addis Ababa, Ethiopia.
FDRE (2002) The Industrial Development Strategy of Ethiopia. Federal Democratic
Republic of Ethiopia Information Ministry, Addis Ababa, Ethiopia.
FDRE (2008a) Raw and Semi-processed Hides and Skins Export Tax. Proclamation
Number 567/2008. FDRE House of Representatives, Addis Ababa, Ethiopia.
FDRE (2008b) The Regulation for the establishment of Ethiopian Horticulture
Development Agency. Regulation No. 152/2008. FDRE Council of Ministers.
FDRE (2012) The Strategy of the Development of Micro and Small Enterprises.
Federal Democratic Republic of Ethiopia, Addis Ababa, Ethiopia.
FIA (2012a) Investment certificates provided to FDI and domestic investors.
Unpublished. Federal Investment Agency, Addis Ababa, Ethiopia.
FIA (2012b) Consultations with business community; and the historical evolution of
FIA and investment law (1992–2013). Unpublished. Federal Investment Agency,
Addis Ababa, Ethiopia.
FIA (2012c) Minutes of meetings of the board of FIA (2008–12). Archives of the
Federal Investment Agency, Addis Ababa, Ethiopia.
(p.323) GOE (1960) The Civil Code of Ethiopia Proclamation of 1960. Proclamation
Number 165/1960. Government of Ethiopia, Addis Ababa, Ethiopia.
LIDI (2011) Draft on the ban of crust and export of leather products. Unpublished.
Leather Industry Development Institute, Ministry of Industry, Addis Ababa, Ethiopia.
LIDI (2012a) Interim Audit Report as per Requirement of Leather and Leather
Products Sector Best Practice Standard for 18 tanneries and footwear firms. Leather
Industry Development Institute, Ministry of Industry, Addis Ababa, Ethiopia.
LIDI (2012b) Minutes of meetings of Tanning and Footwear Industries
Benchmarking, Number 1–7. Leather Industry Development Institute, Ministry of
Industry, Addis Ababa, Ethiopia.
LIDI (2012c) Minutes of meetings of the twinning programme, Number 1–6. Leather
Industry Development Institute, Ministry of Industry, Addis Ababa, Ethiopia.
LIDI (2012d) Draft profile of the leather sector, organized for research interview by
LIDI. Director general’s notes. Leather Industry Development Institute, Ministry of
Industry, Addis Ababa, Ethiopia.
LIDI (2012e) Letters from firms (Number 1–330) in the leather and leather products
sector, 2010–12. Leather Industry Development Institute, Ministry of Industry, Addis
Ababa, Ethiopia.
LIDI (2012f) Facilitation Supports Requested by Firms in Leather Sector. Leather
Industry Development Institute, Ministry of Industry, Addis Ababa, Ethiopia.
LIDI (2012g) Performance data. Unpublished. Planning and MIS Directorate, Leather
Industry Development Institute, Ministry of Industry, Addis Ababa, Ethiopia.
MOA (2012) Review of animal resources. Unpublished. Ministry of Agriculture,
Addis Ababa, Ethiopia.
MOFED (1999) Survey of the Ethiopian Economy: Review of Post-Reform
Developments (1992/3–1997/98). Ministry of Finance and Economic Development,
Addis Ababa, Ethiopia.
MOFED (2002) Sustainable Development and Poverty Reduction Program (SDPRP).
Ministry of Finance and Economic Development, Addis Ababa, Ethiopia.
MOFED (2006) Ethiopia: Building on Progress: A Plan for Accelerated and Sustained
Development to End Poverty (PASDEP) 2005/06–2009/10. Volume I. Ministry of
Finance and Economic Development, Addis Ababa, Ethiopia.
MOFED (2010) Growth and Transformation Plan 2010/11–2014/15. Volume I, Main
Text, and Volume II, Policy Matrix. Ministry of Finance and Economic Development,
Addis Ababa, Ethiopia.
MOFED (2011a) A Study to improve tariff incentives. Ministry of Finance and
Economic Development, Ministry of Industry, and Ethiopian Revenue and Customs
Authority, Addis Ababa, Ethiopia.
MOFED (2011b) Raw and Semi-processed Hides and Skins Export Tax Amendment
Directives. Directive No. 30/2011. Ministry of Finance and Economic Development,
Addis Ababa, Ethiopia.
MOFED (2012a) Macroeconomic Developments in Ethiopia. Annual Report 2010/11.
Ministry of Finance and Economic Development, Addis Ababa, Ethiopia.
(p.324) MOFED (2012b) Ethiopia’s macroeconomic data. Unpublished. National
Economic Accounts Directorate, Ministry of Finance and Economic Development,
Addis Ababa, Ethiopia.
MOFED (2012c) Ethiopian Tariff Amendments (1993–2011). Unpublished. Ministry
of Finance and Economic Development, Addis Ababa, Ethiopia.
MOFED (2013a) Annual GTP performance report. Unpublished. Ministry of Finance
and Economic Development, Addis Ababa, Ethiopia.
MOFED (2013b) A Directive to implement the export tariff incentive system.
Amharic version. Ministry of Finance and Economic Development, Addis Ababa,
Ethiopia.
MOFED (2014) Brief Note on the 2013 National Accounts Statistics. Unpublished.
25 October. Ministry of Finance and Economic Development. Addis Ababa, Ethiopia.
MOI (2012a) Summary of consultations between government and industrialists on 25
October 2012 at Sheraton-Addis. Unpublished. Ministry of Industry, Addis Ababa,
Ethiopia.
MOI (2012b) Global cement consumption. Unpublished. Chemical Industry
Development Directorate, Ministry of Industry, Addis Ababa, Ethiopia.
MOI (2012c) Industrial zone development strategy study. Amharic version.
Unpublished. Ministry of Industry, Addis Ababa, Ethiopia.
MOI (2013) Directives to implement Proclamation 768/2012 on export tariff
promotion. Amharic version. Ministry of Industry, Addis Ababa, Ethiopia.
MOM (2012) Report on mining and exploration licences. Ministry of Mines, Addis
Ababa, Ethiopia.
MoTI (2008) Directive on Petroleum Cost Absorption for Air Freight of Floriculture.
14 April 2008.
MUDC (2013) Data on housing development, cement demand and consumption. July.
Unpublished. Ministry of Urban Development and Construction, Addis Ababa,
Ethiopia.
Mugher (2013) Updated data on firm status. July.
MWUD (2007) Plan for urban development and urban good governance, PASDEP
2005/06–2009/10. Addis Ababa: Ministry of Works and Urban Development and
Berhanena Selam Printing.
MWUD (2009) Reports on capacity building of construction industry. Ministry of
Works and Urban Development, Housing Development Office, Addis Ababa,
Ethiopia.
MWUD (2010) Logistics Review and Plan of Housing Development Programme.
Unpublished. Ministry of Works and Urban Development, Housing Development
Office, Addis Ababa, Ethiopia.
MWUD (2013) Housing Development Programme Updated Data. Unpublished.
Ministry of Works and Urban Development, Housing Development Office, Addis
Ababa, Ethiopia.
Nazret (2010) ‘Ethiopia-Derba Cement to buy 1000 Volvo trucks’. Available at
<http://nazret.com/blog/index.php/2010/01/15/ethiopia_s_derba_midroc_cement_to_buy_10>
(p.325) NBE (2011) Annual Report 2009/10. Addis Ababa: Domestic Economic
Analysis and Publications Directorate, National Bank of Ethiopia.
NBE (2012a) NBE Letter to all banks, List of Flower Exporters with Outstanding
Commitment From May 12, 2008 to April 30, 2012. FEMRMD/745/12. National
Bank of Ethiopia, Addis Ababa.
NBE (2012b) Unpublished data. National Bank of Ethiopia Research Department,
Addis Ababa.
NBE (2013) Regulations and directives of National Bank of Ethiopia. Addis Ababa.
Also available at <www.nbe.gov.et>
NECC (2010) Study on Export Sector Development, Leadership, and Coordination.
Unpublished draft. National Export Coordinating Committee, Prime Minister’s
Office, Addis Ababa, Ethiopia.
NECC (2012) Minutes of meetings (Number 1–92) of the National Export
Coordinating Council, 2004–12. Addis Ababa, Ethiopia.
Oqubay, Arkebe (2012) Quasi-census (Number 1–150), site observations (Number 1–
50), and interviews (Number 1–200) with selected firms in cement, floriculture,
leather and leather products sectors, and related stakeholders. Unpublished. Addis
Ababa, Ethiopia.
PPESA (2012) Privatization performance and reports. Unpublished.
Royal Netherlands Embassy (2012) Surveys of on Dutch Flower Firms in Ethiopia
2010–12. Unpublished, Addis Ababa, Ethiopia.