Structural adjustment and agriculture

Material Information

Structural adjustment and agriculture a comparative perspective on performance in Africa, Asia, and Latin America
Series Title:
International working paper series
Lele, Uma J
Seminar on Food and Agricultural Policies Under Structural Adjustment, (1992
Place of Publication:
Gainesville Fla
Food and Resource Economics Dept., Institute of Food and Agricultural Sciences, University of Florida
Publication Date:
Physical Description:
39, [18] p. : ill. ; 28 cm.


Subjects / Keywords:
Structural adjustment (Economic policy) -- Africa ( lcsh )
Structural adjustment (Economic policy) -- Asia ( lcsh )
Structural adjustment (Economic policy) -- Latin America ( lcsh )
Agriculture -- Economic aspects -- Africa ( lcsh )
Agriculture -- Economic aspects -- Asia ( lcsh )
Agriculture -- Economic aspects -- Latin America ( lcsh )
non-fiction ( marcgt )


Includes bibliographical references.
General Note:
"November 1992"--Cover.
General Note:
"Keynote address to the First Plenary Session of the European Association of Agricultural Economists' Seminar on Food and Agricultural Policies Under Structural Adjustment, September 21-25, 1992, Hohenheim, Germany."
Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.
Statement of Responsibility:
Uma Lele.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact Digital Services ( with any additional information they can provide.
Resource Identifier:
001784450 ( ALEPH )
27377053 ( OCLC )
AJK7832 ( NOTIS )

Full Text
7. /
Uma Lele
IW92-30 November 1992
Institute of Food and Agricultural Sciences
University of Florida
Gainesville, Florida 32611

Uma Lele2,3
'Keynote address to the First Plenary Session of the European Association of Agricultural Economists' Seminar on Food and Agricultural Policies under Structural Adjustment, September 21-25, 1992, Hohenheim, Germany.
2Graduate Research Professor, Food and Resource Economics Department, and Director of International Studies and Programs, University of Florida,
Gainesville, Florida, 32611, Tel. (904)392-3246 or 5068, Fax (904)392-5575 or 392-2395.
3Support for the preparation of this paper was provided by the Rockefeller Foundation and the International Monetary Fund as part of a larger study of Structural Adjustment and Agriculture. For detailed empirical evidence in support of many of the arguments made in this paper see Uma Lele, Kofi Adu-Nyako, and Robert Emerson, "Structural Adjustment and Agriculture", forthcoming. I alone am responsible for the views expressed in this paper and the deficiencies that remain. They should not be attributed to the supporting institutions. I am grateful to Kofi Adu-Nyako for his comments on the earlier draft and to Dieudonne Mann and Rahul Jain for research assistance.

Table of Contents
The Fallacy of Composition Argument (Again?)
Price Stabilization Fertilizer Subsidies
Employment and Income Effects
Agricultural Credit
Land Policy

Structural Adjustment and Agriculture: A Comparative Perspective
on Performance in Africa, Asia, and Latin Americal
Uma Lele2,3
In this paper I argue that structural adjustment has diverted attention from
agriculture. The importance of increasing factor productivity in agriculture
in resuming rapid economic growth, alleviating poverty, increasing women's
participation, and saving the environment is being overlooked. These latter
issues are currently at the forefront and agriculture has become their
unexpected victim. This is an ironic outcome and certainly not the one which
supporters of these various good things (myself included) would wish. After
'Keynote address to the First Plenary Session of the European Association of Agricultural Economists' Seminar on Food and Agricultural Policies under Structural Adjustment, September 21-25, 1992, Hohenheim, Germany.
2Graduate Research Professor, Food and Resource Economics Department, and Director of International Studies and Programs, University of Florida,
Gainesville, Florida, 32611, Tel. (904)392-3246 or 5068, Fax (904)392-5575 or 392-2395.
3Support for the preparation of this paper was provided by the Rockefeller Foundation and the International Monetary Fund as part of a larger study of Structural Adjustment and Agriculture. For detailed empirical evidence in support of many of the arguments made in this paper see Uma Lele, Kofi Adu Nyako, and Robert Emerson, "Structural Adjustment and Agriculture", forthcoming. I alone am responsible for the views expressed in this paper and the deficiencies that remain. They should not be attributed to the supporting institutions. I am grateful to Kofi Adu Nyako for his comments on the earlier draft and to Dieudonne Mann and Rahul Jain for research assistance.

all, structural adjustment was and continues to be directed to (among other things) boosting exports, and agriculture is still an important, albeit declining, source of exports in many developing countries (Figure 1). Through devaluation and elimination of taxes adjustment raises internal terms of trade in favor of agriculture and such intersectoral redistribution of income provides a boost to food and agricultural production. Moreover, the positive impact of a stable macroeconomic environment for healthy growth is well established.
Yet, whereas price adjustments are crucial, partial price
adjustments can harm the process of broadbased economic growth. Moreover, price reforms are by no means sufficient. Long term supply response from agriculture calls for emphasis on nonprice factors. Levels and patterns of investments in irrigation, agricultural research, extension, transport, communications, and human capital are critical in increasing agricultural production. Investments especially with a focus on women and children in health, education, and village water supply determine the size and quality of the population and their access to technologies. They in turn determine whether labor remains in agriculture or migrates to other sectors. Such investments are more important at an early stage of development in which agriculture dominates in GNP, exports, employment, government revenues, food supply, and industrial raw materials, and when agriculture is the major source of demand for goods and services produced in the manufacturing sector. But non-price factors are by no means unimportant at later stages. The share of employment in agriculture still dominates over that in manufacturing in
4'See Mohsin Khan, "The Macroeconomic Effects of Fund-Supported Adjustment Programs," in IMF Staff Paners, Vol. 37, No. 2, June 1990 (pp. 195-231).

countries such as Indonesia and China, and a small percentage growth in agricultural employment is equivalent to a large percentage growth in manufacturing employment. Besides, agricultural growth tends to have strong multiplier effects on the manufacturing and service sectors.
Long run development considerations are increasingly being
incorporated in adjustment operations by both the IMF and the World Bank.5 But the focus of these two institutions is on the short term. The intellectual effort they devote to analysis is on middle income countries and sectors other than agriculture. What little emphasis is given to agriculture focuses on Africa. The lessons from the successful middle income East Asian countries in industrial and financial liberalization are cited frequently, but the important historical experience in their agricultural development is rarely explored. The lessons of their experience are of interest not simply to the small open economies of Africa, but to South Asia and the low end poverty groups in Latin America as well.
That experience stresses the nature of externalities and the
importance of public investments in rural development and an intelligent and enlightened role for the government. But public investments are declining in developing countries and with them those in agriculture. And for quite
understandable reasons governments have become discredited as parasites and predators. Yet without increased and improved government action, the extensive failure in rural factor and product markets, which adjustment aims to improve, will not occur. Moreover the limits of markets in achieving the
5ne IMF, for instance, established the Structural Adjustment Facility (SAF) in 1986 and followed it up with the Extended Structural Adjustment Facility.
6The four so-called Asian Tigers to the list of which must now be added China and Thailand.

many development tasks will not be recognized and acted upon.
To stress these contradictions I first provide the context to my paper by summarizing the findings on agricultural supply response, then based on the evidence of 53 adjusting countries establish agriculture's performance in the three continents before and after adjustment, i.e.*, the changing patterns of international trade, growth in Agricultural value added, growth of agricultural exports and imports, and evaluate it again in the context of the recent literature on response of agriculture specifically to adjustment. I then document the waning attention to agriculture, explore the reasons why, and identify some of the issues pertinent to agriculture in the context of adjustment. Then I summarize and conclude.
Abstracting from the measurement issues discussed in detail by Binswanger, econometric evidence indicates that overall supply response in agriculture tends to be smaller than that of individual crops which (as he correctly observes) economists often mistakenly cite to illustrate the likely impact of policy reforms on agriculture. 7 The export crop sector tends to show a greater long-run response than the food crop sector and annual cash crops tend
to be more responsive than tree crops.8 Long run aggregate responses are much greater than short term responses and nonprice factors are more important in the long run response than price factors, although price factors are not
7See Hans P. Binswanger, "The Policy Response of Agriculture," in Proceedings of the World Bank Annual Conference on Development Economics. 1989
(Washington, DC: World Bank, 1989.
8See Raj Krishna, "Some Aspects of Agricultural Growth Price Policy and Equity in Developing Countries," in Food Research Institute Studies (U.S.), Vol. 18, No. 3, 1986 (pp.219-60).

unimportant. However, recent research has shown that indirect effects of macroeconomic policies (exchange rate, trade, and tax policies) have been more
significant than direct effects.9 The literature also argues that countries with weak representation of rural interests are more prone to tax agriculture heavily than those with a strong representation of rural interests.10 Indeed, subsidies on food, fertilizers, irrigation, and water amount to well over $1 billion annually each in as diverse a set of countries as India, Indonesia, Mexico, Brazil, and Nigeria. These subsidies constitute a large share of overall budget deficits and their reduction constitutes an important element of reforms under adjustment. It is noteworthy from Table 1, however, that whereas taxation of agriculture has been the highest in Africa, it has by no means been insignificant in other countries, many of which have been good performers in agriculture. Taxation of agriculture has diminished significantly since the adjustment process began, most notably in Africa. Indeed, the overall rate of taxation of agriculture may now be comparable in Asia to that in Africa. But both the quantity and quality of government expenditures has been a much more significant factor in the development of agriculture in Asia. Yet how fast, how much, and how should nonagriculture contribute to the development of agriculture as distinct from agriculture's contribution to the rest of the economy remains an inadequately understood issue. In this regard, Taiwan's experience in modernizing agriculture and
OSee Anne 0. Krueger, Maurice W. Schiff, and Alberto Valdez, The Political Economy of Agricultural Pricing Policy (Baltimore: The Johns Hopkins University Press for the World Bank, 1991).
'0See Robert H. Bates, Markets and States in Tropical Africa: The Political Basis of Agricultural Policies (Berkeley: University of California Press, 1981).

agriculture's contribution to its industrialization provide some important insights to contemporary developing countries. Taiwan's average farm size of 1 ha or less is also more in tune with low income countries' agriculture than the frequently cited example of Argentina which enjoys large scale agriculture and well developed factor and product markets.1" Moreover, unlike South Korea (which is a favorite on industrial and financial liberalization) Taiwan did not protect its agriculture, nor was it a large recipient of external capital. On the contrary Taiwan exported capital to its nonagricultural sector, to Japan and more recently to the international capital markets. Taiwan has had neither a Ministry of Planning nor one for Agriculture, and until recently had an undemocratic government. How did Taiwan do it?
In an intersectoral analysis of growth accounting covering a 50 year period (1911 to 1960), T.H. Lee, now Taiwan's President, has shown that agriculture's significant contributions to industrialization were made possible by intelligent public policy and public investments which greatly increased farm productivity.12 Moreover organizational and institutional reforms involving active participation of the farm households in agricultural credit, extension, irrigation, water management, rural infrastructure, marketing, processing, and so on, were an integral part of public investment in agriculture. Labor and capital transfers occurred from agriculture, while the real wage remained near constant, throughout the period of industrialization.
"See Yair Mundlak, Domingo Carvallo, and R. Domenech, Agriculture and Economic Growth, Argentina 1913-84 (Washington, DC: International Food Policy Research Institute, 1988).
12See Teng-hui Lee, Intersectoral Capital Flows in the Economic Development of Taiwan, 1895-1960 (Ithaca and London: Cornell University Press, 1971).

Unlike Taiwan not only do many developing countries experience government failure, but at the international level too they face major barriers to trade. International agricultural markets are thin, volatile, and highly protected. Whereas developing countries have been reducing protection of their agriculture OECD countries continue to subsidize theirs. Clearly continued liberalization in developing countries without access to international markets will not be feasible. We turn to the changing patterns of international trade to illustrate its growing importance.
Following the severe drought of 1973, the share of developing countries in world cereal imports climbed at a rapid rate from 37 percent in 1970 to 44 percent in 1980 at a time when oil shocks reduced their capacity to finance increased food imports. Rising food imports led to a boost to import substitution of food which in Asia was helped by the Green Revolution. The rate of growth of cereal imports declined in the 1980s in part due to a robust agricultural performance in much of Asia which made many importers and food aid recipients self-sufficient. The debt crisis affected the import capacity of countries in Latin America and Africa. Nevertheless, by 1990, the share of developing countries in world cereal imports had increased to 53 percent and for the first time developing country imports exceeded imports by developed countries. The shares of both Asia and Africa increased significantly, in the case of Asia the proportion having increased from a quarter to a third of global imports, whereas in Africa the share doubled from a low 6 percent to a little over 12 percent. The behavior of Latin American countries was more erratic and less homogenous.

The reasons for the increased cereal imports in Asia and Africa were however radically different. In East Asia (China, Taiwan, South Korea, Malaysia, Indonesia, Thailand) and to a lesser extent in South Asia (India, Pakistan, Sri Lanka), imports were the result of increased per capita income, in turn a result of decline in the rate of growth of population, increased agricultural productivity, and rapid industrialization. The high income elasticity of demand for livestock products, edible oils, fruits, and vegetables and derived demand for feedgrains stimulated growth of imports which the Asian countries were able to finance through the increased export of labor intensive manufactured goods. Indeed as Figure 1 shows the share of agriculture in total exports declined more rapidly and consistently in Asia where broadbased development of agriculture had received considerable priority compared to Latin America or Africa where structural barriers (dualism within agriculture, dependence on a narrow set of commodities for exports and acute import substituting industrialization) reduced agriculture-industry linkages.
Asia experienced a substantial decline in the proportion of
population below poverty, although South Asia still has the highest incidence of poverty in absolute terms- -some 300 million people as compared to about 120 million in East Asia, 120 million in Sub-Saharan Africa, and 50 million in
Latin America and the Caribbean. 1 The proportion of the poor in rural areas is highest in Sub-Saharan Africa (80-90 percent), followed by South and Southeast Asia (70-80 percent). In Latin America, the proportion is lower, varying anywhere from 20-60 percent. Alleviating massive rural poverty is directly related to the performance of agriculture.
13See World Develo~ment Report 1990 (Washington, DC: World Bank, 1990), p. 29.

The doubling of the share of world food imports in Africa reflects the failure of domestic food production to keep pace with the accelerated rate of growth of population as well as a rapid shift in the consumption patterns away from traditional foods such as sorghum, millets and cassava to imported cereals such as rice, wheat, and maize. Econometric analysis suggests that declining world cereal prices aided by policies of developed countries, Africa's overvalued exchange rates, and urbanization were major contributors to this growth.14
Africa's food aid dependence increased, nevertheless, due to a
limited import capacity to meet the needed food imports. Food aid needs were estimated to be about 6 million tons in 1991-92 to maintain per capita food consumption at the 1986-1990 average, but 11.4 million tons if the UN minimum calorie requirements were to be met. The prevalence of severe and widespread malnutrition implied in USDA's estimates is also supported by the World Bank's various reviews of food security in a number of African countries. Both the 6 million and 11.4 million estimates were substantially higher than the 3.5 million tons of food aid actually supplied to Africa (or only about a third of the total food aid needs). This was a result of logistical problems (lack of port facilities, fuel, and trucks), economic constraints to the delivery and absorption of aid, acute needs arising from uneven food needs within countries, and emergency food aid. 15 Since Africa has neither the financial
"See William K. Jaeger, The Effects of Economic Policies on Africa Agriculture, World Bank Discussion Paper No. 147, Africa Technical Department Series (Washington, DC: World Bank, 1992).
15 The actual requirements varied considerably among countries. For
instance, whereas Sudan, Somalia, Ethiopia, Liberia, Angola, Mozambique, and Zaire needed food aid in part due to civil wars or significant political disruptions, Cote D'Ivoire and Nigeria, neither of which were previously food aid recipients needed food aid and their requirements were estimated to be

nor the logistical capacity to manage food imports or food aid, but has vast natural resources and a growing population, the importance of increasing food production there simply cannot be overstated.
Tables 2, 3, 4, and 5 show the growth rates of value added in
agriculture in 53 countries (22 in Africa, 10 in Asia, 7 in Eastern Europe and the Middle East, and 14 in Latin America) under adjustment for the period 1970 to 1990 as well as for the various sub-periods representing episodes with regard to external shocks, adjustment etc. Several features about these tables are of interest. First, rates of growth of value added in agriculture in Latin America, Asia, and Africa in the 1983-90 period are similar, the average growth rates being 2.7 in Latin America, 3.0 in Asia, and 3.03 in Africa annually. There are, of course, the.usual expected country differences, for example, in Asia, the East Asian countries having performed a lot better than the South Asian countries. However, the differences among continents are larger when considered in per capita terms. They represent both lower rates of population growth in Asia and Latin America and high labor productivity. Table 2 also shows a clear improvement in the rate of growth of value added in Africa in the 1983 to 1990 period compared to the entire 1970 to 1990 period. Those differences are less significant in Asia and Latin America, although there is considerable variation among countries in each region.
This evidence suggests that the favorable impact of structural adjustment observed by several studies on overall economic and agricultural
considerable (African Food Needs Assessment, USDA Economic Research Service, November 1991, Washington, D.C.). In Nigeria the ban on commercial food imports resulted in substitution of domestic sorghum, millets, and cassava for imported
cereals for rice, wheat and maize. But the opposite was the case in a number of countries where food aid consisted of these cereals.

growth reviewed later is perhaps more noticeable in African agriculture, and perhaps because price distortions prior to adjustment were more acute in the African economies than in Asian agriculture where distortions were lower, but
the reforms have been perhaps slower as well.16 China is the notable exception where a major land reform (the so-called shift to the personal responsibility system) and liberalization of the cereal market provided a major boost to agriculture. Yet China's agriculture remains more controlled today (with 20 percent of cereal production sold to the public sector at a fixed price and another 60 percent sold to the government at a negotiated price, compared to 10 percent in India and a similar percentage share in Tanzania at the peak of interventions). Clearly it is not simply the liberalization of the grain market or the land reform which explains China's better agricultural performance. A great deal of government behavior remains a puzzle, especially in terms of the lessons developing countries should learn from China and other well performing East Asian countries. Evidence also shows that the response of overall economies to structural adjustment has been more impressive in the middle income countries than in Africa. It is to this puzzle that we now turn.
16SeeAdiustment Lending: An Evaluation of Ten Years of Experience, Policy and Research Series No. 1, Country Economics Department (Washington, DC: World
Bank, 1989); Aidjustment Lending Policies for Sustainable Growth, Policy and Research Series No. 14, Country Economics Department (Washington, DC: World Bank, 1989; Africa's Adjustment and Growth in the 1980s, the World Bank and UNDP (Washington, DC: World Bank, 1989); and Riccardo Faini, "Infrastructure, Relative Prices, and Structural Adjustment," in Ian Goldin and L. Alan Winters eds. Open Economies: Structural AdiustMent and Agriculture OECD Development Centre and the Centre for Economic Policy Research (Cambridge: Cambridge University Press, 1992).

When the real growth rates of total agricultural exports (nominal dollar values of total agricultural exports deflated by the international MUV index) are considered, the majority of 53 developing countries experienced a decline in their import capacity in the 1983 to 1990 period relative to the entire 1970 to 1990 period (Tables 6,7,8, and 9).- But the decline in import capacity was greater for Africa than Asia, perhaps reflecting greater terms of trade effects because of Africa's greater concentration in a few agricultural exports with low income elasticities of demand for its exports. More is said on the international demand issues below. Africa showed gains in import capacity (i.e. smaller negative values) in the 1983 to 1990 period compared to the 1978 to 1981 period of acute economic crisis, perhaps suggesting increased export volume effect following structural adjustment in the later period.
While still highly tentative, these results are of considerable interest in understanding the process of economic development in the two sets of countries. First, due to growth in factor productivity, there was greater per capita accumulation of surplus which was ploughed back by the public sector into agriculture in Asia leading to a cumulative income growth compared to Africa. Second, due to a smaller share of agricultural exports in total exports and a greater diversity in those exports, Asia suffered a lower overall loss in income from adverse international terms of trade. However, importantly, when the external environment turned unfavorable in the 1980s Asian countries were more able to turn to the domestic market for a stimulus to their economies than was Africa. 17
Changing patterns of global fertilizer (nutrient) consumption also
17 Based on Recent Economic DeveloRments, various country reports prepared by the IMF.

reflect important differences in the rates of agricultural intensification in the two continents. Whereas Asia's share in global fertilizer consumption increased dramatically from 14 percent in 1970-71 to 36 percent in 1989-90, in the case of Africa, the growth according to FAO data was a mere 2.4 percent to
2.7 percent. Thus, Africa's share in world fertilizer consumption declined significantly.
Through the flows of external finance industrial country policies determine the extent to which developing countries can supplement their domestic investment effort including import of critical agricultural inputs. This is more true of small, low-income open economies of Africa than the large economies of Asia or Latin America. African countries receive more official development assistance (both in per capita terms, as well as share of government expenditures and GNP which is as high as 20 percent in many countries) relative to large Asian countries, although later we will show the effects of external finance on public investment in Asia as well. Donor financed imports of fertilizers on which much of Africa depends have shown great year to year variability due to its inevitable unpredictability. Adjustment loans which focus on liberalization of imports and domestic markets in fertilizers in circumstances of scarce foreign exchange have not focussed adequately on the effects of donor policies towards financing fertilizer
imports. is
18See Uma Lele and Robert Christiansen, "Markets, Marketing Boards, and Cooperatives in the MADIA Countries: Issues in Adjustment Policy in Africa," MADIA Discussion Paper No. 11 (Washington, DC: World Bank, 1989); Uma Lele, ed. Aid to African Agriculture: Lessons from Two Decades of Donor EKperience
(Baltimore: Johns Hopkins University Press for the World Bank, 1992).

The preceding discussion demonstrates that it is difficult to separate the effects of increased availability of foreign exchange and access to imported inputs from that of improved price incentives and investment climate or external shocks such as terms of trade changes or the weather. Adjustment studies tend to compare "before" and "after" adjustment situations because of the difficulty of determining the "without adjustment" situation. Decline in real incomes following adjustment may reduce demand for food commodities and shift relative output prices in favor of export crops. Not only may the output prices of the large foodcrop sector be more cushioned from international markets, but the foodcrop sector may even be adversely affected by the various other price reforms (for example, increase in the prices of inputs due to devaluation, removal of subsidies, and increased internal transport costs). But it may not benefit much from output price increases (which notwithstanding a government monopoly in effect may be uncontrolled to begin with or may lose their price support).
Studies of adjustment show that after controlling for the effects of external shocks and external finance, adjusting countries have performed better than nonadjusting countries, and those early adjusting countries with
three or more loans performed better than those with fewer loans. 19 Not surprisingly countries with a dominant manufacturing sector mostly in Asia, North Africa, and the Middle East performed better than those with a dominant agricultural sector. Adjustment has been relatively less successful in highly indebted countries in Latin America and in Africa- -not a surprising finding in
19See Vittorio Corbo, Stanley Fischer, and Steven B. Webb, Adjustment Lending Revisited: Policies to Restore Growth (Washington, DC: World Bank, 1992).

view of the preceding discussion.2 1
According to Faini's econometric study of 30 countries, which
specifically analyses effects of adjustment on agriculture, the agricultural
sectors of adjusting countries responded positively.22 Importantly, Faini
also shows that while both price factors and the availability of
infrastructure were significant, price factors were more significant than
20Conclusions of the two reviews of adjustment lending carried out in the
World Bank (Adjustment Lending: An Evaluation of Ten Years of Exnerience and Adjustment Lending Policies for Sustainable Growth) were that: the 30 countries
that received adjustment lending before 1985 had higher rates of economic performance than the 63 that did not receive the loans. Performance of the 12 countries that received three or more loans and that were substantial exporters of manufactured goods was much better. Adjustment lending has been relatively
less successful in highly indebted countries in Sub-Saharan Africa. The second review adjusted for the effects of initial conditions, external shocks, and the amount of external financing. It concluded that early intensive adjustment lending countries (EIAL) experienced larger increase in the average rate of growth of GNP than did other countries. Thus, Korea, Mauritius, Morocco, Ghana, and Thailand appeared to have stimulated growth more than the initial conditions, external shocks, and external financing would suggest. Exports as a share of GDP in constant prices increased a lot in these countries. But the picture was less positive in Nigeria, the Philippines, Malawi, Cote D IIvoire, and Mexico (although the situation regarding Mexico has changed since the studies were done). The picture has been less positive in the case of EIAL countries. After adjusting
for these same conditions both private and public investment fell in EIAL countries. The Bank's conclusions on the impact on the poor are more contrived.
---orderly adjustment supported by Bank lending seems to have been less costly for most of the poor and for the general populace than disorderly adjustment without Bank support was" (see Corbo, Fischer, Webb, eds. Adjustment Lending Revisited, p.14). The relationship of adjustment to social indicators is less
clear, but government expenditures on social sectors declined in most countries, leading to a drop in primary school enrollment. Adjustment is taking longer than expected and most Bank documents are placing greater emphasis on a detailed analysis of the social impact, including reallocation of services to the poor, severance payments, and retraining of unemployed workers but most of these are geared to urban workers.
21Conclusions of the two review of adjustment lending carried out in the World Bank.
22 See Riccardo Faini, "Infrastructure, Relative Prices, and Structural Adjustment," in Ian Goldin and L. Alan Winters eds. O~en Economies.

infrastructure in middle-income countries where manufacturing dominated and where presumably markets worked better. The effect of public infrastructure was stronger in countries where agricultural exports dominated, i.e., in lowincome (African) countries where accumulated deterioration of physical capital is now acute and a serious constraint to the functioning of factor and product markets. But due to data problems, econometric studies do not make a distinction between adjustment in the macro economy and in the agricultural sector. Thus, while in the literature on adjustment export crop prices faced by producers are assumed automatically to have increased following devaluations, in reality gains may not always accrue to producers. Tanzania is justifiably credited with a major macroeconomic adjustment but wrongly considered a success in export performance.- By 1992, only cotton production
had increased.23 But it faced major problems in transportation and processing. All other export crops had stagnated because gains from a large devaluation had been absorbed mainly by the financially strapped parastatals who also received the lion's share of credit from the banking sector. IMFimposed credit ceilings which were necessary to control inflation nevertheless caused a major credit crunch in the informal sector which affected the functioning of agricultural markets. 24 In Malawi, for nearly 20 years the government has continued its discriminatory low prices to small farmers growing tobacco while favoring estates and adjustment for nearly a decade had
23 See Adjustment Lending: An Evaluation of Ten Years of Experience.
24 See Uma Lele, "Can Technology Transfer and Macroeconomic Adjustment Sustain Africa's Agricultural Revolution Without an Agricultural Sector Strategy? The Case of Sasakawa Global 2000 Program in Tanzania," An impact evaluation report prepared for the Sasakawa Global 2000 program (University of Florida, Gainesville: International Studies and Programs Office, 1992).

not changed that state of affairs. In Kenya, although prices of the two major exports tea and coffee never experienced much (implicit or explicit) taxation, agricultural adjustment had not changed the character of the price policy much by 1991. Maize marketing had been liberalized partially in Kenya and fully in Tanzania. This greatly increased the producers options as regards marketing channels and helped reduce parastatal losses, but did not necessarily increase producer price levels. Input prices, however, increased sharply in all these countries. Declining international prices of both food and export crops in the 1980s, added to the squeeze on agriculture.
The problem is of course more acute in CFA countries. In the
World Bank sponsored study Managing Agricultural Development in Africa (MADIA) I had documented that the relative producer prices of export crops were already less favorable to their production in West Africa in the 1970s vis a vis food crops especially compared to East Africa. Thus, whereas the ratio of the coffee price to the maize price was nearly 45 to 1 in Kenya in 1977, i.e., at the height of the coffee boom, it was only 7 to 1 in Cameroon. This reflected both the higher explicit and implicit tax on coffee in Cameroon as well as the higher price of maize relative to Kenya (reflecting the higher level of urbanization aided by the effects of the oil boom). The relative disadvantage of export crops is even more pronounced in CFA countries since 1987 due to the increased overvaluation of the exchange rate and the decline in nominal producer prices of export crops required to balance the books of parastatals. The problem of CFA countries has been made more acute by the large devaluations in the neighboring countries. Clearly while the Franc zone has a number of advantages in terms of greater price stability and monetary discipline, its desirability from the viewpoint of competitiveness of

agriculture needs reevaluation.
We will later return to the other agricultural development policy issues faced in transition during adjustment.
The focus in adjustment programs has been more on price policies and short run changes than on investments, institutions, human capital, or technology development. Such focus may be explained by the relative ease in changing nominal prices than in reforming parastatals, strengthening agricultural research systems, building and maintaining a feeder road network, or changing investment patterns away from the urban sectors, tasks that earlier project lending undertook. Second, the difficulty in monitoring non-price reforms adds to the temptation to focus on prices. Third, the greater locationspecific knowledge of developing countries needed to recommend public investment reforms adds to the problem of donors giving advice since most of the staff are located in the capital cities of donor countries. Fourth, prices are more easily amenable to analytical tools of neoclassical economic theory than government expenditures, institutional changes, or technologies. Finally, the importance of macroeconomics has increased in the course of structural adjustment with a concomitant decline in the role of sector economists and, even more important, other technical fields critical for the development of agriculture. Table 10 shows the effect of this phenomenon on the composition of 198 recent publications on structural adjustment. They reflect the bias towards macro economic analysis, middle income countries, and poverty relative to agriculture. Publications on poverty alleviation often have little treatment of agriculture. Africa has relatively more publications

on agriculture than Asia, but the exclusive focus on Africa adds less new insights than would be possible from a more comparative effort to learn lessons from continents where agriculture has been more dynamic, but where governments have played an important role.
Excesses of the 1970s in terms of rapid expansion of public
investments in the absence a conducive policy and institutional environment, however, have also demonstrated that equating increased investment with priority to agriculture (or industry) is not justified. Nevertheless an important indicator of the decline in interest in agriculture is the reduced public sector investment in agriculture, being noted in many developing countries although investment requirements of agriculture remain very
large.25 That decline in agricultural investments seems to be part of a general decline in rates of investment being noted even in adjusting countries
that have considerable access to external finance.26 With lesser access to external finance the decline in investment in non-adjusting countries must of course be greater. This can be seen from Table 11 which also shows that the decline in the rates of public investment are greater in Africa than in Asia and Latin America. The declining share of external assistance to agriculture has of course reinforced the tendency for decline in public sector investments in agriculture of developing countries.
25FAO has estimated total investment requirements of agriculture to be about $1500 billion by year 2000 or about 80 to 100 billion annually and the
Bank's agricultural staff seem to accept these estimates. Changes in lending in only a handful of countries where lending to agriculture has traditionally been
concentrated explains the decline in the Bank's lending. For detailed evidence supporting the arguments in the text, see Lele, Adu-Nyako, and Emerson, "Structural Adjustment and Agriculture," forthcoming.
26See Corbo, Fischer, and Webb, Adjustment Lending Revisited

The World Bank reviews of adjustment lending have justifiably
argued that since growth in adjusting countries has been higher than it would have otherwise been, even with reduced levels of investment, factor efficiency
must have increased.27 Certainly reducing price distortions must have had a positive effect on efficiency. But such efficiency augmenting effect of price reforms tends to be a once and for all increase. Long term increases in output must come from increase in investment in research, extension, education, health, transport, etc.
The World Bank's lending to agriculture is a significant barometer of the overall financing by other donors since they typically follow similar patterns, and co-financing with the Bank has increased after structural adjustment commenced.
Figures 2, 3, 4, 5, and 6 show World Bank lending to agriculture relative to SAL lending and overall lending in the principal regions of the world. The share of agriculture in the total declined from about 31 percent in 1975 to less than 18 percent in 1990. The share of structural adjustment lending in the total increased from about 9 percent to nearly 15 percent over the same period. Increased adjustment lending of course helps the development of agriculture not simply through the removal of macro price distortions but through the increased availability of consumer goods, spare parts, and agricultural inputs. World Bank reports are at pains to stress that many of the changes that have caused the decline in the share of lending are associated with a desire to improve the performance of the agricultural sector rather than a result of its neglect or abandon.
27See Corbo, Fischer, and Webb, Adjustment Lending Revisited

The decline in agriculture's share in World Bank lending has been more evident in Asia and Africa where it reached its peak in the late 1970s although there are important intercountry differences within regions. The decline in the Middle East, North Africa, and the European region has been less significant and overall lending actually increased in Latin America in the 1980s after declining until 1978, although once again there are country differences. Since only six countries (Brazil, China, India, Indonesia, Mexico, Pakistan, and Turkey) received nearly two thirds of total World Bank lending to agriculture changes in assistance to them offer a number of insights. While it is difficult to generalize on the precise patterns, in countries where the World Bank had a substantial presence in agriculture in the 1970s, and familiarity with the problems of agriculture, lending declined although the quality of the countries' overall agricultural performance or the quality of the Bank's lending portfolio in these countries was not necessarily
poor.28 In some countries, however, the World Bank's own portfolio selection did not reflect the best opportunities for investment, as, for example, in Kenya where the country's performance was much superior to that of the Bank's portfolio *29
28For instance lending declined sharply in India, Indonesia, Mexico, and Nigeria (the only poor performing country) but increased sharply in China and
Brazil (a country that failed to adjust). The trends were more mixed in Morocco and Turkey (both of whom were successful adjusters). There it increased in the mid 1980s after having declined before. Among the smaller borrowers lending also declined in Tanzania, Kenya, Malawi, Malaysia, Philippines, Korea, Bangladesh and Rumania, but increased in Sudan (a country devastated by civil war and other political problems) and Tunisia. Lending remained high cut declined slightly in Sri Lanka and Cameroon.
291t could justifiably be argued that donors should be willing to take risks and finance marginal investments unlikely ,to be funded by countries. But this would require greater explicit recognition of the lack of knowledge of precise local constraints, an experimental approach to addressing them, and flexibility
in learning by doing. Whereas these principles are simple enough (see Uma Lele, 23

The waning interest in agriculture is a result of a complex set of interacting factors. Of course, the pressing foreign exchange needs of developing countries following the economic difficulties at the end of the 1970s contributed to the focus away from investment projects. Increased scrutiny of aid agencies by their constituencies in terms of the efficiency of staff resources and the consequent pressure to lend more funds with less staff has reinforced the support for balance of payments over investment projects. But there have also been other factors more specific to agriculture. First, hunger has been perceived to be the result of a lack of effective demand
rather than inadequate supply.30 Therefore emphasis has shifted from production to consumption. While this conclusion is justified at both the international and the national levels, it is an oversimplification of a complex problem in which employment opportunities in agriculture and related fields clearly determine the ability of the rural poor to earn income. In Africa, the failure of the plethora of integrated rural development projects
in the 1970s has led to a shift out of agriculture.31 If ex-post returns calculated in all sectors by the Operational Evaluation Department of the
The Design of Rural Development: Lessons from Africa, Baltimore: The Johns Hopkins University Press for the World Bank, 1975) donors have shown a tendency to finance large projects of inappropriate technology with less flexibility in changing course during the period of implementation when problems occur. But
this is not a problem specific to agriculture. See Uma Lele and L. Richard Meyers, "Growth and Structural Change in East Africa: Domestic Policies, Agricultural Performance, and World Bank Assistance," MADIA Discussion Paper 3 (Washington, DC: World Bank, 1989).
30See Jean Dreze and Azuartya K. Sen, Hunger and Public Action (Oxford: Clarendon Press, 1989).
31See Uma Lele, ed. Aid to African Agriculture.

World Bank are taken at their face value ex-post returns in agriculture were lower relative to ex-ante returns than in other sectors.32 Clearly too high a set of expectations by donors in the 1970s have inadvertently contributed to turning attention away from agriculture. Within agriculture the Bank's emphasis on national agricultural services in Africa effectively focuses on agricultural extension when the challenge of intensifying agriculture requires investment in agricultural research, small-scale irrigation, feeder roads, fertilizer imports and distribution, agricultural finance, and a price policy, including, where appropriate, a selective application of subsidies. Recently concerns about the environment, poverty, and women's participation have taken precedence over agriculture.
In Asia, decline in world cereal prices reduced economic justification for investment in irrigation. Increased salinity and displacement of traditional populations in large-scale irrigation projects reduced donor enthusiasm for financing irrigation in the face of growing criticism from environmentalists. Similarly, the lack of popular support in donor countries for the use of chemical inputs has reduced interest in projects involving fertilizers.
Donors also relied heavily on public sector enterprises as a major conduit for transfer of resources to agriculture. Indeed, elsewhere I have documented that the rapid growth of public enterprises noted in many small developing countries simply would not have occurred without external assistance (Lele and Christiansen). The loss of credibility of the public
32See Gerhard Pohl and Dubravko Mihaljek, "Project Evaluation and Uncertainty in Projects: A Statistical Analysis of Rate-of-Return Divergences of 1,015 World Bank Projects," in The World Bank Economic Review, Vol. 6, No. 2, May 1992 (pp.255-78).

sector and the increased role assigned to the private and the nongovernmental sectors have made it difficult to channel substantial resources to the agricultural sector, although some resources are being provided to private traders to promote input and output marketing and to NGOs. There is a danger that the capacity of local NGOs to utilize resources effectively is outpaced by the number of donors and the volume of resources available to assist them. That may be harmful to the sustainability of genuine local NGOs. The widespread growth of rural development funds being noted in many countries while agricultural investments decline may similarly turn out to be a flash in the pan that keeps the existing governments in power, but does little to improve the capacity of the line ministries to perform their legitimate development functions, unless of course local communities are truly empowered. To illustrate the complex policy problems faced by the line ministries and agencies concerned with agricultural development, I end this paper by highlighting several sector policy issues faced in adjustment.
The Fallacy of Composition Argument (Again?)
Low demand elasticities for tea, coffee and cocoa continue to pose a problem for its producers as a group although individual countries that have not discriminated against the export crop sector have performed better than those that did.3 The problem of fallacy of composition is more acute for African countries with a narrow base of exports and fewer options for diversification
33See Uma Lele, "Agricultural Growth, Domestic Policies, the External Environment, and Assistance to Africa: Lessons of a Quarter Century," MADIA Discussion Paper No. 1 (Washington, DC: World Bank, 1989).

than their Asian and Latin American counterparts who have gained their shares. In a recent cross-sectional study, Evans, Goldin, and van der Mensbrugghe show that an across-the-board tax cut by several small producers of export crops with limited demand will result in a substantial loss of GDP.3 Panagariya and Schiff also show that optimal choice of taxes or quotas and the associated growth in the country's output can lead to a decline in the combined real income of the exporting countries."3 Whereas cross-sectional analysis and game theoretic approaches have their own limitations, they do reflect the donors' long-standing dilemma in Africa. It'is clear that Africa must compete in the production of these crops, by bringing down the cost of its own production so as to make it unattractive for its competitors to remain in production. However, productivity growth has been more rapid in Malaysia than in Cameroon, Nigeria, or Cote d'Ivoire. Apart from price incentives, neglect of agricultural research, extension, credit, and inputs has been a particularly acute problem in African countries with a few exceptions such as Kenya or Zimbabwe. Malaysia, which borrowed oil palm and cocoa technology from Nigeria, by contrast, has excelled in these respects. Despite higher nominal wages, it has had lower unit costs of production. Devaluations in Africa have improved competitiveness but in the absence of technical progress devaluations have caused a decline in real wages.
Since the early 1970s, donors have not helped in Africa's
34 See David Evans, Ian Goldin, and Dominique van der Mensbrugghe, "Trade Reform and the Small Country Assumption," in Ian Goldin and L. Alan Winters eds. Open Economies.
"5See Arvind Panagariya and Maurice Schiff, "Taxes versus Quotas: The Case of Cocoa Exports," in Goldin and Winters, eds. Open Economies.

declining export crop sector. They advised African countries to limit their traditional exports. For example, in 1973 the World Bank, based on an FAO study, adopted a policy not to finance expansion of tea and coffee production-although when it did assist in processing in Kenya, it contributed much to Kenya's excellent tea industry. The EC's position has vacillated. The price policy advice the World Bank and other donors provide under adjustment lending is correctly to encourage producing countries to reduce the level of implicit and explicit taxation, although the Bank has not changed its policy towards financing investments in production.
On the whole donors have swung between export promotion and food security rather than providing sound advice and investment support for efficient production of both. Due to Africa's greater dependence on external advice and capital for financing investments relative to Asia or Latin America it was the loser in the world market shares which the latter two continents gained. Clearly donors will need to assist Africa to diversify out of its traditional exports without discriminating against them. Countries such as Malaysia have much that they can re-export to Africa in terms of the organization of research, extension, agricultural finance, and rural infrastructure.
Price Stabilization
Virtually every country in the world attempts to stabilize prices both to cushion consumers, processors, and producers from the extreme vagaries of international price fluctuations as well as to stabilize prices across regions and seasons. Many governments have a monopoly on domestic procurement as well as on exports and imports, and use quantitative restrictions or tariffs. Most 28

have domestic buffer stocks together with policies of domestic purchases and sales. Marketing boards are another mechanism for stabilization and variable tax is used in several countries.
Clearly the private sector does not indulge in these operations since they are not profitable. Thus some cost to the government to meet welfare, political, and economic objectives is unavoidable. The criticisms of these schemes have been that public sector operations are inefficient, benefit undeserving groups, have high monetary and fiscal consequences, misallocate factors of production, thwart the growth of private enterprise etc. Depending on whether the high operating costs and subsidies of marketing organizations are financed through the budget or the expansion of credit by commercial banks (which is not repaid) the monetary and fiscal costs of these interventions can be very high indeed.
The World Bank's advice to governments, correctly, has been to: a) avoid getting involved in internal or external trade directly rather than using trade based mechanisms including in the case of tradeable goods by relying on a variable levy and market forces; b) to stabilize prices only partially by setting rather wide price bands as net costs to the government are directly related to the extent of price stabilization; c) to the extent possible to relate prices to international prices; d) to protect only the vulnerable groups.
While these principles are undisputable, the difficult problems relate to the transition from a controlled to a partially controlled system. There are many empirical issues which have not yet been fully explored through research. First, in the absence of capacity of governments to regulate the liberalized markets, the predatory behavior now attributed to public agencies 29

may be replaced by that of an oligopolistic private sector. Especially in the absence of competitive goods markets and a severe credit crunch, hurriedly instituted price policy reforms do not always achieve their intended effects. In Kenya, for example, "liberalization" of wheat imports (which were the monopoly of the much criticized maize board) resulted in the government allocating import licences to a favored few. This shifted the profits used to cross-subsidize maize operations (ostensibly for the benefit of urban. consumers) by the maize board to the already well-off, including some politicians. The same phenomenon has been noted in Senegal with regard to the allocation of import licences for rice 36 and in Nigeria. Indeed, the government of President Babangida in Nigeria acquired considerable popularity by vowing to ban rice imports, among other things, because of the public's recognition that a few army generals were the major beneficiaries of import licenses issued during the oil boom3' and a concern that import liberalization would once again transfer rents to them instead of the producers who would respond to higher prices by increased domestic production. By all accounts, import controls have shifted Nigerian food habits to their traditional foods such as plantains, cassava, etc. Raising food prices caused a considerable supply response, although high prices are detrimental to consumers. Nigeria is developing a capacity to diversify its exports out of oil so as to finance increased food imports. The policy of phased liberalization of imports is one which East Asian countries have followed
36 See John Waterbury, "Agricultural Policymaking and Stagnation in Senegal," MADIA Working Paper, Africa Technical Department, Agriculture Division (Washington, DC: World Bank, revised 1990).
37 See Henry Bienen, "Politics and Agricultural Policy in Nigeria," HADIA Working Paper, Africa Technical Department, Agriculture Division (Washington, DC: World Bank, revised 1990).

effectively. The needs of the export sectors were given priority over those which competed with domestic production and consumption. In that vein, Africa, which has become a major importer of canned foods, livestock, and dairy products, may benefit from some protection for its domestic food production, provided it is for a limited period of time and is associated with effective government policy to increase competitiveness, particularly in view of the continued subsidies in OECD countries.
Similarly domestic grain markets are not competitive in many developing countries especially where governments have suppressed private activity over a long period, for example, in Russia as much as in Tanzania. The severe deterioration of physical infrastructure combined with a lack of information, transport, credit, etc. compounds the problem of lack of competitiveness. Clearly, steps need to be taken to increase competition, Including especially massive investments in rural feeder roads. But infrastructure development cannot occur overnight and certainly not in a situation of declining rates of public and donor investments. Much of initial increase in investments in physical infrastructure in Africa, where the need is greatest, focussed on ports and trunk routes. This is not simply a result of a continued urban bias but also reflects an extreme weakening of local and regional governments and a weak private contracting sector. The problem of inadequate rural infrastructure confronts China, East Asia, and South Asia as
Well. 38
Furthermore, even competitive markets do not necessarily mean
stable domestic prices in the presence of unstable production. Unavailability
38See Uma Lele," Can Technology Transfer and Macroeconomic Adjustment sustain Africa's Agricultural Revolution Without An Agricultural Sector Strategy?".

of foreign exchange or food aid to stabilize domestic supplies clearly constrains Africa much more than Asia or Latin America, and this should have some effect on the donors advice on import reliance.
Some price stability is essential to protect the consumption and income of the poor who spend a large share of their income on food. A slow recognition of this fact by the IMF and the World Bank is leading them to develop safety nets through more fine-tuned and targeted food subsidies and this is a development in the right direction. However, there are several ways in which this approach need to be strengthened. First, due to a lack of knowledge of rural areas, adjustment literature frequently ascertains that all rural households are net sellers of food and assumes that adjustment of exchange rates and high food prices will benefit them. The safety nets focus on the urban poor and ignore the large numbers of the rural poor that have become market dependent throughout the world. Indeed the rural poor are even more vulnerable than their urban counterpart due to their dispersed nature and weak political voice. Second, the timing of liberalization often overlooks the government's capacity to sustain the liberalized market through release of adequate food supplies to dampen prices and avoid further speculation. In some cases, the timing of liberalization has been inopportune, as for example in the years of drought, with the lack of food aid or foreign exchange to import food. Third, the focus on consumer subsidies is leading donors to overlook the need for some assurance of minimum prices to producers (perhaps by linking them to a moving average of projected international prices to avoid much protection). Such absence of a producer orientation in price policy advice is a more serious problem in Africa with its poor infrastructure, weak private sector, large internal transportation costs and high price bands.

With all their high fixed and monetary costs, public purchases in domestic markets greatly helped in integrating national markets. Due to their greater leverage (but a lack of practical experience or knowledge of public policy) donors have succeeded in dismantling government interventions to a greater extent in Africa than in Asia where governments still exercise considerable role in stabilizing supplies and prices within and across years, as seen from the contrasting examples of China and Tanzania. China's major land reform after the communist takeover greatly improved rural land distribution and reduced the need for food distribution. Is reducing public food distribution in rural areas the only solution or is there some scope for increasing public sector efficiency and accountability? In discussing the maize price reforms in Zambia, Gulhati has stressed the complexity of the cereal price reform, the lack of information and expertise in donor agencies or governments, and the need for greater political and welfare sensitivity.39
Fertilizer Subsidies
The issue of fertilizer subsidies is particularly pertinent in the context of food security. As in the case of price stabilization, fertilizer subsidies are justifiably being removed because the gains from low prices mainly accrue to middlemen and commercial farmers, administration of subsidies makes it difficult to open up the fertilizer trade to the private sector, and the budget constraint unnecessarily rations the supply and leads to shortages and black markets. These various effects are clearly contrary to those intended. Yet Lele, Christiansen, and Kadiresan have shown that issues related to
39See Ravi Gulhati, The Making of Economic Policy in Africa (Washington, DC: World Bank, Economic Development Institute, 1990).

fertilizer use in rainfed production involve complex and location-specific interactions of technical and economic factors in a situation of inadequate research, poor informational base, high risks, and uncertainty. 40 Fertilizer subsidies may be essential for: (a) households in marginal areas or where fertilizer response coefficients are low, transport costs are high, demand for fertilizers is highly variable and unpredictable due to climatic factors, the private sector does not have the incentive to develop the market, and it is cheaper to provide a subsidy on transportation costs related to fertilizer distribution (thereby also ensuring scope for private sector sales at the regional level and below) than to distribute food to achieve food security of rural households consistently vulnerable to food shortages. A regular public presence in rural areas in various forms enables the successful use of administrative machinery in periods of drought, as demonstrated on several occasions in India and Kenya.
Clearly, effective location-specific agricultural research that increases factor productivity will reduce the need for input subsidies. But agricultural research has typically been underfunded despite consistent evidence of high rates of return. The allocation of funding to national agricultural research has taken a heavy toll since the adjustment process began, in part due to a weak constituency for agricultural research even under the best of circumstances. Moreover, the quality of management of the existing resources to research is declining with resources being devoted to rapidily declining real salaries and few being allocated for operations and maintenance.
40 See Uma Lele, Robert Christiansen, and Kundhavi Kadiresan, *Fertilizer Policy in Africa: Lessons from Development Programs and Adjustment Lending, 197087," MADIA Discussion Paper 5 (Washington, DC: World Bank, 1989).

Employment and Income Effects
Regional income and employment shifts of trade liberalization also pose shortrun problems, although liberalization has a positive, long run effect as shown in Mexico. Despite large overall welfare gains, Levy and Wijnbergen argue for a gradual introduction of the North American free-trade agreement to allow the time for adjustment in peasant households (with little access to formal education) in southern Mexico.' They may be unable to obtain employment in manufacturing and high value commercial agriculture in Northern Mexico that will be the primary beneficiary of trade liberalization. Levy and Wijnbergen acknowledge the lack of understanding of the functioning of labor markets and rural-urban migration.
Agricultural Credit
Asymmetries in the allocation of capital raise similar issues with regard to agricultural credit. The financing requirements of a dynamic agriculture can be very large in macroeconomic terms. Yet, the Philippines agriculture presents the general prototype noted in developing countries with an average share of agriculture in GDP of 30 percent in the period 1966-1984, but agriculture accounting for only 8 percent of the share of formal credit."2 In urban areas large commercial and industrial firms are the main beneficiaries of credit. Within agriculture, the size preference of lenders
"'See Santiago Levy and Sweder van Wijnbergen, "Agricultural Adjustment and the Mexico-USA Free Trade Agreement," in Ian Goldin and L. Alan Winters eds. Open Economies.
42See Sagrario L. Floro and Pan A. Yotopoulos, Informal Credit Markets and the New Institutional Economics: The Case of Philipine Agriculture (Boulder, CO: Westview Press, 1991).

is again visible. The sectoral allocation of credit favors export and commercial crops. The government institutions tend to reach clients who easily meet the bankability criteria whereas the informal market serves the clients whose risk in repayment is greater, and where the formal sector is not likely to venture.
In the context of adjustment, these patterns of credit allocation have been associated with urban bias, financial repression, and overt government acts such as regulation of financial intermediaries, control of interest rates, and government intervention in the credit market. While these are justified concerns they result in an excessive emphasis on adjustment of interest rates and overlook the large institutional and infrastructural gaps that cause fragmented credit markets.
Floro and Yotopolous demonstrate the fragmentation of credit markets in the Philippines dictated by the current nature of the economic environment and the existence of a great variety of interest rates, i.e., interest rates being lower for linked than unlinked loans, higher in marginal than developed areas, higher for poorer than richer farmers, etc.43 Although rural traders/lenders try to avoid adverse selection risk, often farmers who lend to other poorer farmers invite it. They offer low interest rates in order to facilitate debt accumulation that triggers in collateral and enables eviction from land.
Based on the experience of the Southern Cone countries, Cho and
Khatkhate also illustrate the problems of excessive increase in interest rates
43See Floro and Yotopoulos, Informal Credit Markets.

following liberalization which is unjustified by the fundamentals.44 They point out that high interest rates are as detrimental to investment as low interest rates are to resource mobilization.
Another major problem in the course of liberalization is that
budget deficits and the losses of parastatals crowd out private (especially rural) demand for credit as shown by the examples of Tanzania and Ghana.'5
Yet donor reports on financial reforms have little to say about
the extent to which demand for credit by the informal sector, particularly the private traders and farmers, is being met. As the government removes itself from rural finance, input and output trade, there are various ways in which lack of credit at the microlevel can have a severe impact on agricultural output and income and asset distribution, for example by limiting access to modern inputs, forcing the sale of farm assets, reducing maintenance expenditures, leading to suboptimal use of inputs or forcing shift to a suboptimal crop-mix, and increasing land concentration through foreclosures of small farms. Broad availability of agricultural credit on affordable terms to small farmers has a considerable positive value for social welfare. Floro and Yotopoulos argue and I concur that financial markets require both government regulation and government assistance for improving their performance.
Land Policy
The discussion on credit indicates that unequal access to finance may
'4See Yoon-je Cho and Deena R. Khatkhate, Lessons of Financial
Liberalization in Asia: A Comparative Study (Washington, DC: World Bank, 1989).
ASSee Uma Lele, "Can Technology Transfer and Macroeconomic Adjustment Sustain Africa's Agricultural Revolution Without an Agricultural Sector Strategy?"; and "The Sasakawa-Global 2000 Project in Ghana: An Evaluation," March 1991.

inadvertently skew land distribution. On the other hand, redistribution of land rights has been at the forefront of adjustment in many countries- -most notably in China, but also in Poland and the former Soviet Union. Evidence universally suggests that the short term impact of land redistribution on production can well be adverse, with considerable balance of payments implications, although in the medium and long term assured land rights ensure incentives to invest in land. Information on the best practices during the transition from public to private ownership of land,however, is limited. Such information needs to be collected to be of assistance to countries in transition.
Agriculture is important in a macroeconomic context. The transition of an economy from a predominantly agricultural to a manufacturing economy depends on the supply response of agriculture. That response depends on price as well as nonprice factors including especially public policy and investments in physical and human capital and improvements in regulatory and organizational capacity and an important regulatory and facilitating role for the government. Nonprice factors are more important than price factors for countries at an early stage of development and more important for a long run supply response.
With the growing concern about macroeconomic disequilibria since the early 1980s, attention to the complex problems of developing agriculture has diminished and public investments in agriculture have declined. Analytical work has shifted to middle-income countries, industry, and finance and focuses more on price than nonprice factors. This is a serious problem in understanding the reform process, including in particular the best practices 38

in agriculture of the more successful countries which have resulted in rapid overall growth. Agriculture clearly played a major role in East Asia relative to South Asia which in turn performed better than Africa. Latin America and Asia performed better than Africa in per capita terms although problems of distribution remain acute in Latin America.
Future rates of growth may be adversely affected by the decline in the rates of agricultural investments that appear to be part of a general decline in the rates of investment being noted in developing countries. It is clear that the capacity of countries to weather multiple external shocks is greater when agriculture is developed than when it is not, and that requires a substantial increase in investment. Experience of the 1970s has also shown that, without an effort to increase the absorptive capacity of the agricultural sector, the investment of additional resources may simply be wasted. But to improve the efficiency of needed investments requires a renewed focus on the performance of governments. The timing, speed, and the extent of liberalization raise many specific issues which often appear not to have been addressed prior to the introduction of reforms. There is clearly a vast research agenda of a comparative nature to make the reform process both more efficient and more humane.

Table I: Direct, Indirect, and Total Nominal Protection Rates for Agriculture (Selected Countries, Selected Years)
Degree of Tax Total Of
Over- Caused By Direct &
Country and Period Indirect Valuation Tariff Direct Indirect
Cote D'Ivoire (1960-82) -23.3 -29.6 -23.2 -25.7 -49.0
Ghana (1958-76) -32.6 -38.1 -32.4 -26.9 -55.1
Zambia (1966-84) -29.9 -50.6 -21.4 -16.4 -46.3
Average for Group I -28.6 -39.4 -25.7 -23.0 -51.6
Argentina (1960-84) -21.3 --17.7 -39.5 -17.8 -39.1
Colombia (1960-83) -25.2 -18.8 -37.8 -4.8 -30.0
Dom. Republic (1966-85) -21.3 -19.8 -20.8 -18.6 -39.9
Egypt (1964-84) -19.6 -17.4 -27.5 -24.8 -44.4
Morocco (1963-84) -17.4 -21.0 -13.4 -15.0 -32.4
Pakistan (1960-86) -33.1 -31.0 -44.9 -6.4 -39.5
Philippines (1960-86) -23.3 -19.3 -33.0 -4.1 -27.4
Sri Lanka (1960-85) -31.1 -14.8 -40.1 -9.0 -40.1
Thailand (1962-84) -15.0 -16.0 -13.9 -25.1 -40.1
Turkey (1961-83) -37.1 -30.9 -57.4 5.3 -31.8
Average for Group II -24.4 -20.7 -32.8 -12.0 -36.4

Table 1 : contd.
Degree of Tax Total Of
Over- Caused By Direct &
Country and Period, Indirect Valuation Tariff Direct Indirect GROUP III
Brazil (1969-83) -18.4 -12.8 -21.4 10.1 -8.3
Chile (1960-83) -20.4 -17.6 -37.4 -1.2 -21.6
Malaysia (1960-83) -8.2 -7.3 -9.9 -9.4 -17.6
Average for Group III -15.7 -12.6 -22.9 -0.2 -15.8
Rep. of Korea -25.8 -36.4 -26.7 39.0 13.2
Portugal -1.3 -2.3 -1.0 9.0 7.7
Average for Group IV -13.6 -19.3 -13.9 24.0 10.4
Average for all Groups -22.5 -22.3 -27.9 -7.9 -30.3
Source: Krueger, Schiff, and Valdez, "A Synthesis of the Political Economy
in Developing Countries," p. 61.

Table 2:
Agriculture, Value Added Growth Rates -- Africa
Country 1970-90 1970-72 1973-77 1978-81 1983-90 Post
Botswana -0.69 25.13 4.37 -4.66 1.38
Cameroon .4.07 31.53 1.99 11.72 2-.37
Cote d'Ivoire 2.58 4.06 1.33 5.76 5.23
Gambia 3.73 9.28 2.07 9.84 5.34
Ghana 0.30 4.29 -6.38 1.19 2.99
Kenya 3.92 13.75 4.19 2.99 3.53
Madagascar 1.08 0.02 -0.48 1.69 2.73
Malawi 2.41 13.17 6.36 -4.26 1.72
Mauritania 1.38 10.97 -3.28 7.75 0.77
Mauritius -1.05 10.09 -7.48 -9.15 1.76
Niger -0.54 -2.46 -2.61 -0.07 1.44
Nigeria -0.59 -1.23 -0.47 -4.29 3.25
Senegal 0.95 1.64 8.38 -2.58 2.52
Sierra Leone 4.53 0.62 3.81 1.72 2.87
Sudan 1.88 4.11 10.13 -9.19 1.18
Tanzania 2.04 3.37 0.21 2.07 5.06
Togo 3.57 2.44 -1.63 4.79 6.95
Uganda 0.04 0.66 0.42 -5.09 2.11
Zaire 1.96 2.12 0.83 2.65 2.58
Zambia 1.98 3.44 4.61 -0.08 6.04
Zimbabwe 1.54 19.76 -0.08 4.71 4.73
Region 1.67 0.79 3.03
Source: FAO Trade Yearbook

Table 3
Agriculture, Value Added Growth Rates -- Asia
Country 1970-90 1970-72 1973-77 1978-81 1983-90 Post
Bangladesh 2.24 -7.66 4.48 1.19 1.44
China 4.56 0.41 1.27 3.24 4.39
India 2.43 -3.46 3.52 2.29 2.75
Indonesia 3.90 3.79 2.39 6.20 3.-r
Korea 2.16 2.69 6.27 -2.72 1.33
Malaysia 4.14 4.49 4.44 3.68 4.64
Pakistan 3.83 0.15 2.01 6.67 4.56
Philipines 3.59 4.53 5.23 4.47 2.38
Sri Lanka 3.11 0.31 2.78 3.91 0.95
Thailand 4.10 1.18 4.16 1.71 4.07
Region 3.4 3.07 3.00
Source: FAO Trade Yearbook

Table 4
Agriculture, Value Added Growth Rates -- Eastern Europe & Middle East
Country 1970-90 1970-72 1973-77 1978-81 1983-90 Post
Algeria 5.76 2.34 11.13 7.62 8.04
Egypt 2.76 4.63 1.71 3.15 2.36
Hungary 2.77 4.49 2.36 2.27 0.99
Morocco. Z.79 3.71 0.09 -5.24 9.46
Tunisia 2.72 22.91 4.06 4.28 0.59
Turkey 3.02 5.56 7.30 1.54 3.13
Yugoslavia 2.05 5.21 3.31 1.94 0.49
Region 3.12 2.22 3.58
Source: FAQ Trade Yearbook

Table 5
Agricultural, Value Added Growth Rates -- Latin America
Country 1970-90 1970-72 1973-77 1978-81 1983-90 Post
Argentina 1.49 1.75 1.58 -0.84 -1.72
Bolivia 2.55 5.89 4.41 1.28 2.73
Brazil .3.93 7.02 5. 7.62 3.67
Chile 3.34 -4.78 7.85 4.52 5.79
Colombia 3.45 4.26 4.41 3.29 3.53
Costa Rica 2.33 5.02 1.15 1.45 3.33
Dominican Rep. 2.46 4.57 1.73 -2.59 -0.16
Ecuador 3.07 4.51 3.79 5.01 7.45
Honduras 2.42 5.44 0.39 3.22 2.54
Jamaica 0.08 7.39 2.26. -4.26 -0.18
Mexico 2.49 3.18 2.89 3.96 -0.04
Panama 1.89 2.09 3.49 -0.57 1.51
Peru 1.49 -3.97 1.31 1.22 5.38
Venezuela 2.96 -0.01 3.69 1.04 4.32
Region 2.43 1.74 2.72
Source: FAO Trade Yearbook

Table 6
Total Agricultural Exports Growth Rates -- Africa
Country 1970-90 1970-72 1973-77 1978-81 1983-90 Post
Botswana 0.93 1.55 1.76 10.58 -3.86
Cameroon -0.75 -11.07 7.01 -13.14 -2.45
Cot- d'Ivoire 2.15 -6.18 19.83 -7.05 -5.32 -3.02
Gambia -9.73 -1.52 9.10 -20.44 -17.84 7.4&
Ghana -5.38 -18.33 6.12 -20.85 -1.88 -6.45
Kenya 0.17 4.32 15.89 -9.79 -5.64 -2.42
Madagascar -4.38 1.13 4.19 -14.12 -10.67 -14.63
Malawi 1.49 11.76 9.28 4.82 -2.37 -1.39
Mauritania -3.30 1.72 -0.80 1.97 -7.04 -1.02
Mauritius -0.27 15.88 0.04 -10.17 2.14 1.71
Nigh, -5.25 -6.31 1.71 9.01 -9.93 -5.65
Nigeria -8.76 -22.43 -0.78 -27.01 -11.40 -10.62
Senegal -5.26 9,.45 25.62 -32.45 -2.58 -3.56
Sierra Leone -5.38 0.42 3.41 -19.96 -14.26 -26.74
Sudan -4.65 1.69 1.37 -7.64 -3.29 -3.35
Tanzania -5.86 3.06 3.32 -1.73 -7.36 -7.56
Togo -0.56 -19.38 10.75 -16.52 -0.45 -5.05
Uganda -6.26 -3.81 6.67 -18.24 -15.79 -18.61
Zaire -4.84 0.34 6.08 -14.76 -11.00 -9.53
Zambia -3.26 0.70 -17.19 -16.86 6.72 -3.52
Zimbabwe 0.05 19.36 0.89 6.61 0.20 0.15
Region -10.37 -5.91
Source: FAO Trade Yearbook

Table 7
Total Agricultural Exports Growth Rates -- Asia
Country 1970-90 1970-72 1978-81 1983-90 Post
India -0.27 0.91 6.66 -2.83
Indonesia 2.77 -7.26 -5.49 -1.43 -10.35
Korea 4.41 17.96 -0.39 6.81 6.1I
Malaysia 2.41 -9.79 -0.31 '.02
Pakistan 2.42 10.92 30.00 1.21 -1.57
Philipines -3.42 -1.69 2.09 -8.61 -8.61
Sri Lanka -2.49 -9.13 -8.73 -7.03
Thailand 4.51 10.79 10.72 2.07 2.07
Region 4.32 -1.73
Source: FAO Trade Yearbook

Table 8
Total Agricultural Exports Growth Rates -- Eastern Europe & Middle East
Country 1970-90 1970-72 1978-81 1983-90 Post
Algeria -15.21 -26.08 -11.52 -7.51
Egypt -6.97 -6.73 -3.07 -11.93
Hungary -0.88 10.85 -1.09 -3.97 2.53
Morocco -3.44 6.82 -5.13 3.09 2.77
Poland -2.40 14.18 -20.06 5.80
Romania -6.61 21.66 -5.36 -17.99
Tunisia -3.25 46.51 -4.64 4.03 4.91
Turkey 2.26 12.38 8.67 -1.81 -2.12
Yugoslavia -0.17 1.43 11.69 -5.79 -5.07
Region -4.79 -4.00
Source: FAO Trade Yearbook

Table 9
Total Agricultural Exports Growth Rates - Latin America
Country 1970-90 1970-72 1973-77 1978-81 1983-90 Post
Argentina 0.14 -6.49 0.90 3.36 -6.04 14.58
Bolivia -1.51 41.23 5.15 -17.56 21.62 10.23
Brazil '0.45 10.59 4.88 6.65 -6.41 -7.45
Chile 8.99
Colombia -11.93
Costa Rica 0.84 2.24 11.66 -6.79 -0.56 -2.23
Dominican Rep. -4.58 11.24 1.97 9..35 -10.98
Ecuador -2.64
Honduras 2.61 0.27 13.88 -0.69 -1.95
Jamaica -2.94 0.94 -1.84 -11.16 -0.62 1.32
Mexico -0.40 4.01 n.34 -8.18 3.60 -0.23
Panama -2.37 -1.99 -2.09 4.24 -5.69 -5.36
Peru -5.11 -3.44 -1.18 -19.65 -2.58
Venezuela 0.05 6.36 5.02 -14.52 6.45
Region -5.04 0.33
Source: FAQ Trade Yearbook

Number of Publications on Adjustment 1984 to 1992
Africa Non-Africa
Agriculture 11 5
Poverty 16 22
Macro 32 112
Total: 1%8

Table ii: Developing Countries: Public Sector Investment as % of GDP
1985-87 1988-90 1985-87 1988-90
ALL COUNTRIES 8.6 7.9 8.7 8.2
AFRICA 8.8 8.1 7.9 7.3
ASIA 9.0 8.6 9.8 9.6
MIDDLE EAST 9.6 7.7 8.5 7.4
WESTERN HEMISPHERE 7.8 7.3 8.1 7.8
Source: World Economic Outlook, International Monetary Fund, May 1992.

Figure 1
16711 '13 '1675 1977 1g79 1961 M93 136 M37 19m
..AFR ICA -*-A
nmocMs PAO. lofto 6-MYw

Figure 2
World Bank Commitments to Africa
035 I I I I
7E& 719 77 78 75F 80 8?t 8= M 84 8as8 87 88 as 90 4- Ag~cuibr s gr. -A4 Lancar a6 C
Source: World Bank data.

Figure 3
World Bank Commitments to Asia
M 6 7 &7 08 B 3 8 5BS8 9 9
2 ~ ux r A4U-gA Ti

Figure 4
World Bank Commitments to EMENA
j h
0T T I T I I I I I I I I 1
75 76 77 78 79 80 91 82 83 84 85 86 87 a88 9 90 4- *dqcubu. 0 9r. A* LAding A Tdd
Source: World Bank data.
Note: EMENA refers to Europe, Middle East, and North Africa.

Figure 5 World Bank Commitments to Latin America
W 97-9
75 76. 77 78 9 80 81 82 83 84 815 86 897 88 89 90 4- rcuftr w k. 4 Lwdn A Tdd
Source: World Bank data.

Figure 6 World Bank Commitments To All Regions

7&- 76 77 781 79- 80 61 812 80 84 MIS 816 837 88 8 90 4- Ag~kulbw 0 Si. A4 Laudinq A Tdd
Source: World Bank data.