...-!.

The Role of Foreign Assistance in the Economic Growth of Developing Countries

by

C. Peter Timmer1

"Little else is requisite to carry a state to the highest degree of opulence from the lowest
barbarism than peace, easy taxes, and tolerable administration ofjustice; all the rest
being brought about by the natural course ofthings." Lecture by Adam Smith in 1775

I. The Underlying Forces that Drive Economic Growth

Adam Smith was making several points about economic growth (i.e. "the highest degree
of opulence") when he spoke these prescient words in 1775 (cited by E.L. Jones, 1981, p.
235). The unit of observation is the state and it is also the decision maker in Smith's
world, because it provides what are now considered key elements of good economic
governance-"peace, easy taxes, and the tolerable administration of justice." Democracy
is not mentioned in Smith's list; as the political mechanism to ensure good economic
governance, that comes much later. Smith's final point is also among the most
controversial: economic growth will be "the natural course of things" because of how
people behave and the pressures of competition from the "invisible hand."

The human behavior that led Smith to make these observations seems to be wired very
deeply in our brains (Jared Diamond, 1997; E. L. Jones, 1988). From this behavioral
(and historical) perspective, development is seen as a long-run sequence of decisions by
economic agents, acting in their own self interest, that culminate in rising investment
levels and higher labor productivity. The "miracle" of advanced living standards is the
set of institutions that permit, even encourage, these decisions for very long periods of
time-a century or longer (Jones, 1981 ). At a growth rate of just two percent per capita
per year, a $1,000 income becomes $8,000 in that century, and $64,000 after two
centuries. That is how the United States and Europe became rich-two major wars and,
in Europe, increasingly heavy tax regimes, notwithstanding.

The difficult tasktoday is for less-developed countries to sustain the "Smithian

conditions" over a long enough period of time to realize the promised opulence. To say
that "good governance" is the bridge that connects an investment-promoting tax regime
and a system of justice that protects property rights (as well as human rights) within a

1
This is the third draft of a chapter for the "Natsios Report" on Foreign Assistance and the U.S. National
Interest, commissioned by the United States Agency for International Development (USAID). The ideas in
this chapter reflect extensive discussions with colleagues inside and outside ofUSAID, but the agency
takes no responsibilitY for the arguments advanced herein. The author is Professor of Development Studies,
Graduate School of International Relations and Pacific Studies, University ofC!llifornia, San Diego. E-
mail address: [email protected]
stable, peaceful (and democratic) country is to say the obvious, hut it is no less important
for that.

In view of the record in transition economies and much of Sub-Saharan Africa, it seems
that we know little of how to help countries build this bridge, at least, how to build it
quickly when few foundations are in place. Chapter 2 provides a number of helpful
guidelines on what we do know (L. Diamond, 2002). Still, the "mechanics" of economic
development, to use Robert Lucas' provocative term, are of virtually no use without
effective economic governance (Lucas, 1988). If political, ethnic, tribal, or religious
rivalries prevent a central government from implementing a coherent and stable
macroeconomic strategy, supported by the rule of law, there is little hope for a rapid
escape from poverty. Without effective economic governance, virtually nothing of what
follows in this chapter will matter.

With effective economic governance, however, much is on promise, because the only
long-run hope of the poor is to live in a growing economy. In some circumstances even
economic growth may not be enough to reduce poverty significantly and additional
efforts will be needed. Analyses carried out since Smith was writing have identified the
circumstances and we review them in what follows.

The primary focus of this chapter, then, is on starting or speeding up growth in the
economies of poor countries. There is some scope for making growth policies "pro-
poor ," but the real challenge in most poor countries is to get any sustainable growth at all.
At least for the next generation, U.S. strategy for poverty reduction in developing
countries must be focused on poor economies. Clearly, the next step is to figure out how
to get economic growth going in the many countries where the ingredients are missing.

There are two ways to approach this problem. The first is to look at current experience,
country by country, treating each as a special case, and assess what these countries are
doing right or wrong. The second is to look at the long-run record of economic growth to
determine the underlying causes of growth and identify what is missing in countries with
poor growth performance.

The first approach will keep myriad policy analysts, consultants and academic
researchers productively engaged for decades; the time- and site-specific findings will no
doubt be useful in their time and place. But such short-run analysis has not solved the
development problems in these countries in the past and there is no reason to think it will
solve them now. Somehow the problems run deeper than current political squabbles,
debt ratios, or changes in the terms of trade. The second approach is taken here.

2
A. The Evolution of Thinking on Economic Growth

Economists have conceptualized the process of economic growth around three basic
models: specialization and trade; investment in machines; and increasing returns to
knowledge. Although the models are overlapping to an extent and potentially
complementary, each model has been held out at some point in the history of economic
thought as the fastest road to riches. Some variants of the "machine model" even
promised a turnpike (Dorfman, Samuelson and Solow, 1958; Rostow, 1990).

Much of classical economics was devoted to understanding the process of economic

growth. As already noted, Adam Smith was highly optimistic about the prospects for
higher living standards, to be achieved primarily by the higher labor productivity that
resulted from specialization in, for example, the many tasks involved in pin making.
Specialization could only succeed through trade and the process was then limited by the
size of the market. By lowering trade barriers-artificially imposed by governments or
naturally caused by long distances or difficult terrain-larger markets became accessible
to manufacturers. Competition-the invisible hand-would force them to more and more
specialized forms of production, thus raising labor productivity and living standards. To
be successful, this trade-intensive strategy of economic growth required many .
transactions, increasingly at long distance, so the institutions defending property rights
and lowering transactions costs, such as rule of law, came to be seen as the foundation of
a market economy. That is, economic governance has long been seen as an essential
starting point for the economic growth process, not something that is tacked on in mid-
stream.

The visible success in the 19th century of the Industrial Revolution, first in Britain and
then in France and Germany, changed how economists thought about the growth process.
Technological change, created by a new scientific enterprise and embodied in machines,
became .the driving force of development (Landes, 1969, 1990, 1998). And if not all
countries could invent and produce their own machines, all were free to import them and
reproduce the factory system that was making Europe so rich and powerful. A "capital
fundamentalism" emerged that stressed the accumulation of savings to be used to invest
in machines that embodied the latest technologies, the origins of which were exogenous
to the day-to-day activities of factory managers or national economic planners.

The "machine model" was clearly open to countries in a hurry to catch up with their rich
neighbors or distant trading partners. Many institutional elements of Adam Smith's
"trade model" could be circumvented, or substitutes found, ifthe investment and
production process did not have to rely on the profit motive of private investors, but
instead relied directly on decisions of national planners. Early German industrialization
was a "deliberate act of policy" (Cole and Deane, 1965) and it changed the balance of .
both economic and military power in Europe by the end of the 19th century. The pace of
Soviet industrialization changed the balance of power again during the middle half of the
20th century.

3
The "machine model" worked reasonably well when two conditions held: first, when
relatively little trade was needed to permit labor specialization as the source of higher
productivity. The higher productivity was generated instead through intensification of
capital goods, with production primarily for domestic markets; and second, when
appropriate technology in the form of capital equipment was readily available
domestically or from the world market.

As the pace of scientific innovation accelerated in the advanced countries, however, and
productivity growth relied increasingly on knowledge directly rather than on technology
embodied in machines, the model stumbled in country after country. In those countries
with nascent institutions to support low-cost trade and absorption of western knowledge
rather than just machines, such as Korea, Taiwan, Israel and Brazil, the transition to
export-led growth was feasible if not always smooth. In countries without these
institutions, including nearly all of Africa and most of the Islamic world, the failure of
economic governance led to rapidly failing economies. After years, even decades of
steady economic growth, they have slipped back into economic decline and rising levels
of poverty. A number of these countries have slipped even further, into chaos and
conflict.

The new model of economic growth that explains this performance is based on increasing
returns to knowledge. That is, instead of diminishing marginal returns as more and more
identical machines are used for a given labor force, marginal returns to knowledge
actually increase with greater use because of spillovers, at no marginal cost, to additional
users. Large economic payoffs from new knowledge, especially in the early years of
application when patent rights provide market power, encourage economic entrepreneurs
to develop it, using the fundamental science produced in modern research universities
and corporate research centers.

Thus economic growth, instead of depending on exogenous technical change, is now seen
to be an endogenous process of response to incentives throughout the entire economic
system, not just at the level of firm investments or consumer decision making (Easterly,
2001 ). The modem concern for enforcing intellectual property rights as well as property
rights for land, goods and financial assets is easy to understand from the perspective of
this "knowledge model." A failure to defend intellectual property rights will slow the
search for useful new knowledge, and hence the rate of economic growth.

The difficulty of endogenous growth for the still poor countries is that knowledge
generation and development of sophisticated human capital depend at least as strongly on
the "foundation institutions" that ensure property rights and low transaction costs as the
Smithian "specialization and trade" model that gave rise to them. It was the long time
needed for each society to evolve its own such institutions and investments that gave rise
to the search for substitutions that could speed the economic growth process
(Gerschenkron), a search that largely failed in the third world. There is an uneasy sense
in the development profession that we are back to square one (Landes, 1990; Easterly,
2001). As Easterly reminds us, the "quest for growth" in this world of increasing returns
to knowledge is quite elusive.

4
B. The Empirical Record of Economic Growth

All models of economic growth emerge as attempts to explain the empirical record,
however stylized the facts are that describe it (Solow). Modem growth empirics trace
from the pioneering work of Clark and Kuznets (1955, 1966), through the increasingly
sophisticated frameworks of growth accounting developed by Denison, Abramovitz,
Baumol and Nelson, to the high-tech econometrics of Barro and his associates (1994,
1995, 1997), Mankiw and his colleagues, and a legion of users of the Summers-Heston
data set. Only the key results for our understanding of the long-run growth process are
highlighted here.

First, it is clear that empirics have driven theory, rather than the other way around, an
entirely appropriate approach for a discipline that tries to explain human decision making
in the face of scarce resources. The contrast with the field of international trade, heavily
dominated by theorists, is striking (Davis and Weinstein, 2001).

Second, given the diversity of experience with economic growth-over time and across
countries-most "sensible" variables can be shown to influence the growth process under
some set of circumstances, ifthe researcher looks hard enough. Capital, labor, education,
government investment spending, low inflation, macroeconomic stability, openness to
trade, the quality of institutions, and democracy all contribute positively to economic
growth in some set of countries or time period.

Third, with a specialist's insights and motivation, and hard work assembling the
necessary data sets, more particular or specific factors can be seen at work. Unstable
prices for commodity exports slow down economic growth (Dawe, 1993, 1996; Collier
and Denn, 2001). A favorable balance between rural and urban education, serving as a
proxy for reduced "urban bias," speeds it up (Lipton, 1977, 1993; Chai, 1995). Trade
openness turned out to be bad for economic growth in the 1930s and 1940s, of little
importance one way or the other in the 1950s and 1960s, but highly significant in
explaining rapid economic growth in the 1970s and 1980s (Vamvak:idis, 1997).

Finally, one factor emerges from both the long-run historical record of individual
countries and the shorter post-World War II cross-section record for developing countries
as crucial to sustained growth~ Rapid factor accumulation, driven by high domestic
savings rates or foreign capital inflows, can lift a poor country onto the first rung of the
development process. But eventually, improvements in total factor productivity, or th.e
efficiency with which all inputs are used, including capital, must become the main source
of higher incomes. This realization was the real driving force behind the development of
the "knowledge model" of development (Romer, 1986, 1990; Prescott, 1998).

The difficulties in making the transition from the "machine model" to the "knowledge
model" in a trade-driven global economy are clearly seen in the growth record of the
1990s. According to statistics in the World Bank's World Development Report, 2002,
more than a third of the 108 developing or transition economies had lower per capita
incomes in 2000 than in 1990, and none of these were countries directly affected by the

5
Asian financial crisis that started in 1997. The decade saw some of the fastest growth in
global output, and in volumes of international trade, on record, so the external
environment was favorable to growth. The sources of poor performance must be sought
within the countries themselves.

None of the 38 poor performers are in Asia. Nearly half (18 of38) of these countries
were part of the former Soviet Union, testimony to the failure of shock therapy to speed
the transition to a market economy in the absence of market institutions. Africa
accounted for another 14 countries, testimony to the continuing, widespread difficulty in
arresting the economic erosion that has been continent-wide for decades. And despite
reasonable economic performance for the region as a whole, four countries in Latin
America and the Caribbean suffered decade-long declines.

Such long-term economic problems point to deep-seated failures to establish the core
elements that support modern economic growth. The list of these elements is not long,
but they are basic: provision of public goods and social infrastructure, a stable
macroeconomic environment, and a conducive business climate. Why do governments
fail to provide these essentials for growth? Indeed, why do some governments actively
seek to undermine them?

Modem political economy has tried to answer these qu€stions. Rational choice models of
political actors explain why state agents collect private goods from public resources, but
they are less successful in explaining how to correct the vicious circle of corruption
within the dynamics of a society's own political system (Srinivasan, 1985; Alesina and
Rodrik, 1994; L. Diamond, 2002). Even democratic systems with regular elections have
not been immune from these dynamics, so it is not possible to recommend with
confidence any single approach as a short-run remedy.

C. Mapping a Way Forward: The Task Ahead

To be successful, this chapter must establish the general validity of three relationships:

1. The connection between economic growth and reduction in poverty. This

connection might be absolute, in the sense that any economic growth is good for
the poor and no further attention to poverty reduction, as opposed to economic
growth, is needed. Alternatively, the connection might be conditional, in the
sense that economic structure, initial conditions of asset and income distribution,
the sector in which growth takes place, or even government policies, could
influence the extent to which the poor participate in economic growth. The
debate over the "quality of growth" is about the nature of conditions, if any, that
connect the poor to the process of economic growth (World Bank, 2000);

2. The relationship between what governments do and the pace of economic

growth. It is obvious that government action, or inaction, can slow down the
growth process or stop it altogether. In the extreme, partisans in failed states can
fight over the immediate distribution of the spoils from war, immizerizing all in

6
 their path. The relationships sought here are more positive: the need (and
potential) for governments to provide public goods and a stable economic
environment that stimulate trade and investment. Good economic governance
will be essential for governments to realize this potential (L. Diamond, 2002); and

3. The relationship between foreign assistance, broadly construed, and poverty

reduction in developing countries. Multiple mechanisms will be at work within
this relationship, and they complicate the incentive systems that should link donor
behavior to appropriate responses by recipient governments and civil servants.
Assistance mechanisms include direct humanitarian assistance to feed the hungry,
educate or train the poor and illiterate, or cure the sick; remittances from abroad
that provide consumption goods or capital to establish family-based
microenterprises; technical assistance from donor countries to improve policy
design and program implementation; loans to build roads, hospitals, irrigation
systems and telecommunication networks; and balance of payments support to
ward off financial crises. Clearly, designing an incentive system to call forth
efficient delivery of food aid will be very different from one that maximizes the
productivity of private remittances. Repeatedly bailing out countries from failed
macroeconomic policies, or forgiving their international debt, creates perverse
incentives-moral hazards-that encourage the very policies and behavior that
are the problem in the first place (Meltzer, 2000). A recent and burgeoning
empirical literature on this topic attests to its importance and, alas, to its
complexity and contradictory signals (Collier, 2002; Collier and Dollar, 2001;
Crosswell, 1999, 2001 ).

There is a logic to the ordering of these topics. Considerable controversy remains over
whether economic growth reaches the poor, over whether growth worsens income
distribution (and thus might make the poor even poorer), even whether growth is
desirable at all (Kuznets, 1966; Ravallion, 2001). This controversy must be addressed
first. Then, because the record shows clearly that economic growth is the main
mechanism for reducing poverty, we must understand how to do it. The role of
government in this will be crucial, even if a key lesson is that the governments of many
poor countries are making things worse by increasing economic risks and transaction
costs for entrepreneurs seeking to do business in their countries.

Finally, how can we help--the United States government collectively and our citizens
individually? In setting up the questions in this order, and asking them in this way, an
important assumption has been made: it is in our interests that the economies of poor
countries develop and that their citizens participate equitably in the process. To many,
especially after September 11, 2001, this is obvious, and substantial resources should be
devoted to making it happen. To many others, a clear articulation of the national interest
in helping the world become less poor is needed to justify the distortions to private
incentives that taxation, necessary for the mobilization of those resources, imposes.
Unless we know for sure that foreign assistance can speed the reduction of'poverty, this
debate is moot. Thus, to chart a way forward, we first look more closely ~i the link
between economic growth and the reduction of poverty.

7
II. Reducing Poverty

Defining poverty, establishing its causes, and finding mechanisms to reduce its incidence
are major conceptual and empirical undertakings, the results of which can only be
highlighted here. Fortunately, the literature is vast and easily accessible. A guide to
web-based accessability is provided in Box 1.

[Box 1 on sources and access, especially from the World Bank]

This literature is not, however, easily summarized because the messages are contentious
and overlaid with ideology and political rhetoric. Still, several robust themes can be
noted here.

First, how poverty is defined matters to such elemental issues as whether the numbers of
poor are increasing or decreasing, in absolute terms or relative to population. When life
expectancy and literacy, for example, are included with income in defining poverty, then
unambiguous progress at a global level has been made in the past several decades in
reducing the numbers of the population defined as poor (Fox, 2002). By contrast, if
poverty lines are established in U.S. dollars at market exchange rates, even the share of
the population defined as poor seems to have risen in recent years (Wade, 2001).

The confusion resulting from such disparate measures and results is important because
the mechanisms by which poverty can be reduced are obscured. Is income growth
everything? It is, if it is all that counts. Or can initiatives to direct health and education
programs to the poor substitute effectively for the more difficult task of getting economic
growth going? This debate can easily paralyze governments and aid agencies alike.

The only effective way forward is to make the discussion more regionally specific and
causally concrete. In any given household, village, province, city, country or region, the
reality and causes of poverty will be easier to identify than at a global level. At the lower
levels poverty becomes more visible because it is more human. Careful research can
reveal its determinants, even ways forward to reduce its incidence (Morduch, 1994). An
added advantage of this perspective is that it also avoids the rather sterile debate over
global trends in poverty and income distribution. The important action is at the country
level, or within, because that is where national policies can be effective. There are few
global policy instruments for reducing poverty.

Such a micro perspective can be frustrating, however, in two ways. First, it condemns
"one size fits all" approaches as irrelevant in many empirical settings. Second, the very
complexity of local determinants of poverty can mask the powerful but indirect impact of
i·
macroeconomic forces, trends in commodity prices, or evolving demands for skill levels
in labor markets. These factors are the ultimate mechanisms for an escape from poverty.

Balancing detailed understanding of the determinants of poverty at the local level with an
understanding of national and global economic forces is one of the most difficult tasks

8
facing poverty analysts and policy makers. It is easy enough to list plausible
determinants of poverty, harder to quantify their significance in general. The..richness of
the micro empirical reco!_ch however""~gues that low ~~6norili£...l2Dldw:rti.Yicy of poor
household~US&-ef·poverty. ·· ,

Behind this low productivity can lie supply factors, such as limited availability ofland,
skills, or appropriate technology. Demand factors, such as prices for commodities grown
and sold, availability of productive jobs, and access to urban markets for handicrafts can
also sharply influence the incidence of poverty in any given setting. Key to both supply
and demand factors is the importance of local markets to provide a low-cost and
convenient arena of exchange for the goods and services produced by the poor. It is
1 tJ,,..f
q

virtually impossible to escape from poverty except through market exchanges.

The importance of market exchange illuminates the role of governance in causing and
reducing poverty. Bad governance means poorly defined property rights, high /.
transaction costs, large economic risks, and outright theft. Markets disappear in such
environments and with them the hopes of the poor for an escape route from poverty. '
With lost hope often comes despair and fatalism. Sometimes it leads to migration to
better opportunities, whether legal or not. On occasion, it breeds violence (Collier, '
2001).

There are also more technical determinants of poverty, or factors that are visible to
observers but are hard to incorporate in general models of development. Cultural or
religious factors are often high on the list. At least in the short run, attitudes, levels of
trust, traditions, religious taboos, preferences for leisure and the like can be barriers to
rapid change. But the evidence is not persuasive that these factors are long run barriers to
economic behavior, when the long run spans centuries rather than decades (E. L. Jones,
1988).

Environmental degradation is often thought to be both cause and effect of poverty.

Especially in settings where the poor depend on common property resources to survive,
individually rational behavior with respect to resource exploitation and family fertility
can lead to a downward spiral in productivity and environmental degradation (Dasgupta,
1993). This is one form of poverty trap that is inescapable without new resources, new /
technology, or migration to better opportunities. The environmental dimensions of
economic growth and poverty reduction are vastly complicated, from deforestation and
loss of species diversity, to loss of productive agricultural land, to global warming and
climate change (Vincent and Mahler, forthcoming). This chapter recognizes these
complex dimensions but argues that sound economic growth policies are the solution to __J
these environmental problems, not their cause.

A final determinant of poverty is perhaps its most visible-hunger and malnutrition. The
development profession continues to argue over which causes which, but hunger as a
measure of poverty is widely established. Most poverty lines have an explicit or implicit
food component. Preventing famines, children from becoming acutely malnourished, and
mothers from delivering underweight babies has motivated much of the humanitarian

9
assistance delivered around the world. With abundant food in rich countries, it seems a
tragic waste not to feed the hungry in poor countries. With powerful political forces
aligned behind this reasoning and much popular support for foreign assistance driven by
television images of starving children, it would be foolish, even dangerous to ignore the
link between hunger and poverty.

And yet the link is more tenuous than supposed. The evidence for nutritional poverty
traps, where workers are too malnourished to work hard enough to feed themselves and
their families, has strong historical dimensions (Fogel, 1991, 1994; Bliss and Stem, 1978; .
Strauss, 1986; Strauss and Thomas, 1998). But simple energy shortages cannot account
for very much of the chronic poverty observed over the past several decades because the
cost of raw calories, in the form of staple foods, has fallen too sharply relative to wages
for unskilled labor (Fox, 2002). If inadequate food intake is the prj.mary cause of
poverty, the solution would be in sight. If, however, poverty is the main cause of
inadequate food intake, hunger will be much harder to end. ·

A. How Well Do the Poor Connect to the Growth Process?

The publication by the World Bank in 1996 of the Deininger-Squire data set on income
distribution and levels of poverty allowed preliminary testing of many theoretical models
that attempted to explain links between economic growth and reduction in poverty. At
one level, the new data have supported a very comforting story. There is no longer room
to doubt that rapid economic growth reduces poverty. Even cursory analysis of the
Deininger-Squire data set on changes in income distribution over time reveals only a
small handful of examples where economic growth on average failed to increase per
capita incomes in the bottom twenty or forty percent of the income distribution.

But most economists and nearly all policy makers are reluctant to stop at this point in the
analysis because modem economic history--since World War II-has left large, and in
some regions of the world, growing numbers of poor people despite global economic
growth that has been rapid by earlier historical standards. Are these poor trapped in low-
growth environments? Are there circumstances where economic growth does not reach
the poor? Do alternative measures of poverty change the story? For example, stories
about widening income gaps and the poor being left behind have powerful political
resonance even when the percentage of overall income earned by the bottom quintile has
not deteriorated. Absolute poverty carries a powerful moral and emotional charge.
Income distribution, despite its central role in neoclassical economic theory, is politically
and economically messier to handle. Fortunately, it is possible to examine the empirical
growth record using the Deininger-Squire data set to understand the relationship between
reductions in poverty and changes in income distribution and to improve the connection
between economic growth and poverty alleviation revealed by this record.

Income distribution matters because it affects how well the poor connect to the growth
process. Society might care little about income distribution per se, but a great deal about
those living in absolute poverty. Analyzing the prospects of these families requires data
on their circumstances, such as from the Deininger-Squire data set, as well as an

10
understanding of the economic and political mechanisms that connect the poor to
economic growth. One of these connection mechanisms is mediated by income
distribution, as the following analysis indicates.

Following on a burgeoning literature using the Deininger-Squire data set (reviewed in

Gugerty and Timmer, 1999), Timmer (1997) examined the question of how well the poor
share in economic growth by directly estimating the "elasticity of connection," or the
degree to which a percentage increase in overall per capita incomes translates into a
percentage increase in the per capita income of the poorest quintile. In addition, this
analysis addressed the question of whether the sectoral composition of growth matters for
the incomes of the poor, thus moving beyond absolute connections to conditional ones.

1. The Impact of Income Distribution on the "Elasticity of Connection"

The necessary technical details of this analysis are shown in Box 2. The conclusions,
challenging as they are for the ''growth-is-all-that-matters school," are easy to summarize.

[Box 2: Estimating the "Elasticity of Connection"]

Both the sector in which growth originates and the initial distribution of income matter
greatly to how well the poor connect to overall economic growth. Indeed, two
fundamentally different growth processes seem to be at work with respect to the roles of
labor productivity in agriculture and non-agriculture, and how these affect incomes in each .
of the five quintiles of the income distribution. In countries where the gap between the
incomes of the bottom quintile and the top quintile is less than twice as large as average per
1
capita income-that is, where the income gap is relatively small, labor productivity in
agriculture is slightly but consistently more important in generating incomes in each of the·
five quintiles than growth in labor productivity in the nonagricultural sector.

Furthermore, agricultural productivity has a noticeable "anti-Kuznets" effect in these

countries, i.e. economic growth actually improves income distribution rather than worsens
it. A similar "anti-Kuznets" effect is seen from the non-agricultural sector and this impact
is even more important to the poor in the long run because the non-agricultural sector
makes up, on average, 75 percent of the overall economy. It also has the capacity to grow
significantly faster than the agricultural economy over sustained periods of time. When the
starting point for economic growth is a reasonably even distribution of income, the growth
process itself reaches the poor in an effective manner. Agricultural growth is more
effective than nonagricultural growth in such circumstances.

The contrast with countries where the relative income gap is large--more than twice the
average per capita income--is striking. In the poorest quintile, workers are virtually
disconnected from the national economy (see Figure 1 in Box 2). The impact of growth in
either agriculture or nonagriculture is the same for the poor, a statistical disconnect.· In
economies with sharply unequal distributions of income, the poor do not participate
significantly in economic growth. However, the elasticity of connection rises sharply by
income class and exceeds one for the top quintile. There, agricultural productivity

11
growth is especially favorable to the rich, no doubt because of unequal asset distribution,
particularly of land. These results show the importance of understanding the impact of
asset distributions on income prospects of the poor.

2. Asset Distribution and Poverty

It is virtually impossible to understand the impact of economic growth on income

distribution and of income distribution on the rate and distribution of economic growth
without incorporating the distribution of assets held by the society. Assets are important
because they are a measure of the capital available to an individual, or society, for the
production of goods and services. Assets are likely to be distributed even more unequally
than income. In a world of perfect data, one would rather examine the distribution of
assets than income, but in reality, data on the distribution of assets are almost non-
existent, particularly for developing countries.

What empirical evidence there is, however, suggests important linkages between assets
and incomes of the poor. Deininger and Squire (1998) find that initial income inequality
and initial land inequality both have negative impacts on the incomes of the poor, but not
on the rich. Using the initial distribution of land as a proxy for the distribution of assets,
they find that asset inequality has a significant negative effect on subsequent growth and
this effect is stronger in low-income countries than in high-income countries. In addition,
initial land inequality has a negative effect on rates of schooling, suggesting that the link
between inequality and growth for the poor is mediated through credit rationing; the poor
are unable to borrow to make investments in human capital.

Birdsall and Londono (1997) also examine the impacts of asset inequality on the income
of the poor using the Deininger and Squire data. They find that inequality in the
distribution of land and education negatively impact income growth of the poor. Datt and
Ravallion (1997) examine the effects of inequality on the elasticity of poverty reduction
in India using a model similar to Timmer (1997). They conclude: "[c ]ertain inequalities
can severely impede the prospects for poverty reduction through non-farm
growth .. .Initial inter-sectoral disparities in earnings .. .influence how much non-farm
economic growth reduces the incidence of poverty. In addition, the higher the initial
poverty rate, the less effective is non-farm economic growth in reducing poverty." Non-
farm productivity is less effective in poverty alleviation in states with "poor" initial
conditions.

Additional research extends this result. Ravallion and Datt (1996) have shown that the
sectoral composition of growth matters to poverty reduction in India: poverty measures in
India have responded far more to rural economic growth than urban economic growth.
In addition, their work indicates that the connection of the poor to rural economic growth
is quite robust over time, at least in India.

Both theoretical and empirical work, then, suggest that inequalities may persist over time,
and that certain inequalities particularly penalize the poor. The next step in the research
agenda is to better understand the underlying distribution of wealth in an economy and its

12
implications for the economic and political sustainability of growth (Alesina and Perotti,
1993; Anand and Kanbur, 1993). There is virtually no data available on asset distribution
in developing economies, but it is possible to use the Deininger and Squire data on
income distribution to develop a simple, stylized model of asset distribution and its
evolution over time. The details of this model are illustrated with an example in Box 3.
The underlying framework and the key results are discussed below.

[Box 3: An Empirical Example Comparing Brazil and Thailand)

For the analysis here, capital assets are divided into four categories: physical labor,
human capital, financial capital, and social capital. Some simple assumptions are made
about the returns to these various forms of capital to generate several striking lessons.

1. Physical labor is what an individual can exert without using any other form of
capital to raise productivity. Somewhat arbitrarily, this physical labor is valued at $365
per year in terms of purchasing power parity (PPP), which is simply one of the World
Bank's poverty lines If a worker's income depends entirely on competing with a horse,
tractor, or bulldozer, by expending physical energy, the expected income is likely to be
low indeed. Incomes below $365 per year reflect significant poverty and the likely
depletion of human capital in the form of reduced health and nutritional status.

2. Human capital comes from education and on-the-job training (in addition to
physiological contributions from health and nutrition). It is useful to consider three
categories of human capital: (a) that arising from literacy and numeracy, both of which
should result from a primary education; (b) more formal analytical and reasoning skills
that result from a high school education; and (c) advanced professional skills and
research training that come from college and post-graduate education.

Again, somewhat arbitrarily, primary education in a developing country is assumed to

generate $1,000 per year (in $PPP) for the holder,. whereas finishing high school results
in an additional $5,000 per year in earnings. Thus, by assumption, a worker with a
completed high school education, or the equivalent in on-the-job training, is expected to
earn $6,365 per year ($365 for physical labor returns, $1,000 for primary school returns,
and $5,000 for high school returns). This simple assumption about returns to human
capital has powerfal implications for the distribution ofassets, including financial assets.
Because earnings from college and post-graduate education vary so widely, and are often
seen as a return on financial investment, they are included in the financial category.

3. Financial capital permits ownership of land, industrial plant and equipment,

and other financial assets. This category, of course, is what most people think of as
"assets," and determining their distribution has bedeviled both theorists and empiricists
for decades. As a simple example of an age-old controversy in economics, should capital
assets be valued at what they cost, minus depreciation, or at market value as determined
by the discounted flow of income? The market value approach has the obvious merit of
putting all assets on a similar valuation basis, and of linking directly income flows with

13
asset values. The disadvantage is the near tautology implied between incomes and asset
values. The link can be altered only when the discount rate changes.

The empirical work reported in Box 3 does not break down financial capital into more
workable components, especially land, industrial capital, and financial assets, because
this whole category of capital does not become important to income generation until well
into the development process. Lack of access to land, or industrial jobs, will obviously
reduce the earnings of the poor with no other capital at their disposal. But the surprising
fact is that variations in human capital seem able to account for most of the differences in
income distribution among poor countries, at least when income is disaggregated only to
the quintile level. This fact opens important policy opportunities.

4. Social capital has taken the economic development profession by storm

(Woolcock, 1998, 1999). By various measures, it seems to account for order-of-
magnitude differences in incomes among individuals in African villages (Narayan and
Pritchett, 1999), as well as similarly large differences in incomes among countries
(Knack, 1998). The social networks, institutional infrastructure, and level of trust among
economic agents that might account for these differences in productivity are the subject
of major research efforts, much of it in the field of economic history (North, 1992).
Without a consensus yet on how to define social capital or attribute productivity
differences to it, this chapter merely observes that there are likely to be at least two
different levels at which social capital operates, with substantially different policy
implications.

First, social capital seems to exist at the micro level, connecting individual villagers
whose knowledge of each other can be turned into collateral for loans, for example. An
entire microfinance industry is growing around this realization (1999). At the other end
of the spectrum, social capital in the form of deeply-rooted institutions that support
property rights and rule by law also seem to have macro level implications for
productivity and economic growth (Olson, 2000). Indeed, North argues that these
institutions fully account for the differences in welfare levels between rich and poor
countries. Empirical analysis of the lasting effects of different types of institutional
investments by colonial powers shows both the huge quantitative impact of institutions
on income growth as well as their lasting footprints to the present (Acemoglu, Johnson,
and Robinson, 2001; Easterly, 2002). Consequently, it does not seem outlandish to
suggest that societies with a full "portfolio" of social capital might have labor
productivity that is twice as high as in a similar society with serious shortfalls in social
capital, holding other forms ofcapital constant. Translating this reality into effective
development policy is well beyond the scope of this chapter, but the historical perspective
it requires is an important lesson in itself.

B. Is there a Special Role for Agriculture in Economic Growth and Poverty

Reduction?

Many readers will be surprised to find a prominent section on the role of agriculture in
this chapter on economic growth and reduction of poverty. Economists will be surprised

14
because the argwnents so far have emphasized the primacy of market forces, :which
leaves little scope for special attention to a particular sector of the economy. P<ilicy
makers will be surprised to find emphasis on a sector which holds little promise for rapid
economic growth or contribution to a knowledge-based economy: Agricuitu!e' is, after
all, a declining sector during a successful structural 'transformation, at lefl;st in relative
terms. Even political scientists, who might be counting _the large nwnbet of rural votes in
newly democratic poor countries, will be dubious about the desire of most governments
to redress decades of urban bias. · '

So, highlighting the role of agriculture in poverty reduction and economic growth is
tricky. But it is not wrong. The ~ectru:.specifically, and the rurai'economy
more b!_~adly..L~~Jtniqu~Wmlmrtantlo..co~hapo.or..t_g__the e~!l2mi~g~~~ ~~ - ~
pr?~USe-80?'11~HhemJiye,jn or come fro~ niial areas. Further.. g~wtb in -
agricultural product1v1ty has demonstrable economy-wide benefits, many o(which
receive no value in commodity markets where farmers sell their output. The'case for
developing an agricultural strategy is Clear, even if it must be carefully constrained by /
market realities and institutional capabilities. This case, summarized here, is presented in.
detail in Timmer (forthcoming).

1. Agriculture and Economic Growth

How does agriculture affect economic growth? First, there is the obvious national
income accountin~ identity:, the Ch?flge in national income is eqtJal t~ the ~ro~ rates in
the agricultural and nonagricultural sectors, weighted by their respective· shares in
aggregate GDP. It is worth noting that where the agri~~ltural share is large, the direct
contribution of agrjculture to to~. ecq~?mic_gro~ cru:i also ~e, s~b'sfantiaL This obvious
but often forgotten fact is also an ingr~~nt ~riJh~Jast growth of city-states such as
Singapore and Hong Kong, which never faced the drag of a large, slow-growing
agricultural sector or the need to make large infrastructure investments to modernize that
sector.

For countries where the share of agriculture is still significant, changes in agricultural
productivity can influ~n~e,the growth process tlirottgh a set of indirect and roundabout
linkages. They can be categorized by each' of the variables in a standard production
function: the location of the local technolo~isal'frontier; the rate of physical capital
deepening; the rate of hwnan capital 'deepening; and any changes in the economic or
institutional "environment" that influences how closely an economy operates relative to
its local frontier (Mundlak, 2000, 2001).

Impact on the Rate of Shift in the Local Technological Frontier.--Increases in

agricultural value added earn foreign exchange that permits imports of foreign
technology, where new technology is errd~qdi~d ip. physical,capital. It s@>_uldbe noted
that agricultural exports have a very high ratio ofvalue added, unlike man manufactured
expo s a re y eav1 y o~rted materials and compQ!].ents. On the other hand, there
may<:he few lffi:Owledge spillover effects from exporting traditional agricultural products,
thus forgoing one of the main growth stimulants from foreign trade specified in the

15
endogenous growth literature. Earning foreign exchange is one of the standard Johnston-
Mellor linkages (Johnston and Mellor, 1961). Th~re i~, how~v<;r,_evidence that very
heavy dependence on primary exports is a _signif1~!'Jllt factor influencing the probability of
violent conflict within a country (Collier, 200 i ).

Impact on the Rate of Physical Capital Deepening.--Savings from the agricultural

sector are a function of agricultural value ;illded, and- in a selni-closed economy' or orie
with imperfect capital markets, higher savings translate mto. faster physical capital
deepening (Feldstein and Horioka, 1980). the .secfor'in which the investment takes place
will depend on financial intermediaries (for private savings), or, mechanisms of savings
extraction (for example, taxation or pricing policy). This is a standard Lewis linkage
(Lewis, 1954). ·

Savings may be less productive for growth if in government hands rather than private
hands, after minimum government revenues are available tO fund affairs of state. It
should be recognized, however, that these public revenues can have very high
productivity when invested in public goods and.irifrastructure that raise the profitability
of private sector investment in agriculture (Tera.Q.ishi, 1997). If agriculture is more easily
taxed than nonagriculture in the early stages of ~eyeJopment, _perhaps by border taxes on
exports, the agricultural sector may well provide revenue for this important, initial stage
of public sector investment.

Impact on the Rate of Human Capital Deep.ening.--Rural education levels can be

influenced by growth in agi;icultur.alpJl?ductivity and rural incomes <Cfiai~ 1995; Birdsall
et at,1995). Such education can raise farm productivity directly (Jamison and Lau,
1982). It can also make the migration process much less painful and more economically
rewarding for children who leave the farm (Johnson, 1997; Larson and Mundlak:, 1997).

Improved nutritional intake can raise labor productivity through the processes examined
in historical England and France by Fogel (1991, 1994). Although iµ principle staple
foods are tradable, in fact there is a very high correlation between increases in food
production and increases in food consumption within regions and countries. The "Fogel
linkages" can thus be stimulated by growth in agricultural output, especially food output.

Impact on the Rate of Change in "Environmental" Variables.--A wide range of

variables might cause economies to produce at less than their technically efficient level.
Lack of economic freedom, poor institutions, ineffective e~onomic policies, and political
instability all have been shown to slow down economic gro~, yvhen controls are
included for initial conditions and factor accumulation (Barro, J997). How changes in
agricultural productivity might affect these "efficiei:icy shifters" ,is a matter of
considerable speculation and relatively little empiric.al evidence. Two mechanisms for
which evidence is accumulating involve price stability, ~ecause investments are more
efficient when signal extraction problems are reduced'(Lucas, 1973; Dawe, 1996; Rock,
2001), and political economy considerations, because restive rurcll populations can
challenge political leaders if they are left behind during the process .of rapid economic
growth (Anderson and Hayami, 1986).

16
Other linkages that Johnston and Mellor identified i:iiight also work through these
"environmental" variables. For example,, producing ra\\'. materials for industrial
processing suggests that capacity utilization in_the ·industrial ~ector might depend on
agricultural productivity. Earning foreign e:x;change might have the same impact on
imported intermediate goods, which are .often cruCial for produCing manufactured
exports. · ·

It is, of course, important not to forget the c~tical direct contribution that agricultural
development has made historically to economic growth. As stressed by Lewis (1954)
analytically and by Johnson (1997) empirically, lower food prices stimulated by rapid
technological change in agriculture have contributed substantially to higher living
standards directly, especially for the poor who spend a large share of their budget on
basic foodstuffs, and indirectly by keeping real wage costs low in the industrial sector,
thus fostering investment and the structural transform'.adon. It is argued, however, that
these benefits oflow food prices are.as ~asiJy a~cessed by t~l!4e a~ by investing in the
domestic agricultural 'sector (Sachs, 1997). What is the significance of other
contributions from agricultural modernization· that would be missed with a pure trade
strategy?

Plausible candidates include the loss of backward and forward linkages that connect cities
with the countryside. Without these linkages, societies ,ris1': greater vulnerability to
fluctuations in world markets, inequities between riltal and urban inhabitants, more under
employment in rural areas, and exc~ss migration. The returns to good rural-urban
linkages include a relatively smoo~h structur~l transformation, as seen in Taiwan, in
contrast to the difficulties seen in Thailand (Tabor, 2002; Timmer, 1988).

Surprisingly, in view of the length 9ft~~-tqe,depate }fas been go~ng ~n, there are still no
satisfactory tests of the imp~ct of p~'g~~Jp:_~gric:tJltmliLproducti.vity.on,the~~eral
mechanisms of"catch-up growth" outlined above, or on the value of good rural-urban
linkages. There is evidence generated. from a large data-gathering project at the World
Bank led by Don Larsen, Will Martin, and Yair Mundlak, that total factor productivity in
agriculture tends to grow faster than in manufacturing (Martin and Mitra, 1996). This
result alone argues that past investments in agriculture have had large economic returns
(Mundlak, 2001).

2. Does Rural Growth Mediate Poverty Reduction?

The work of Ravallion, Mellor and Timmer shows the empiric~I_roJe of rural gro · ~
connecting the poor to economic growth, both within andou s1 e t e rura · economy.
Without firm theoretical underpinnings, however, these empirical 'observations provide
only casual guidance to policy makers seeking to make the groWth process more pro-
poor. It would be better to uriderstahd the mechanisms .l:J.t ~ork ~s _well as the facts
(Sarris, 2001 ).

17
Fortunately, much progress has been made in the past decade in identifying these
mechanisms. Foremost in this ~f~ort i~.t4e.~e.ct(nt moc;iel ~f agricultural growth, rural
employment and poverty reductiqn that ymphasizes ~he roJe of.n~Lntr~dables in pulling
underemployed workers out of agricµlture into the nonagricultural rural economy
(Mellor, 2000). This in~del, dra"'.'ing ~n Mell.o~'s earlier work in India (Mellor, 1976),
shows the importance of rural inc.omes as the dxiver pf demand for the goods and services
produced in the nonagricultural rural eco~~my ~d ~o~ th!s econpmy J~§ to urban
demand, especially when it is driven by rising incomes from workers in labor-intensive _J
export industries.

The Mellor model is the first to expl~..r.e~ogni~etw0en

manufactur§i'.texpol'[p.e~pe role of the nontradables sector in the rural
eoorfomy;- and subsequent reductions in poverty. Thus the model explains why countries ,
with rapid growth from labor:igtensive_rnanuf.agtured f!Xports,Jhat a!so ~~stantial
agricultura! sectors, ~ad suc}J. good xecqrds of poyert~duction. But the nontradable '
sector is often ignored by policy makers and donors pr~c~sely because so much emphasis
is placed on the role of expqrts and open-economy.strategies for economic growth. ·~
Retargeting public expenditures in support of a more balanced strategy will not sacrifice ~
overall growth performance but it will increase its impact on reducing poverty (Mellor,
2000).

Two other components of the relationship betyv~algr.o.:wtlumd povroy reductign

should be noted. First, political commitme11ts 'to rural growth imply a more balanced
politteatetronomy, with less urban bias than has been seen in tzj.ost qevdoped countries
historically (Lipton, 1977, 1993; Timmer, 1993). T~ develop~g world has already seen ,,
a notable reduction in the macroeconomic biases a ainst a ·culttire such as overvalued #
currencies, repress10n o mancta s stems ploithre terms.G.i4Fati~~her ,· --.
prog~ e expected as de~gcracy spread$.....and.empowers~thaIUfal.populatianJn..,,
p~~---~~-~~~-

The second important component is the linkage between urban and rural labor markets,
often in the form of seasonal migration and remittances. rhere is no hope of reducing
rural poverty without rising real wa es for rural workers. Rising wages have a demand
and a supply 1mension, and migration can affect both in ways that support higher living
standards in both parts of the economy. Migration of workers from rural to urban areas
raises other issues, of course, but those issues depend fundamentally on whether this
migration is driven by the push of rural poverty or the pull of urban jobs. Either way, the
food security dimensions of rural-urban migration are clear. Urban markets become
relatively more important in supplying food needs for the population. Whether the rural \
economy or the world market is the best source of this supply will be one of the prime
strategic issues facing economic policy makers (Naylor and Falcon, 1995; Tabor, 2002).

C. Implications for Policy

This perspective on the l~ pet\yeen economic groyyth and poveey r~duction suggests
two policy arenas where government actions might strengthen the connection even when
• " " t

18
the starting point with respect to income and asset distribution is unfavorable. In tum, we
discuss the mechanisms for "getting agriculture moving," to use Artlitir Mosher's
memorable phrase (Mosher, 1966), and initJatives to develop human capital, especially
for the rural population. ·

1. Getting Agriculture Moving

There is no great secret to agricultural development. Mosher (1966) and Schultz (1964) /
had identified the key constraints and strategic elements by the mid-1960s. New
agricultural technology and incentive prices in local markets combine to generate
profitable farm investments and inc;:ome streams that simultaneously increase commodity
output and lift the rural economy out qf poverty (Haya:riii and Ruttan, 1985). The process
can be speeded up by investing in the human capital of rural inhabitants, especially
through education, and by assistance in the development of new agricultural technology,
especially where modem science is needed to play a key role in providing the genetic _.)
foundation for higher yields.

Beyond this level of general understanding, however, the diversity of rural circumstances
has sharply impeded its implementation. At the regional level, for example, neither the
agricultural technology nor the incentive prices in rural markets have been reliably
available in Sub-Saharan Africa. In Asia, success in linking the nontradable sector in
rural areas to urban markets and labor-intensive export growth has been mixed at best.
And in Latin America, extreme rural poverty has largely migrated to urban areas, so the
poverty problem is now primarily an.urban problem (Painter, 2001; Haddad, Ruel and
Garrett, 1999). Central America and Mexico continue to face acute rural poverty,
however, and rural strategies will l;>e needed to reduce it (de Janvry and Sadoulet, 2000,
2001). .

The mechanisms for both technology de:velopment and provision of rural price incentives
are no longer as clear as they were hi the 1.960~. The CGIAR system has a laudable
record of important breakthrough~ for_niflllY ~f the world's staple foodcrops. But funding
for the system \las b_een threatened as the market prices of these crops .have dropped to
in
historic lows, under the weight of productivity gains developing countries and publicly
subsidized crop surpluses in rich countries. Few countries have the scientific resources to
conduct basic crop research on their own, so a large question·1ooms. Where will
agricultural technology come from for the additional 3 billion people expected in the next
50 years? Biotechnology holds out both promise and concern; even in the best of
circumstances it is largely a product of scientific enterprise, public and private, in rich
countries. ·

There is an obvious role for the United States in answering this question. First, starving ·
the CGIAR centers of funding to pursue essential and basic crop research with spillovers
to many countries is very short-sighted, and U.S. leadership in restoring budgets is likely
to have major add-on effects. Second, our university system is the best in the world at
training scientists in basic biology and applied agricultural fields: We have an
opportunity to provide the next generation of these scientists for. the entire world. 1I

19
Apart from its indirect impact on funding for the CGIAR system, the sharp drop in
commodity prices in worl~ mark~Js also h~ a more immediate impact (Fox, 2002). Open
borders and flexible markets for (oreign ~xch~ge.tran~Il!it t4e,se.low prices directly into
the markets of poor countries, often with devastating impact on local farmers (Dawe,
2001; Tabor, 2002). Rich countries :fjnd ways to protect their farmers a,gainst such low ·
prices, but poor countries cannot afford the subs~dies, or defepd th~ trade interventions,
that would be needed to do the same. "Agriculture-led economic growth" is impossible
1
unless it is profitable. ··

Instability in food prices also remains a, cpnce.m, especially with open borders and the
possibility of sharp movements in exchange rates (Islam and Thomas, 1996). In 1998,
for example, the collapse of the Indonesian rupiah during the financial crisis caused the
landed price of imported rice to increase more thah four-fol<l."Ind~ed, for a time, it was
profitable to export rice in the same year as' one of the worst el Nino-induced droughts in
history! In open economies, food price instability ha& macroeconomic roots as well as
local supply and demand roots. If some degr~e of food price stability is a political
imperative, new tools will be needed to provide lt (Tiinnier:'1989).

Many place their hopes for solving the problems of pric~ levels and price instability on
new rules regulating agriculnire in the WTO. Bufnegotiations leading to useful new
rules are likely to be very difficult, with Europe and Japan still extremely reluctant to
abandon their :farmers to free markets. The United States i's under great pressure from
Congress not to give away its subsidy tools as mechanisms to ~eep U.S. farming
profitable, but the U.S. Department of Agriculture (USD'A) has taken a clear stance in
favor of reduced subsidies and freer trade in agriculture.' '

Lowering tariffs and other barriers to trade is fundament(;!.l to expanding "')

exports. The average food and agricultural tariff is 62 percent, much higher than
tariffs on manufactured items. Both developed and developing countries have
high tariffs. Exports to the large potential markets in South Asia (including India)
and to South America must overcome tariffs of 113 and 40 percent, respectively/
The United States has one of the lowest food and agricultural tariffs,~ 12 _
percent, and thus-standSmgaiifffiiiifensely from ambitious tariff cuts. However,
the United States still maintains some high tariffs that protect specific
commodities.
In addition to tariffs, high levels of domestic support for agriculture and
export subsidies distort agricultural markets. In contrast td tariffs that are applied
by almost all countries, developed countries account for virtually all domestic
support and export subsidies. The Organization for Economic Cooperation and 'J
Development (OECD) estimates that in 2000,. dev~loped countries' total support
for agriculture was $327 billion. In that same year, total production supports by
the European Union were $90.2 billion, compared to $49 billion by the United .
States. The European Union dominates use of export subsidies, accounting for
approximately 90 percent of total annual spending since the Uruguay Round /
Agreement on Agriculture (URAA) took effect.

20
 USDA research shows that removing all forms of agricultural protection
and support could raise world prices 12 percent, over half of this from removing
tariffs alone. Our producers and the industries they support could see the value of
U.S. agricultural exports grow 19 percent. Global economic welfare would
increase by $56 billion annually by removing existing distortions (USDA, 2001,
p. 40).

The way forward on these negotiations is badly in need ofleadership. USAID is in a

strong position to provide this leadership, speaking as it does for development interests rp
l
within the U.S. government. Furthermore, USJ\!D's recognition of the "new agriculture'~ ..
in its Background Paper 011; Future Directions for A,iriculture is a.'signal of intellectual '\ e
·and bureaucratic readiness to take on this challenge (USAID/EGAT, 2002). ~ff ••• /

Part of the challenge may involve acceptance by the United States of ~griculture's
"multifunctionaliiy" as' tl:ie basis for· do~es!i<-: .policies that have clear social,
environmental, or secririty rationale.' With thU? acceptance, the United States could take
the lead in the Doha Round of WTO negotiation8 to design rul~s ~xplicitly recognizing
what reasonable functions might be f9r agrfoulture" iri different countries and' at different
stages of development. For example, environmental prote9tion would be anaceeptable
role for domestic agricul~:;tl policies in aJf _co00.'ti:ies, wlierea~ policies to stimulate basic
grain production to enhance. domestiC' food segurity' would be restricted to countries with
limited access to world markets or' poorly developed internal marketing systems.

In the end then, what are the components of an..agric11ltural strate..gy and how can USAID
help countries develop one? First, obviously, is a supportive macroeconomic policy, one
that yields low inflation, a reas~na~_lY. ~taple _exchange ra!e, positive real interest rates,
I
and perhaps some monitoring of short-run capital flows. Second, "getting prices right"
extends good macro policy to the trade arena, where an o en econom with.low barriers
to internal and external trag~~@Q.!!lcigenerate a leveJ .pl~iog field.fOLpr.Qd~.

-
consumers alike.

What remains after this? The externalities from rural growth outlined above argue for
pol.icy. att .. and bu.dget. priorities fi.or the rural nontradables sec.!QL.once ~~
. en.t.ion
technology is it1J~I~e ~ the_gasis for pJoflt€!ble..§rm!ng. Part of the profitability for this
I
sector wfll come from a labor-intensive export sector that is successfully linked into the
global economy. Rapid growth in this export sector ereates demand for labor directly as
well as for the goods and services of the rural economy that raise demand for labor
indirectly. ·

Improving the rural finan~i~ system, both ~o vermit farmers toJnake long-run
investments and as a vehicle for handlin~ intersf'.ctoral financial flows such as savings
and remittances, will take time, buf is essenti~l to a successful structural transformation.
None of this is rocket sci{f~ce, but .al.I of it requires talented ,eolicy analysts and
government administrators. Training them in U.S. universities and empowering them
when they return home is a powerful form of U.S. foreign assistance and one in which
l
USAID has considerable experience.

21
 2. Investing in Human Capital

Investments in human capital improve the distribution ofassets in the early stage of
economic development, and therein is a clear policy message. For "pro-poor growth," a
country must invest in the human capital of its poorest citizens. At the earliest stages this
will involve primary health clinics, household food security, and access to rural schools.
Policies that encourage the efficient functioning of rural financial markets can also play a
role in increasing the poor's access to capital. Later it will mean opportunities for high
school education and on-the-job training as unskilled and semi-skilled labor. Such
investments, if broad-based and of adequate quality, will keep the distribution of income
from becoming highly skewed until well into the development process, and thus lead to
the near elimination of absolute poverty. Taiwan and South Korea managed such
investments until middle income status; Brazil, the Philippines, and Thailand did not.

An optimistic policy interpretation of these results is that fiscally manageable investment

strategies are available for even the poorest countries to set themselves on a growth path
that includes the poor. The pessimistic interpretation suggests that political forces or bad
governance will keep this from happening where the "starting point" in income and asset
distribution already prevents the poor from connecting to the growth process. But surely
this is a result that the donor community can grasp--it provides donors with a rationale for
investing in the very people that countries' leaders themselves might choose, or be forced,
to ignore. Then the policy dialogue, and the resources that could be mobilized behind it,
can have dramatic effects.

III. The Role for External Actors in a World of Endogenous Growth

If economic growth is endogenous to the knowledge-generation process in each country,

the institutions of good economic governance must be built locally, and domestic
agricultural potential can only be realized in each specific ecological setting, how do
external actors play a role in economic development? The answer comes at three levels:
through direct engagement by individuals, through trade and investment by firms, and
through collective choices exercised through governments.

A. Through Direct Engagement

Americans are deeply troubled by images of starving children with flies swarming around
their vacant eyes, by ten-year old boys with Kalashnikov rifles fighting brutal and
seemingly mindless wars against their own people, and by angry rhetoric from the Third
World that holds the United States responsible for these ills. Americans as individuals,
as 50 years of polling evidence attests, are eager to help. The nation has a deep well of
humanitarian concern that wants to end hunger and poverty, resolve bloody conflicts, and
extend human rights to all. But the polling evidence also consistently reports that the
public doubts the ability of the U.S. government to achieve these goals. So, how do we
help?

22
Opportunities abound for Americans to be engaged. As private individuals, we send
money directly to poor people, poor villages, or poor countries through our churches,
temples, synagogues and mosques. We join and support service organizations such as
Rotary clubs and special-function, non-governmental organizations (NGOs) such as Save
the Children or Project Concern International. We even send money directly via Western
Union or other channels to our friends or families abroad. Such direct remittances now
total more than $50 billion per year to developing countries and rival official
development funding.

When viewed from this perspective of individual freedom of choice, the opportunities for
U.S. residents to help solve the problems of poverty in less developed countries run
across an incredibly wide spectrum, a nearly complete private market of mechanisms far
personal involvement in international development.

At one end of this spectrum, the array ofNGOs involved in overseas development is vast
and growing daily: an individual can chose among those focusing solely on emergency
humanitarian assistance, such as the International Red Cross, or NGOs dedicated to
longer-term development efforts, such as OXFAM, WorldVision, or Catholic Relief
Services, with programs on education, on-the-job training, or agricultural extension.
Recently, two former World Bank officials have launched www.DevelopmentSpace.com, "the
e-Bay of development assistance." The site links social investors in rich countries with
development entrepreneurs in poor countries, by-passing the bureaucracy of official
development agencies. The opportunities for individuals to be directly engaged in
development activities according to their own preferences and financial abilities have
never been greater.

B. Engagement through Trade and Investment

Americans also enjoy the bountiful and cheap consumer goods delivered from factories in
developing countries. At the same time, they are concerned by reports of child labor and
sweat-shop working conditions employed to produce these goods. This tension
highlights the other end of the spectrum by which Americans, as individuals, can
participate in the development process: through trade and investment. The tension is
reflected in the debate over the costs and benefits of globalization, the potential to use the
Doha Round of WTO negotiations to further development agendas, and even whether
pulling the most backward economies into world markets might reduce the level of
international terror and conflict.

The market mechanisms that have developed the economy of the United States generate
opportunities to invest in U.S.-based companies that have operations in the developing
world, providing them with capital to deepen their involvement. These operations range
from the labor-intensive manufacture of garments, shoes, and electronic goods for export
to rich consumers, to the provision of capital-intensive infrastructure such as electricity
generating plants, communications networks, and transportation systems, to the
exploitation of natural resources that are needed to keep the global economy functioning.

23
Individuals are free to invest in whatever companies best satisfy their own personal
understanding of how economic development takes place, or how best to reduce poverty.
Nike, The GAP, and Nordstrom's procure low-cost shoes and clothing from labor-
intensive companies abroad. Bechtel builds airports, transit systems, and dams. Boeing
supplies aircraft that link the Third World to the First. Investment in any such company
is to some extent an investment in development in this era of globalization.

More broadly, and no doubt riskier, investments can be made in single country funds sold
on U.S. stock exchanges or even in particular companies in poor countries themselves
that are listed on indigenous stock markets. The integration of global markets has
brought not just exotic products to the shelves of U.S. retailers, but financial
opportunities to our investors.

The point is that citizens of the United States have ample opportunity to exercise their
personal preferences on how to be involved in the development of poor countries, as
donors, consumers, and/or investors. This poses a difficult question: what remains for
the U.S. government to do, using taxpayers' money on behalf of development in these
same poor countries? A satisfactory answer to this question must involve actions or
investments that private individuals acting alone, or private markets collectively, cannot
accomplish on their own.

C. Public Choice Mechanisms and the Role of Government

Since one major function of government is to solve problems that require collective
action in addition to, or even to correct for, private actions, answering this question is
basically the same as asking "what is the appropriate role of government in the
development process?" For citizens deeply skeptical that governments can be trusted to
do anything but police their own borders, the answer will be "none." For citizens who
see chronic poverty in the Third World as a failure of markets in the first place and a
potential breeding ground for terrorists, the answer will be for the government to address
the underlying causes of poverty.

At least two dimensions of government role are applicable here. First, what can third
world governments themselves do to speed the development process and improve the
distribution of its benefits, so that the poor in these countries benefit sooner rather than
later? Ancillary to answering this question is the related issue of whether .foreign
assistance directly to these governments can help them help themselves. The empirical
evidence suggests that best results occur when these two factors work together (Collier
and Dollar, 2001; Collier, 2002)

The second dimension of governmental role is whether there are international institutions
and publicly provided services that rich countries such as the United States should be
building and providing to improve the environment in which national development takes
place. Examples of such "global public goods" might include the International Financial
Institutions (IFis), with their mandates to stabilize the global economy and assist in the
development process; internationally funded and commanded peacekeeping forces to

24
separate warring civil factions; agricultural research with regional or global applications;
or funding for development of vaccines against tropical diseases that are of little threat to
citizens of rich countries that are home to the major pharmaceutical companies. These
global public goods, and the institutions to develop and deliver them, will not be provided
in adequate amounts by private markets and individual action (Dalrymple, 2001).
Governments, using taxpayers' money, will be needed to provide these global public
goods.

USAID can play a crucial role in this process for the United States government through
three activities. First, the agency can help develop the intellectual case, analytically and
empirically, for the role and content of these public goods. Second, USAID can provide
public leadership among the donor community in the mobilization of resources to fund
these public goods and in the design and management of the institutions that will develop
and deliver them. And third, USAID can be the appropriate vehicle to channel U.S.
taxpayer support for these public goods to the international institutions that need them.
For example, despite historically tight budgets, USAID-funded support for the
Consultative Group for International Agricultural Research (CGIAR) has consistently
been one of the major, and most stable, sources of on-going funding for this critical
institution.

IV. The Role of Development Assistance

Development assistance is under challenge in most western societies. One set of critics
argues that the funding levels are inadequate-Western European leaders are pushing for
a doubling of official development assistance (ODA). In the United States, there is
widespread doubt that development assistance works at all, so the case for any increase is
dismissed. Analysts in the World Bank have been working hard to sort out what works
and what does not. Their answer, perhaps not surprisingly, is that despite mistakes in the
past, the donors in general and the World Bank in particular now know how to help poor
countries get on a sustainable development path. More money, they argue, can be used
very productively (Collier, 2002). It is useful to review some of this historical experience
before reaching a conclusion for USAID.

A. Finding Focus

In the four decades since USAID was established in 1961, the goals and mechanisms of
development assistance have broadened considerably. From an early emphasis on growth
in gross domestic product (GDP) and containing communism, USAID's mandate grew to
include, among many other things, reductions in poverty, improvements in child health,
gender equity, environmental sustainability, transition to market economies and
democratization.

Many institutions involved in development activities saw similar broadening of agendas.

The Development Advisory Service (DAS), founded by Harvard University in the early
1960s to help poor countries prepare economic development plans, expanded its scope in
1975 to become the Harvard Institute for International Development (HIID). New

25
activities in health, education, and rural development were integrated into the Institute's
traditional core ofmacroeconomists. The University's program on Women in
Development was housed in IDID. An environmental program started in the late 1970s
with the arrival of Theo Panayotou. Both in academia and government, development
came to be seen as a multifaceted and complex process.

This progress came at a cost, however. Focus was lost as agendas multiplied. Harvard
closed HIID in 1999, arguing that it was managerially too complex for an academic
institution. Outside observers note that a failure of managerial oversight of HIID projects
in Russia, funded by USAID, may have contributed to Harvard's decision. USAID was
certainly distressed by apparent conflicts of interest among advisors and inappropriate,
perhaps corrupt, outcomes. Managing such complex relationships was clearly difficult,
perhaps beyond the range of an institution organized around academic procedures.

But the difficult management tasks extended to USAID as well. In the early 1990s, Brian
Atwood tried to sharpen USAID's increasingly blurred focus by withdrawing the Agency
from its economic growth agenda and emphasizing several themes of great interest to
Congress: short-run humanitarian assistance, especially food aid; health care, especially
child survival and family planning programs; environmental sustainability, especially the
development of agricultural technology for poor farmers, including women, working in
fragile ecosystems; and gender issues more broadly. As the challenges and opportunities
presented by the collapse of communism in the former Soviet Union became apparent,
democratization was added as a USAID objective.

Somehow lost in the multiple agendas and Agency efforts to program effectively in the
face of developmental complexity was the need for poor countries to have growing
economies as the only sustainable solution to all of their broader problems. The Natsios
Administration recognizes the simple fact that economic growth is essential to any lasting
reduction in poverty. Equally simply, the Administration recognizes the power of the
increasingly integrated global market economy to drive the growth process in rich
countries and poor alike (Hannon and Rhee, 2002).

The failure of most countries that are still poor to connect to global markets illuminates
the task. It is to "get economies moving" if poverty is going to be reduced. This need for
economic growth, and the focus on integration into global markets that it suggests, has
been the organizing rationale for this chapter. Other chapters develop additional
dimensions to the rationale for USAID and program elements to operationalize it. But to
tum on its head the title of Paul Streeten's famous book on meeting basic needs, "first
things first" means reestablishing economic growth as the foundation of development
(Streeten, 1986).

B. New Thinking on the Role of Foreign Assistance

Does foreign assistance have any role in turning around the rather difficult record of
economic performance over the past decade? Obviously, if entire societies chose to make
themselves progressively poorer, the donor community is relatively powerless to

26
 intercede. Even when the general population desperately wants better governance and
economic opportunities, a venal government can make foreign assistance problematic, as
the waning days of Mugabe in Zimbabwe indicate.

This chapter concludes with a focus on the role of foreign assistance in stimulating
economic growth in poor countries. That is a narrow task when matched against disaster
relief and humanitarian assistance, health and child survival interventions, strengthening
democracy and governance, and fighting terrorism. Indeed, one might ask, why bother?
If the tasks just listed are successfully implemented and governments of poor countries
create a favorable business environment and capacity for integrating into the global
economy, what else does foreign assistance need to do?

Even ifthe answer was not clear when the Woods Report was drafted in 1989, it is now.
Without visible, sustainable economic growth, none of the above tasks can be successful
for long. Economic growth is not sufficient to solve any of these problems, but it is
necessary to fund their domestic solution and to generate the political support to continue
solving them. A failure of economic growth means no sustained health gains, weaker
governance and faltering democracies, and the emergence of rogue states that harbor
narcotics dealers and terrorists. Economic growth is everything in the sense that nothing
lasts without it.

1. Foreign Assistance and Economic Growth

So the important question is, how can foreign assistance stimulate economic growth in
poor countries? The empirical record is mixed at best, but that is hardly surprising in
r
view of the perverse incentives that linked donors and governments of developing
countries over the past five decades. After all, ~~!Y little fon~igtl.fi~~J§~Ce, ~t least as
measured as Official Development Assistance, or ODA, has specifically been justified as
S!9Wth~enhanGit1J~· and . a SllJ:~~L~hareof .that,h~~ ~<::tti(llly'~~~.ii used in envirlT:ti11'!~11ts··:···-­
.where ithad mucli.cllance:ofenhancing growth (Alesina and Dollar, 2000). The Collfor.:
Dollar results from World Bank research show there was no statistical relationship
between volumes of ODA to individual countries and their record of economic growth in
the 1960s and 1970s. By the 1980s and 1990s, however, ODA was growth-en)lancing
conditional on favorable economic policies. Crosswell finds a similar result for USAID
.assistance:· the part of l]SAID funding that mightlegiti:rrtafolybe srud· to be targeted to
/ poor countries to assist with their economic growth was successful in that task.
<. Unfortunately, only a small share ofUSAID assistance was so targeted (Crosswell, 1999,
~\ 2?01 ). Even worse, wh~n ass~s~ce ?id work, the effec~ve "t~ rate" ?n success was so
····highasto.serve as a senous dismcentive to good economic policy mak:mg (Meltzer,
2000; Fox, 1997).·

Historically, the shifting ideologies that justified foreign assistance to an American

U opulace concerned about domestic iss.ues, especially jobs, inflation, and security, have
ever focused squarely on economic growth in poor countries. Reconstruction of Europe
after World War II, the containment of communism during the Cold War, the transition to
democracy and market economies in the former Soviet Union, and now a commitment to

27
 a broad war on terrorism have all trumped, at least in rhetoric, the rationale for long-run
investments in the economic growth process in poor countries. But at least the record
now shows that foreign assistance can speed up that process in the right settings.

2. The Mechanisms of Foreign Assistance

There are at least four mechanisms by which U.S. foreign assistance can speed up
economic growth in poor countries, and the list does not include improved access to U.S.
markets, the flow of direct foreign investment, remittances, and the actions and
involvement of U.S. based or funded NGOs, all items that are likely to be more important
for economic growth than official development assistance. The list does include:

1. Direct financial support of policies, programs and projects through bilateral

assistance activities, often implemented by private sector U.S. contractors;
!~"'2. Engagement of governments of developing countries in policy dialogues,
·1, often with the. explicit or implicit promise of greater aid if policy conditions
L aremet;
3. Production of new knowledge about the development process through
research or project activities funded by USAID or other agencies of the U.S.
government; and
4. Bringing the United States "to the table" in broader discussions, often in
multilateral settings, for diplomatic or trade negotiations. The value of this
seat, of course, is its capacity to help open the door to the "$10 trillion
economy." Connecting to the U.S. economy through trade and investment
flows will be the main engine by which poor countries speed up their
economic growth.

The budget of US AID is no longer large enough for the agency to be a significant player
in the direct provision of financial resources to poor countries, even in those
circumstances where the payoff in faster economic growth might be high. Further, the
agency no longer has the in-house analytical capacity to determine which countries might
offer those circumstances. Neither situation is likely to change significantly within the
time horizon of this report. Accordingly, USAID should seek its influence over the
development process primarily through policy dialogue, the production and
dissemination ofnew knowledge, and as an advocate for trade-led growth both at home
and abroad.
<: ..
--· · ·

Of course, engaging in meaningful policy dialogue also requires extensive knowledge of

a country's political economy and the local capacity for pragmatic policy analysis (as
opposed to ideologically driven policy recommendations not based specifically on local
conditions). But here, USAID's premier role as a contracting agent can be a very
efficient mechanism for accessing knowledgeable analysts, especially if there has been an
on-going effort in key countries to build the research and knowledge base as a way to
produce and retain such analysts. Thus policy dialogue and knowledge generation should
be thought of as mirror images that require coordinated and integrated support for lengthy
periods of time. An especially poignant example that links the deteriorating capadty for

28
 economic growth in Africa, because HIV/AIDS is "hollowing out" institutions of
economic governance, to an indigenous research program to understand the
consequences, is presented in Box 4.

[Bo:x 4: Letter from Malcolm MacPherson to Andrew Natsios on HIV/AIDS]

Many environments are hostile to economic growth but still require a U.S. presence for
diplomatic or security reasons. Even in such settings it is possible to provide support to
reform-minded elites through travel and study grants or U.S.-supported forums as
platforms for reformers' messages. "Engaging civil society" has become something of a
donor platitude because it is meant to serve as a substitute for discussions with venal or
unresponsive governments. Still, building civil institutions, including legal and political
institutions, takes time and investments, and donors can help. In addition, no one can
predict when a "window of opportunity" will open for effective dialogue with a
genuinely reform-minded government, and having effective civil institutions to help
shape policy and implementation can be crucial at such times (Grindle and Thomas,
1991).

Having the United States "at the table" as policy discussions are held, trade openings are

l
~
negotiated, and international treaties are brokered is no doubt the most important way the
U.S. can speed economic growth in developing countries. Indeed, the role is so important
for trade and investment discussions that a separate chapter of this report is devoted to the
topic (Hannon and Rhee). But a serious issue arises immediately with this recognition
because the United States does not speak with one voice. Even the United States
government has many agencies with many agendas, and outsiders are often bewildered
by which one is actually "speaking" for the country. As a very simple example, who
speaks for the United States Government on economic assistance to poor countries? On
different occasions, the correct answer can be the President, the Secretary of State, the
Secretary of the Treasury, the Secretary of Agriculture, or even the Administrator of
US AID.

The issue is important because the United States is the most powerful nation in the world.
Obligations come with this power. One is to provide leadership on issues that affect
global and national security, whether directly as in the fight on terrorism, or indirectly,
through efforts to develop poor countries. Thus clear guidelines on which agency will
take responsibility for which dimensions of foreign assistance, even narrowly in the field
of assistance for economic growth, will be welcome both within the U.S. government and
by participants beyond.

C. So? Missed Opportunities, New Opportunities, and Next Steps

There is an understandable political impatience with academic debate. While the main
points of agreement are ignored as uninteresting, controversy rages over the decimal ·
points. Some of that is on view here. Economic growth is good for the poor. But it
could be better. Agricultural development is good for food security, poverty reduction,
and economic growth. But, in theory, the same or larger benefits could be achieved by

29
importing cheaper food from world markets and investing the difference in export
processing zones. And so it goes, the best as the enemy of the good.

And yet. It is only from the perspective of theory that we know what is possible, what we
have missed, what new opportunities await. From this perspective, the global economy
missed three opportunities to assist economic development over the past several decades.
First, two decades intervened between the first and the second world food conferences
with little to show in terms of increased food security and reduced poverty in the most
vulnerable countries, those that might have hoped that Henry Kissinger's promise in 1976
that no child would go to bed hungry within a decade actually would translate
.
into action. 1.1
Other countries, especially in East and Southeast Asia, used the two decades to improve J'
their rural infrastructure, agricultural technology, and economic competitiveness. They ~
were rewarded with reduced poverty, improved food security, and rapid economic fl
growth, but the global promises figured little in this performance. ~

Second, subsidies to farmers in rich countries have become larger over the past two
decades, n9t smaller, despite promises made at the Uruguay Round. The result has not
just been a large budget burden in OECD countries. More importantly for developing
countries, the result has been increasing surpluses dumped on world markets, depressing
world prices and the incomes of farmers in poor countries who have to compete with
these prices. The best guess is that every dollar of agricultural subsidies in rich countries
costs farmers in poor countries a similar amount. And official development assistance is ·
only one quarter of this total. It is not a fair trade.

Third, the Cold War took a terrible to1l on good governanc~. If we now recognize how
important good economic governance'is to the foundations o~economic development, we
are just coming to realize how the willingness of governments in the West to do business
with any government ostensibly in the anti-communist camp undermined those
institutional foundations. Many decades have been lost in the creation of sound
economic governance and they cannot be recaptured overnight.

From this rather desolate historical record comes unique historical opportunities. The
United States in particular, and the West in general, won the Cold War. With that victory
comes the freedom to seek new goals. Angry individuals, whole societies left behind by
the sweep of globalization, even trading partners frightened by the technological
superiority of the United States, are challenging this freedom. When the challenge has
been in the marketplace, it has been met in the marketplace. When the challenge has
been violent, it has been met with a military resolve that has surprised even our allies.
But the war on terrorism must not mask the unique opportunity open to the world at the
start of a new millennium--to build peaceful, prosperous, and open societies that engage
the rest of the world as willing partners.

This chapter has tried to identify this potential, the constraints on reaching it, and the
realistic alternatives before the United States and the developing world. But realism is in
the eye of the beholder, and it is the nature of constraints to keep changing. So the real
purpose of this chapter has been to present a framework to organize and understand

30
alternatives, conditioned by their settings, as a platform for building country-specific
strategies that address their unique challenges in getting economies moving while
bringing the poor into the process. "Unique challenge~" is the message here. No single
key will fit as countries seek to' unlock their potential for rapid economic growth. ·

USAID also has unique opportunities in this challenging context. It can provide political
leadership within the U.S. government, and the world community at large, on..fost~rjllgJt.)
development-oriented. intematio!!~!rade !~gime_o_ut Qf..t_M Doha Round of WTO
negotiations ..,It can lirlktne knowledge generated in the wOtla's'best sy'stem~ofresearch
universitiesfo the technical assistance required on a day-by-day basis in developing J
countries. Perhaps most importantly, USAID has the opportunity to reclaim the
intellectual leadership that was its birthright-to speak out on the importance to the
United States of fostering economic development and reducing poverty everywhere in the
world.

31
 Box 1
International Development Resources on the Internet

[This annotated list was taken from Perkins, Radelet, Snodgrass, Gillis and Roemer, Economics of
Development, 51h Edition, New York: W.W. Norton and Co. 2001.]

The World Bank

http://www. worldbank. org/htmllextdrlregions.htm
http://www. worldbank. org/htmVextdr/thematic. htm
http://www.worldbank.org/proverty/

The World Bank website contains a wide range of information on specific countries and key development
issues. The first of the three sites just listed contains brief economic overviews (including basic data) on
every developing country. The second is a gateway to documents and information on a large assortment of
development themes and issues. The third is a link to PovertyNet, a World Bank site aimed at providing
resources for people and organizations working to understand and alleviate poverty.

The International Development Research Centre (IDRC)

http :llwww.idrc.ca/library/world/

Sponsored by the Canadian government, the IDRC gateway provides links to a large number of
publications, databases, and development institutions.

Global Macroeconomic and Financial Policy Site

http://stern.nyu.edu/globalmacro/

Maintained by Professor Nouriel Roubini of New York University's Stem School of Business, this site
provides links to newspaper and magazine articles on current issues, as well as academic and government
analyses of exchange rate regimes, financial sector policies, financial crises, and the international fmancial
system. It also provides links to information on specific countries and to selected macroeconomic and
fmancial databases.

Netaid.org
http://www.netaid.org/

Netaid's mission is to use the power of the Internet to create opportunities tend the cycle of extreme
poverty and to provide information on successful development projects and innovative organizations.

The World Factbook

http://www.odci.gov/cia/publications/factbook/index. html

Produced by the CIA, The World Factbook contains maps and information on government structures, key
personnel, the economy, and other basic information on countries around the world.

IMF Directory of Economic, Commodity, and Development Organizations

http://www. imf org/np/seddecdo/contents. htm

The IMF directory provides background information on and links to over 100 regional economic
organizations and intergovernmental commodity and development organizations.

ELDIS: The Electronic Development and Environment Information System

http://www.ids.ac. uk/e!disleldis. html

32
The ELDIS gateway provides a wealth of information on development and the environment. Hosted by the
Institute of Development Studies and the University of Sussex, ELDIS provides links to country-specific
pages, full-text reports and research papers, recent news items, and other information.

Oneworld.net
http://www.oneworld.org/

Oneworld is an international network of cooperative centers with the objectives of promoting human rights
and sustainable development. This site focuses on current news items and key development issues and
includes information from a large number of development organizations.

Box2
The Elasticity of Connection Between Economic Growth and Poverty Reduction

To estimate the elasticity of connection, Timmer regressed the level of income of each
quintile on overall per capita GDP. This "levels" estimation includes country and time
fixed effects (dummy variables for each developing country included and for each decade
from the 1960's to the 1990's). The country fixed effects allow shifts in the regression
intercept for each country, but assume the same slope, or elasticity of connection, for all
countries. The fixed effects for decades allow a shift in the regression intercept for each
10-year decade.

The paper also restricted the sample of countries to those that have a significant
agriculture sector, are reasonably large, and are considered developing countries. For this
reason, countries such as Hong Kong and Singapore were excluded, as were most
countries with populations smaller than 6 million (Costa Rica and Jamaica are the
exceptions to include better representation of Latin America and the Caribbean).

To examine the impacts of inequality on income levels of the poor, Timmer constructed a
variable that measures the relative income gap between the rich and the poor. A dummy
variable was then created that is equal to one when the gap in income between the highest
and lowest quintiles is more than twice as large as average income. Timmer then
disaggregated income into sectoral components from agriculture and non-agriculture in
order to examine whether the sectoral composition of labor productivity matters to the
incomes earned by each quintile.

Earlier results from asking a similar question had already indicated that growth in the
agricultural sector seems to have a much larger impact on growth of incomes in the bottom
quintile than growth in services or industry (Ravallion and Datt, 1996; Gallup, Radelet, and
Warner, 1997; Mellor, 2000). The question here is framed in terms ofrelative labor ·
productivities. Do the per capita labor productivities of workers in agriculture and non-
agriculture have differential effects on the average earnings in each income quintile? Put
another way, do the poor benefit more from growth in the agriculture or the non-
agriculture sector?

33
Timmer found that in unequal countries, that is, where the relative income gap is large,
there is a pronounced Kuznets effect: the elasticity of connection for the poorest quintile
is significantly lower than for the higher quintiles; the poor appear to be nearly
disconnected from the growth process in these economies. The elasticity of connection
for the poorest quintile is 0.257 for agriculture and 0.449 for non-agriculture. In contrast,
for those economies with better income distribution, the elasticity of connection for the
poor in the agriculture sector is 1.146 and 1.018 for non-agriculture. This is slightly
higher than the elasticities for the upper quintiles, suggesting a slight but significant
"anti-Kuznets" effect in these economies. These results are illustrated in Figure 1 below.

Figure 1

Source: Timmer (1997)

Box3
An Empirical Example Comparing Brazil and Thailand

Building on the earlier analytical and empirical work in Tirnmer (1997), it is possible to
use the definitions and returns to capital discussed in the text to construct crude estimates
of the value of human and financial capital assets by income quintiles (for details, see
Gugerty and Timmer, 1999). As a particularly interesting comparison over time and
space, Table 1 shows these asset values by quintile for Thailand and Brazil over a three-
decade period.

34
 Table 1

Changes in income and asset distributions over a three-decade period in Brazil and Thailand

Per capita Per capita Per capita

Quintile income human capital financial capital
($PPP) ($000) ($000)

Thailand, 1962-1992 Annual per capita

1962 1992 quintile growth 1962 1992 1962 1992

I 397 726 2.03% 0.64 7.22 0 0

n 427 1491 1.24 22.52 0 0
III 600 2276 4.70 38.22 0 0
IV 1066 3649 14.02 65.68 0 0
v 2470 11478 5.25% 42.10 120.00 0 102.26
RELGAP 2.089 2.740 3.470 2.431

Brazil, 1960-1989 Annual per capita

1960 1989 quintile growth 1960 1989 1960 1989

I 285 534 2.19% 0 3.38 0 0

II 614 1047 4.98 13.64 0 0
III 970 1965 12.10 32.00 0 0
IV 1700 3909 26.70 70.88 0 0
v 5331 13927 3.37% 99.32 120.00 0 151.24
RELGAP 2.835 3.135 3.306 2.223

Both Brazil and Thailand grew fairly rapidly during this three-decade period, Brazil from
per capita income of$1,780 in 1960 to $4,272 in 1989 (3.06% annual growth), Thailand
from $992 in 1962 to $3,924 in 1992 (4.69% annual growth). Income distribution in both
countries, as measured by RELGAP, worsened, from an average level of2.089 in
Thailand at the start to a highly unequal level of2.740 at the end. 2 In Brazil, inequality
was already very high at the start of the penod, and worsened to a level of 3 .13 5 at the
end, one of the worst distributions of income in the entire Deininger-Squire sample.

Not surprisingly, asset distributions in both countries also changed quite significantly, but
not always in the expected direction. Levels of human capital increased dramatically for
all income classes, but much faster for the poor than for the rich, who were already closer
to the plateau levels used in this analysis. Increases of two to three times were the norm
in Brazil; full order of magnitude increases occurred in Thailand. Even as income
distribution worsened, the distribution of human capital became more equal, as the poor
were finally included in the growth process to some extent.

2
RELGAP is defined as the average per capita income in the top quintile minus the average per capita
income in the bottom quintile, divided by the average per capita income for the society. When RELGAP is
greater than two, economies have a difficult time sustaining growth and connecting the poor to it (Timmer,
1997).

35
This result is supported by empirical research that indicates that changes in primary
school enrollment are strongly positively associated with growth in lower income
countries (Easterly, 1997; de Gregorio and Lee, 1998). Because of the self-limiting
nature of human capital accumulation, however, this dimension of asset distribution is
also limited in its potential contribution to future ~arnings. In addition, given the
constraints on investment in human capital by the poor suggested by the theoretic
literature, it is clear that government policy will play an important role in human capital
accumulation at lower levels of development.

The open-ended nature of financial assets avoids the ceilings inherent in accumulation of
human capital. If the distribution of financial assets is or becomes highly skewed during
the growth process, at some point the income-earning potential of these assets will lead to
a self-reinforcing skewing of incomes. Both Brazil and Thailand seem to have reached
such a point by around 1990. In Brazil, a simple dynamic calculation shows that if all
incomes above the human capital level of $6365 are saved and invested in financial assets
that earn the assumed five percent per year, within a decade the upper quintile of income
earners will receive the entire additional income generated by an economy growing at
five percent per capita per year.

At some point in the development process, concentration of ownership of financial assets

can lead to sharply skewed income distribution as an inevitable result of economic
growth, a result that is not typically seen in the early stages of growth when the
dependence on investments in human capital are far more important for the distribution
of income and for lifting the poor out of absolute poverty (Ravalli on, 1996).

Box4
HIVIAIDS and Capacity Building in Africa

Dear Mr. Natsios:

Thank you for taking the time to talk to me after the panel discussion at the Council for
Foreign Relations supper at Harvard on Monday evening. Your comments touched on
two issues that relate to my experience in Africa. The first is the potential for HIV/AIDS
to contribute to state failure. The second is the problem of mass depression. (Rwanda
was the example you gave.) The first point becomes increasingly clearer to people who
work within key organizations in Southern Africa (central banks, ministries of finance).
These.organizations are often hollow shells operating with depleted human capacity and
little hope under current circumstances of that capacity being replenished. The second
point also comes from working in these organizations (and living and traveling in
Southern and East Africa). A little noted effect of the HIVIAIDS epidemic has been
profound sense of sorrow and gloom that pervades these organizations in particular, and
societies more generally. The human spirit may be resilient but the toll of death and
debility over the last decade has been extreme. Unfortunately, as we are all aware, the

36
immediate future is unlikely to show any improvement, even if the broad-based provision
of anti-retroviral drugs were feasible, as some people noisily argue.

I have been dealing with the consequences of HIVIAIDS in my work in Africa for almost
a decade. Over that period, several questions have bothered me. My work has been
primarily within ministries of finance and central banks doing two things: helping
countries create the conditions for sustained economic reform; and providing training to
build the capacity for these organizations to continue the reforms. HIVI AIDS has cut
across this effort in fundamental ways. The most obvious is the loss of skilled personnel
and labor time. Less obvious has been the impact of behavioral changes that occur as
growing numbers of people recognize (or suspect) that their productive life spans are
being dramatically truncated. The questions I have been trying to address relate to the
economic effects (at both the micro and macro levels) that occur when decision horizons
are prematurely foreshortened. There are many dimensions involved. For example, why
should people support economic reforms (some of which have harsh immediate impacts)
in the hope that conditions will improve at some future date that many of them will not be
around to witness or benefit from? Similarly, why should individuals who already have
some skills (e.g., a BA) forego income and leisure in order to deepen their capacities if
the prospects of enjoying the higher income and status associated with those skills has
sharply diminished? This point can be flipped around to ask the same question about
opportunism. Why should any one who knows their time is limited refrain from
opportunism (goofing off, prolonged absenteeism) or even overtly criminal behavior?
What punishment can an organization or society prescribe that is worse than the fate they
expect? These examples can be readily extended. Indeed, because so much of economic
behavior is based on expectations of the future and the anticipated flows of benefits and
costs, many human activities - investment, migration, savings, education, portfolio
allocation, trade and exchange, cooperation, to name a few - are covered. Curiously,
researchers who model the impact of HIVI AIDS on economic growth (whether they
make the connection with capacity deepening or not) do not draw on these economic
principles. The standard projection models (ING Barings, Abt Associates, Arndt/Lewis
at the World Bank, and even a recent paper from the IMF) are relatively mechanistic.
They typically involve computable general equilibrium or neoclassical growth models
upon which most of the behavioral consequences are imposed to produce scenarios that
compare situations "with" and "without" AIDS. Not surprisingly, the 'with AIDS'
growth paths are lower than the 'without AIDS' growth paths. Ultimately, however, the
epidemic is assumed to work its way through the system and growth resumes.

All of this is far too clinical and from my experience in Southern Africa, beside the point.
It is not clear what are the policy implications. For example, what can be done to change
the incentives that are driving the economic behavior that results in lower growth?
Furthermore, I am completely at a loss to understand what a 'without AIDS' scenario can
mean in countries such as Botswana, South Africa, Zambia, and Zimbabwe. These
countries have had their growth paths fundamentally altered precisely because the spread
of HIV/AIDS has changed the expectations and behavior of most, if not all, members of
the respective populations.

37
For these reasons, I have urged that USAID should be encouraging groups of African
researchers to study in more detail the behavioral dimensions related to the impact of
HIV/AIDS on economic growth. Such study, I believe, would highlight the need for a
radical rethinking of the way we train, manage, motivate, discipline, and otherwise deal
with the growing number of people (in all skill categories) whose productive lives are
being drastically shortened. The problems of capacity deepening have especially
concerned me. In this regard, how do we cut across the adverse expectations that leads
from declining skills (and the diminished incentive to invest in skill acquisition) due to
HIVIAIDS and economic growth. History has shown that a key element in sustained
economic growth is improved productivity. As the spread of HIV/AIDS erodes human
capacities, a major question that needs to be addressed is what modifications can we
make in training activities, work flow, organizational structure, and management
procedures to compensate for these losses? To illustrate the relevance of this point, much
has been made of the importance of extension in raising agricultural output and
improving food security. But, what do we extend to children and grandparents who now
head many agricultural households when the parents (the object of standard extension
programs) have died or become debilitated? More important, since the education levels
of the children and grandparents are generally lower than their parents, how do we extend
whatever research may prove to be useful or productive or profitable? Questions such as
these have been at the core of my work on tracing the effects of HIV/AIDS on sustained
economic growth. My conviction is that unless we can determine ways of maintaining
and even deepening capacity in the face of the HIV/AIDS epidemic, African countries
cannot grow and develop.

USAID has devoted so much effort and resources to HIV prevention, social marketing,
strengthening of health systems and related activities. This effort would be greatly
strengthened if attention were also given to the questions of how capacities can be
deepened; how HIV -positive but still productive workers can be managed, motivated, and
induced to add to national income rather than subtract from it; and how overall
productivity and growth can be maintained in the face of mounting losses due to
HIVIAIDS. The Agency has appropriately placed heavy emphasis on the health related
aspects of HIV/AIDS. But, some effort also should be given to finding ways of reducing
the negative effects of HIV/AIDS on economic growth.

I would be happy to expand further on these ideas if you think it useful. Thank you for
your interest and help. Kind regards.

Malcolm McPherson
Senior Fellow in Development
John F. Kennedy School of Government
Harvard University

38
 References

Abramovitz, Moses. 1989. Thinking About Growth. Cambridge, UK: Cambridge University Press.

Acemoglu, Daron, Simon Johnson and James A. Robinson. 2001. The Colonial Origins of Comparative
Development: An Empirical Investigation. American Economic Review, Vol. 91, no. 5
(December), pp. 1369-1401.

Alesina, A. and David Dollar. 2000. "Who Gives Foreign Aid to Whom and Why? Journal ofEconomic
Growth. Vol. 5, no 1 (March), pp. 33-63.

Alesina, A., and R. Perotti. 1993. "Income Distribution, Political Instability, and Investment."
Cambridge, MA: National Bureau of Economic Research (NBER) Working Paper no. 4486,
October.

Alesina, Alberto and Dani Rodrik, 1994. "Distributive Politics and Economic Growth." The Quarterly
Journal ofEconomics, May 1994.

Anand, Sudhir, and S. M. R. Kanbur. 1993. "The Kuznets Process and the Inequality-Development
Relationship." Journal of Development Economics. vol. 40, no. 1, pp. 25-52.

Anderson, K., and Y. Hayami, with associates. 1986. The Political Economy ofAgricultural Protection:
East Asia in International Perspective. London: Allen & Unwin.

Barro, R.J. 1997. "Determinants of Economic Growth: A Cross-Country Empirical Study,'' Development
Discussion Paper No. 579, Harvard Institute for International Development, April, 116 pages.

Barro, R.J., and J.-H. Lee. 1994. "Sources of Economic Growth,'' Carnegie-Rochester Conference Series
on Public Policy. Amsterdam: North-Holland/Elsevier, pp. 1-46.

Barro, R.J., and X. Sala-i-Martin. 1995. Economic Growth. New York: McGraw-Hill.

Baumol, William J. 1986. "Productivity Growth, Convergence, and Welfare: What the Long-Run Data
Show," American Economic Review, Vol. 76 (December), pp. 1072-85.

Birdsall, N., D. Ross, and R. Sabot. 1995. "Inequality and Growth Reconsidered: Lessons from East
Asia." World Bank Economic Review. Vol. 9, no. 3, pp. 477-508.

Birdsall, Nancy, and Juan Luis Londano, 1997. "Asset Inequality Matters: An Assessment of the World
Bank's Approach to Poverty Reduction." American Economic Review, Vol 97, No. 2.

Bliss, C., and N. Stem. 1978. "Productivity, Wages and Nutrition: Parts I and II." Journal of
Development Economics. Vol. 5, no. 4, pp. 331-98.

Chai, C.P. 1995. Rural Human Capital and Economic Growth in Developing Countries. Senior Honors
Thesis. Cambridge, MA: Department of Economics, Harvard University.

Clark, C. 1940. The Conditions ofEconomic Progress (3rd Edition, 1957). London: Macmillan and Co.

Cole, W. A., and Phyllis Deane, 1965. "The Growth of National Incomes,'' in Habakkuk and Postan, eds.,
The Cambridge Economic History ofEurope, vol. 6: The Industrial Revolution and After (Part I),
Cambridge, UK: Cambridge University Press, pp 1-55.

Collier, Paul. 2001. "Conflict and Development." Development Research Group, The World Bank,
November.

39
Collier, Paul. 2002. "Making Aid Smart: Institutional Incentives facing Donor Organizations and their
Implications for Aid Effectiveness." Forum Series on the Role of Institutions in Promoting
Economic Growth. The IRIS Center, University of Maryland. February.

Collier, Paul, and Jan Denn. 2001. "Aid, Shocks and Growth." Development Research Group, Policy
Research Working Paper no. 2688. The World Bank. October.

Collier, Paul, and David Dollar. 2001. "Can the World Cut Poverty in Half? How Policy Reform and
Effective Aid Can Meet International Development Goals."
World Development, Vol 29, no 11, pp. 1787-1802.

Crosswell, Michael, 1999. "The Development Record and the Effectiveness of Foreign Aid," Praxis: The
Fletcher Journal of Development Studies, Vol. XV.

Crosswell, Michael, 2001. "Aid to the Least Developed Countries: What do the Data Tell Us?
USAID/PPC, December.

Dalrymple, Dana, 2001. "Global Public Goods: Some Outline Notes and References." Office of
Agriculture and Food Security, Global Bureau, USAID, Washington, DC. March.

Dasgupta, P. 1993. An Inquiry into Well-Being and Destitution. Oxford: Clarendon Press.

Datt, Guarav, and Martin Ravallion. 1997. "Why Have Some Indian States Done Better Than Others at
Reducing Rural Poverty?" Economica, vol. 64.

Davis, Donald R., and David Weinstein. 2001. "What Role for Empirics in International Trade?" NBER
Working Paper no. 8543. Cambridge, MA. October.

Dawe, D. 1993. The Effects ofExport Instability on Investment: The Information Content ofPrices and
Growth. Ph.D. Dissertation. Cambridge, MA: Economics Department, Harvard University.

Dawe, D. 1996. "A New Look at the Effects of Export Instability on Investment and Growth." World
Development. December.

Dawe, David. 2001. "Long-Run Changes in the World Rice Market," International Rice Research
Institute, Los Banos, the Philippines. Processed.

De Gregorio, Jose, and Jong-Wha Lee. 1998. "Education and Income Distribution: New Evidence from
Cross-Country Data." Department of Economics, Korea University, Seoul, Korea. July.

Deininger, Klaus, and Lyn Squire. 1998. "New Ways of Looking at Old Issues: Inequality and Growth."
Journal ofDevelopment Economics, Vol 57.

Deininger, Klaus, and Lyn Squire. 1996. "A New Data Set Measuring Income Inequality." The World Bank
Economic Review. vol. 10, no. 3 (September), pp. 565-591. Data available at
http://www.worldbank.org/html/prdmg/grwthweb/growth_t.htm
'
de Janvry, A., and E. Sadoulet. 1987. "Agricultural Price Policy in General Equilibrium Models: Results
and Comparisons." American Journal ofAgricultural Economics. May, pp. 230-246.

De Janvry, A., and E. Sadoulet. 2000. "Rural Poverty in Latin America: Determinants and Exit Paths."
Food Policy. Vol. 25, pp. 3 89-409.

40
De Janvry, A., and E. Sadoulet. 2001. "Concepts for an Approach to Rural Development in Mexico and
Central America: Regional Development and Economic Inclusion." Prepared for the IDB
regional workshop, "Desarrollando la economia rural de Puebla a Panama," Guatemala City,
March5-7.

Denison, Edward F., 1974. Accounting/or United States Economic Growth, 1929-1969. Washington, DC:
Brookings Institution.

Diamond, Jared. 1997. Guns, Germs and Steel: A Short History ofEverybody for the Last 13,000 Years.
Vintage Books, Random House, London.

Diamond, Larry. 2002. "Democracy and Governance." Draft Chapter for the ''Natsios Report," USAID,
Washington, DC. February.

Dorfman, Robert, Paul A. Samuelson, and Robert Solow. 1958. Linear Programming and Economic
Analysis. New York, NY: McGraw-Hill.

Easterly, William, 1997. "Life During Growth" Mimeo, World Banlc Development Research Group.

Easterly, William. 2001. The Elusive Quest/or Economic Growth. Cambridge, MA: MIT Press.

Easterly, William. 2002. "Inequality does Cause Underdevelopment: New Evidence." Working Paper no.
I. Center for Global Development, Institute for International Economics. Washington, DC.
January.

Feldstein, M., and C. Horioka. 1980. "Domestic Saving and International Capital Flows." Economic
Journal. Vol. 90 (June), pp. 314-29.

Fogel, R.W. 1991. "The Conquest of High Mortality and Hunger in Europe and America: Timing and
Mechanisms." In P. Higonnet, D.S. Landes, and H. Rosovsky, eds., Favorites ofFortune:
Technology, Growth, and Economic Development since the Industrial Revolution. Cambridge,
MA: Harvard University Press, pp. 35-71.

Fogel, RW. 1994. "Economic Growth, Population Theory, and Physiology: The Bearing of Long-Term
Processes on the Making of Economic Policy." {Nobel Prize Lecture] American Economic
Review. Vol. 84, no. 3 (June), pp. 369-395.

Fox, James W. 1997. "What do Economists Know that Policymakers need to? American Economic
Review, Vol. 87, no. 2 (May), pp.49-51.

Fox, James W. 2002. "Development Overview." Draft Chapter for "Natsios Report." USAID,
Washington, DC. January.

Gallup, John, Steven Radelet, and Andrew Warner. 1997. "Economic Growth and the Income of the Poor."
Harvard Institute for International Development. Prepared for the CAER II Project, August, mimeo.

Gerschenkron, Alexander. 1962. Economic Backwardness in Historical Perspective. Cambridge, MA:

Harvard University Press.

Grindle, Merrilee S., and John W. Thomas. 1991. Public Choices and Policy Change: The Political
Economy ofReform in Developing Countries. Baltimore, MD: The Johns Hopkins University Press.

Gugerty, Mary Kay, and C. Peter Timmer. 1999. "Growth, Inequality and Poverty Alleviation: Implications
for Development Assistance." Prepared for CAER II, USAID, Washington, DC. November.

41
Haddad, L, M. T Ruel and J. L. Garrett. 1999. "Are Urban Poverty and Undemutrition Growing? Some
Newly Assemble Evidence." Food Policy. Vol. 27, no. 1 (November)., pp. 1891-1904.

Hannon, Brian, and Yung Whee Rhee. 2002. Draft Chapter on Globalization, Trade, and Investment for
"Natsios Report." Washington, DC: USAID.

Hayami, Y., and V. Ruttan. 1985. Agricultural Development: An International Perspective. Revised and
expanded edition. Baltimore: Johns Hopkins University Press.

Islam, N. and S. Thomas. 1996. Foodgrain Price Stabilization in Developing Countries: Issues and
Experiences in Asia. Food Policy Review no. 3. Washington, D.C.: International Food Policy
Research Institute (IFPRI).

Jamison, D., and L. Lau. 1982. Farmer Education and Agricultural Productivity. Baltimore: Johns
Hopkins University Press.

Johnson, D.G. 1997. "Agriculture and the Wealth of Nations (Ely Lecture)," American Economic Review.
Vol. 87, no. 2 (May), pp. 1-12.

Johnston, B.F., and J.W. Mellor. 1961. "The Role of Agriculture in Economic Development." American
Economic Review. Vol. 51, no. 4, pp. 566-93.

Jones, E. L. 1981. The European Miracle: Environments, Economics and Geopolitics in the History of
Europe and Asia. Cambridge, UK: Cambridge University Press.

Jones, E. L. 1988. Growth Recurring: Economic Change in World History. Oxford, UK: Clarendon Press.

Knack, Stephen, (1999). "Social capital, growth and poverty: A Survey of the cross-country evidence."
World Bank, Social Capital Initiative, Working Paper #7.

Kuznets, S. 1955. "Economic Growth and Income Inequality." American Economic Review. Vol. 49, no.
1, pp. 1-28.

Kuznets, S. 1966. Modern Economic Growth. New Haven: Yale University Press.

Landes, David. 1969. The Unbound Promethus: Technological Change and Industrial Development in
Western Europe from 1750 to the Present. Cambridge, UK: Cambridge University Press.

Landes, David. 1990. "Why Are We So Rich and They So Poor?" American Economic Review. Vol 80,
no. 2 (May), pp. 1-13.

Landes, David. 1998. The Wealth and Poverty ofNations: Why Some Are So Rich and Some So Poor.
New York, NY: W.W. Norton and Co.

Larson, D., and Y. Mundlak. 1997. "On the lntersectoral Migration of Agricultural Labor," Economic
Development and Cultural Change, pp. 295-319.

Lewis, W.A. 1954. "Economic Development with Unlimited Supplies of Labor." The Manchester School.
Vol. 22, pp. 3-42.

Lipton, M. 1977. Why Poor People Stay Poor: Urban Bias in World Development. Cambridge, MA:
Harvard University Press.

Lipton, M. 1993. "Urban Bias: Of Consequences, Classes and Causality." In A. Varshney, ed., Beyond
Urban Bias. London: Frank Cass, pp. 229-58.

42
Lucas, R.E. 1973. "Some International Evidence on Output-Inflation Tradeoffs." American Economic
Review. Vol. 63, pp. 326-34.

Lucas, R. E. 1988. "On the Mechanics of Economic Development." Journal ofMonetary Economics.
Vol. 22, no. 1 (July), pp. 3-42.

Mankiw, N.G., D. Romer, and D.N. Weil. 1992. "A Contribution to the Empirics of Economic Growth."
Quarterly Journal ofEconomics. Vol. 107, no. 2 (May), pp. 407-437.

Martin, W., and D. Mitra. 1996. "Productivity Growth in Agriculture and Manufacturing," International
Trade Division, The World Bank, preliminary draft (September), 30 pages.

Mellor, John W. 1976. The New Economic of Growth: A Strategy for India and the Developing World
Ithaca, NY: Cornell University Press.

Mellor, John W. 2000. "Agricultural Growth, Rural Employment, and Poverty Reduction: Non-Tradables,
Public Expenditure, and Balanced Growth." Prepared for the World Bank Rural Week 2000, March.

Meltzer, Allan H. 2000. Statement by Allan H. Meltzer on the Report of the International Financial
Institution Advisory Commission. Senate Committee on Banking, Housing, and Urban Affairs.
March 9.

Morduch, Jonathan. 1994. "Poverty and Vulnerability," American Economic Review: Papers and
Proceedings, vol. 84, no. 2 (May), pp. 221-5.

Morduch, Jonathan. 1999. "The Microfinance Promise," Journal ofEconomic Literature. Vol. 37, no. 4,
pp. 1569-1614.

Mosher, A.T. 1966. Getting Agriculture Moving: Essentials for Development and Modernization. New
York: Praeger.

Mundlak, Yair. 2000. Agriculture and Economic Growth: Theory and Measurement. Cambridge, MA:
Harvard University Press.

Mundlak, Yair. 2001. "Explaining Economic Growth," American Journal ofAgricultural Economics.
Vol. 83, no. 5, pp. 1154-67.

Narayan, Deepa, and Lant Pritchett, 1999. "Cents and sociability: household income and social capital in
rural Tanzania." Economic Development and Cultural Change. Volume 57.

Naylor, Rosamund, and Walter P. Falcon. 1995. "Is the Locus of Poverty Changing?" Food Policy, Vol 20,
no. 6, pp. 501-518.

Nelson, Richard R 1993. Nationa/1nnovation Systems: A Comparative Analysis. New York, NY: Oxford
University Press.

North, Douglass C. 1992. Transaction Costs, Institutions, and Economic Performance. San Francisco, CA:
ICS Press.

Olson, Mancur. 2000. Power and Prosperity: Outgrowing Communist and Capitalist Dictatorships. New
York: Basic Books.

Painter, David. 2001. "Urban Poverty and the Role ofUSAID." Washington, DC: USAID, processed.

Prescott, E. D. 1998. "Needed: A Theory of Total Factor Productivity." International Economics Review.
Vol. 39 (August), pp. 525-51.

43
Ravallion, M. 1996. "Can High-Inequality Developing Countries Escape Absolute Poverty?" Policy
Research Department, Mimeo, World Bank, Washington, D.C., mimeo.

Ravallion, M. 2001. "Growth, Inequality and Poverty: Looking Beyond the Averages," World
Development, Vol. 29, no. 11, pp. 1803-15.

Ravallion, M., and G. Datt. 1996. "How Important to India's Poor Is the Sectoral Composition of
Economic Growth?" The World Bank Economic Review. Vol. 10, no. 1, pp. 1-25.

Reynolds, L.G., ed. 1975. Agriculture in Development Theory. New Haven: Yale University Press.

Rock, Michael T. 2001. "Exploring the Impact of Selective Interventions in Agriculture on the Growth of
Manufactures in Indonesia, Malaysia, and Thailand," Journal ofInternational Development
(forthcoming).

Romer, Paul M. 1986. "Increasing Returns and Long-Run Growth," Journal ofPolitical Economy. Vol.
94 (October), pp. 1002-37.

Romer, Paul M. 1990. "Endogenous Technological Change," Journal ofPolitical Economy, Vol. 98
(October), pp. S71-Sl02 .

. Rostow, Walt W. 1990. Theorists of Economic Growth from David Hume to the Present. New York, NY:
Oxford University Press.

Sachs, J. 1997. "Nature, NUrture, and Growth," Economist, June 14, pp. 19-27.

Sarris, Alexander H. 2001. "The Role of Agriculture in Economic Development and Poverty Reduction:
An Empirical and Conceptual Foundation." Prepared for the Rural Development Department of
the World Bank, Washington, DC. March.

Schultz, T.W., 1964. Transforming Traditional Agriculture. New Haven: Yale University Press.

Solow, Robert M. 1956. "A Contribution to the Theory of Economic Growth." Quarterly Journal of
Economics. Vol. 70 (February), pp. 65-94.

Srinivasan, T.N. 1985. "Neoclassical Political Economy, the State, and Economic Development." Asian
Development Review, Vol. 1, no. 1, pp. 38-58.

Strauss, J. 1986. "Does Better Nutrition Raise Farm Productivity?" Journal ofPolitical Economy. Vol.
94, no. 2, pp. 297-320.

Strauss, J., and D. Thomas. 1998. "Health, Nutrition, and Economic Development." Journal of Economic
Literature. Vol. XXXVI, no. 2 (June), pp. 766-816.

Streeten, Paul. 1986. First Things First: Meeting Basic Human Needs. Baltimore, MD: Johns Hopkins
University Press for the World Bank.

Summers, R., and A. Heston. 1991 (updated 1996). "The Penn World Table (Mark 5): An Expanded Set
of International Comparisons, 1950-1988." Quarterly Journal ofEconomics. Vol. 106, no. 2
(May), pp. 327-68.

Tabor, Steven R. 2002 "Indonesian Rice Policy and the Choice of Trade Regimes." Prepared for the
BULOG seminar on alternative options for rice trade policy. Jakarta, Indonesia. March.

44
Teranishi, J. 1997. Presentation at World Bank seminar on research project documenting importance of
investment in agricultural infrastructure.

Timmer, C.P. 1988. "The Agricultural Transformation." In H. Chenery and T.N. Srinivasan, eds.,
Handbook ofDevelopment Economics. Vol. 1. Amsterdam: North-Holland, pp. 275-331.

Timmer, C.P. 1989. "Food Price Policy: The Rationale for Government Intervention." Food Policy. Vol.
14, no. 1 (February), pp. 17-42.

Timmer, C.P. 1993. "Rural Bias in the East and Southeast Asian Rice Economy: Indonesia in
Comparative Perspective." In A. Varshney, ed., Beyond Urban Bias. London: Frank Cass, pp.
149-76.

Timmer, C. P. 1997. "How Well do the Poor Connect to the Growth Process?" Harvard Institute for
International Development, -mimeo.

Timmer, C. P. forthcoming. "Agriculture and Economic Growth." In Bruce Gardner and Gordon Rausser,
eds., The Handbook ofAgricultural Economics, Vol. II. Amsterdam: North-Holland.

Timmer, C.P., W.P. Falcon, and S.R. Pearson. 1983. Food Policy Analysis. Baltimore, MD: Johns
Hopkins University Press for the World Bank.

United States Department of Agriculture (USDA). 2001. Food and Agricultural Policy: Taking Stock for
the New Century. Washington, DC: USDA. September.

United States Agency for International Development (USAID). 2002. Future Directions for Agriculture:
A Background Paper (draft). Bureau for Economic Growth, Agriculture and Trade (EGAT),
Washington, DC: USAID, February.

Vamvakidis, Athanasios. 1997. "GATT was Right: Historical Evidence on the Growth-Openness
Connection and Discriminatory versus Nondiscriminatory Liberalization." Economics
Department, Harvard University, Cambridge, MA: processed.

Vincent, Jeffrey and Karl-Goran Mahler, eds. forthcoming. Handbook of Environmental Economics.
Amsterdam: North-Holland.

Wade, Robert, 2001. "The Rising Inequality of World Income Distribution." Finance and Development.
December, pp. 37-39.

Woolcock, Michael. 1998. "Social Capital and Economic Development: Toward a Theoretical Synthesis
and Policy Framework." Theory and Society. Vol 27, no. 2, pp. 151-208.

Woolcock, Michael, 1999. "Managing Risk, Shocks, and Opportunity in Developing Economies: The Role of
Social Capital." Mimeo, World Bank Development Research Group.

World Bank. 2000. The Quality of Economic Growth. Oxford and New York: Oxford University Press for
the World Bank.

World Bank, 2002. World Development Report. Oxford and New York: Oxford University Press for the
World Bank.

45