Singh T GREA ON wm "1as I tThe Rural Poor in South Asia U,INDERJIT SINGf" 9025 HOPKINS AWrdBn uUafm z The Great Ascent A World Bank Publication I The Great Ascent The Rural Poor in South Asia Inderjit Singh Published for the World Bank The Johns Hopkins University Press Baltimore and London Copyright © 1990 by The International Bank for Reconstruction and Development/The World Bank 1818 H Street, N.W., Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing October 1990 The Johns Hopkins University Press Baltimore, Maryland 21211, U.S.A. The findings, interpretations, and conclusions expressed in this study are en- tirely those of the author and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Ex- ecutive Directors or the countries they represent. The maps that accompany the text have been prepared solely for the convenience of the reader; the des- ignations and presentation of material in them do not imply the expression of any opinion whatsoever on the part of the World Bank, its affiliates, or its Board or member countries concerning the legal status of any country, terri- tory, city, or area, or of the authorities thereof, or concerning the delimitation of its boundaries or its national affiliation. The textile pictured on the cover shows rice growers in an Indian village. It is one of more than 250 applique and embroidered pieces produced by Les Ateliers au Fils d'Indra in Pondicherry, India. Founded in 1968, the workshop provides employment for women and preserves embroidery techniques and styles that are centuries old. The artwork is distributed in Western Europe through nonprofit associations. Library of Congress Cataloging-in-Publication Data Singh, Inderjit, 1941- The great ascent: the rural poor in South Asia I Inderjit Singh. p. cm. Includes bibliographical references and index. ISBN 0-8018-3954-8 1. Rural poor-South Asia. 2. Agricultural productivity-South Asia-Regional disparities. 3. South Asia-Economic conditions- Regional disparities. I. Title. HC430.6.S59 1990 338.954'09173'4-dc2O 90-5122 CIP Contents Foreword by Michael Lipton ix Preface xv Acronyms and Abbreviations xxiv Maps India xxvi Pakistan xxvii Bangladesh xxvii 1 Introduction 1 2 Rural Poverty: Incidence and Trends 8 Patterns of Poverty 9 Trends in Agricultural Output and Incomes 18 Agricultural Growth and Poverty 26 Agricultural Wage Rates 29 Who Are the Rural Poor? 35 Conclusions 40 Notes 41 3 The Agrarian Structure 47 Two Theories of Agrarian Change 48 The Structure of Ownership Holdings 51 The Structure of Operational Holdings 62 Landlessness 72 Rural Wage Labor 77 Rural Unemployment 81 V vi Contents The Occupational Distribution of Rural Wage Employment 87 Conclusions 93 Notes 96 4 Prerequisites for Sustained Gains in Output on Small Farms 100 The Small-Farm Sector 100 Better Cultivating Practices: The Training and Visit System 108 Scientific Multiple Cropping 115 Irrigation 128 Access to Credit, Markets, Research, and Extension 135 Notes 151 5 The Rural Poor and the Green Revolution 155 Patterns of HYV Adoption 157 Yield Gains and Nutrient Use 165 HYvs and the Landless 176 Distribution of Gains from the Green Revolution 197 Conclusions 200 Notes 201 6 Noncrop Employment for the Rural Poor 204 Livestock and Dairying 205 Small Stock Development 226 Poultry 227 Fisheries 228 Forestry 229 Conclusions 231 Notes 231 7 Rural Industrialization and Employment 233 Employment Linkages 233 Rural and Small-Scale Industrialization 244 Rural Public Works Programs 259 Specialized Poverty Programs 271 Notes 275 Contents vii 8 Land Reform: The Missed Opportunity 277 Tenancy as an Institutional Constraint 281 Tenancy Reform 291 Ceilings on Landholdings and Redistribution of Land 299 Consolidation of Fragmented Holdings 308 Conclusions 314 Notes 318 Appendix A: Caveats in Using Data on Poverty in South Asia 321 Note 323 Appendix B: Hypothetical Redistribution Schemes: Data Sources and Policy Issues 324 Land to the Landless First 325 Land to Marginal Cultivators First 329 Notes and Sources for Table B-1 329 Notes 332 Statistical Appendix: Tables Al through A45 333 Bibliography 377 Index 437 Foreword The rural performance since independence in India, Pakistan, and Bangladesh has inspired thousands of papers, books, and surveys. The flow has increased since high-yielding cereals were introduced in the mid-1960s. For scholars and administrators, the task now is to evaluate this work. Three things are needed for such a task, and Inderjit Singh's book brilliantly provides them. The first is a comprehensive, ordered guide to the literature. Singh thoroughly searches official series and research studies, neoclassical and Marxist approaches. Furthermore, he organizes them into a co- herent story with four central themes: the trends in rural poverty as they are affected by agrarian structure, landlessness, and unemploy- ment; the sources of growth for small farms, especially the role of technical progress; the impact of nonfarm employment and rural public works; and the scope and limits of land reform. The second requirement is a critical examination of the literature. It is necessary to weigh the worth, not merely measure the mass, of the research materials. This book does more than count the pieces of evidence on each side of an issue; it assesses their quality and con- siders the number of rural people to whom they apply. Third, a guidebook needs a clearly stated point of view. Only thus can it organize evidence or challenge readers to disagree intelligently. This book's viewpoint is that, although massive poverty persists in South Asia, economic growth has helped to ease it. "Where agricul- tural output per capita has been stagnant or declining," says Singh, "the living conditions of the rural poor have deteriorated," and where it has been growing at over 2 percent a year, all rural classes, even the poorest agricultural labor households, have gained (p. 41). Singh's optimistic view of what rapid farm growth can do must be set against his recognition that many rural South Asians, "unable to support themselves by farming because their landholdings are very .x x Foreword small or nonexistent, . . . seem increasingly to suffer from underem- ployment and the erosion of the real value of their earnings" (p. 93). Indeed, rural unemployment-while low overall-is highest among the poorest, is rising over time, and is concentrated in places and times of low wage rates and low work force participation (see my World Bank working paper, Labor and Poverty). Many underemployed people are poor not only because their landholdings are inadequate, but also because the "dramatic expansion of [nonfarm incomes] has been confined . . . to areas in which agricultural growth has been high" (p. 93). Thus Singh's main recipe for reducing rural poverty in South Asia is to promote fast growth of output on small farms. Because such growth is usually needed for farm and nonfarm employment, both poor farmers and even poorer rural employees benefit. Singh, how- ever, is worried that concentrating on small-farm growth, so desirable on equity grounds, may sometimes damage efficiency. He argues that the "inverse relationship"-the tendency of small farms to show higher yields per acre than large farms-may be reversed by the Green Revolution (p. 101). Fortunately, growing recent evidence sug- gests that once cultivation with improved seeds settles down, small farmers once again do better (as R. Longhurst and I argue in New Seeds and Poor People, pp. 142-44). Using data from Bangladesh and Pakistan, Singh also shows that small farmers are dramatically more likely to double-crop than large farmers. Moreover, the gap has been widening (p. 117 and tables 4.9- 4.10). "Multiple cropping promises to be more effective in creating [rural] employment and income . . . than either rural industries or rural work programs" (p. 128). It depends critically on improved water control (including drainage) and further research into appro- priate short-duration, and hence drought-avoiding, crops and varie- ties. What of the problems small farmers have with rural credit and marketing? Singh does not fully review the evidence of bias against small farmers in markets. He does note that India's commercial banks have served farmers better since nationalization, but other institu- tional agencies still neglect smallholders (p. 137). Yet farmers opera- ting less than two hectares are much likelier to get credit from a bank or a coop, instead of a moneylender, if they use high-yielding seeds than if they do not (p. 142). This finding lends support to the view that small-farm growth in South Asia need not face a credit con- straint, provided that in areas where such a constraint might indeed bite, there is no dogmatic objection to modest, carefully managed credit subsidies. South Asia's small farms are at least as efficient as big ones-largely Foreword xi because they employ more labor per acre-and, if growth prospects are attractive, small farms can often overcome credit and marketing biases against them. How, then, have small farmers been affected by the new cereal varieties? In a few areas these varieties have brought a veritable revolution in output (for example, by doubling wheat yields and almost tripling rice yields in the Indian Punjab between the mid-I 960s and the early 1980s). Small farmers' adoption of these varieties "did lag, but . . lags diminished rapidly" (p. 159), although sometimes the lagging, poorer adopters then faced a cost-price squeeze, in contrast to the economic rents enjoyed by earlier and better-off innovators. In regard to the use of fertilizer, however, small farmers may fare even better than is suggested here (p. 168); in Southeast Asia they use at least as much inorganic fertilizer per hect- are as big farmers do (Herdt and Capule 1983). Mechanization and eviction sometimes erode the small farmer's gains from new seed technology and may even reverse the employee's. Singh argues that the gains "arise more from the agrarian structure ... and from the economic policies" than from the new varieties them- selves (p. 177). But a more socially conscious approach to farm re- search would steer it toward varieties-and also crops, seasons, tech- niques, and environments-where the interaction with such structure and policies was relatively likely to favor the poor. Singh argues that the Green Revolution has pushed up wage rates, kept down food prices, and hence favored the poor as workers and consumers (pp. 176, 185, 195). The poor would certainly be worse off without the new seeds, but these particular benefits are precari- ous. A poor and landless work force, especially one that is growing fast because the population as a whole is rising, readily moves toward areas with a higher demand for labor. As a result, employment may increase, but even with a Green Revolution wages may not rise signifi- cantly over the long term. If food prices are restrained, the ready availability of labor means that employers can restrain wage increases. Partly to boost the demand for labor against such eventualities, Singh argues for the Chinese approach to rural industrialization-"upgrad- ing village crafts by transforming their technologies [rather than preserving] outmoded and stagnant production methods"-together with "medium-term relief to the unemployed" through rural works, along the lines of Maharashtra's Employment Guarantee Scheme (pp. 256, 259). Yet, if both farm and nonfarm employment-and hence poor peo- ple's incomes-depend on agricultural growth through transformed technology, and if small farms are at least as good at attaining such growth as large, then is there not an overwhelming case for land re- form? Singh rightly rejects reform based on restrictions on tenancy. xii Foreword There is little tenancy anyway, and it is declining. Some of it entails rentals by the poor to the rich; the rest often helps the poor and land- less by allowing them to "sell" their managerial skills for usufruct claims (pp. 291-92). Anyway, can such reforms be imposed on pow- erful landlords, who enjoy increasing competition among tenants to rent land? Given the rural "balance of economic and political power . . . further tenancy reform is useless," Singh concludes (p. 299). "It is poverty that leads to sharecropping, and not sharecropping that causes poverty" (p. 287). Yet Singh accepts the strong case for redistributing owned hold- ings, though he does so more for reasons of equity than in confident anticipation of productivity gains. "The 'structuralist' viewpoint ... that the only long-run remedy to the problems of rural poverty is ... radical land reform ... is correct [because] the numerous [other] remedies ... may not be enough.... The case for redistributive land reforms is further strengthened by the evidence . .. that smaller hold- ings could enhance employment and [cropping] intensities" (pp. 288-90). Land redistribution has in fact achieved little, though more than Singh suggests here (see my World Bank working paper, Land Assets and Rural Poverty, p. 28). Singh contends that "a relatively modest re- distribution program could usefully give the landless plots for simple homesteads and space for small-scale vegetable or livestock farming" (p. 307). The experience of Kerala supports this approach, but also reveals the political obstacles (described by Ronald Herring in Land to the Tiller). Singh makes a convincing and thoroughly documented case that "the great ascent" of South Asia's rural poor, though patchy and slow, has indeed begun. It is most noticeable where agricultural growth is really rapid (such growth does indeed "trickle down," at least to the moderately poor) and where, despite unsatisfactory growth, the poor are both literate and organized, as in Kerala. The book argues force- fully for labor-intensive, agriculture-based technical progress to bene- fit the poor. Rural employment programs also have a role. History strongly suggests two further requirements. One is much more politi- cal organization among the rural poor. The other is a shift in the pat- tern of research and technical change. High-yielding cereals were in some ways ideal innovations for the rural poor: divisible to the level of the smallest farm, generally reduc- ing risks of moisture stress and (at least in the short run) pest attack, permitting the buildup of cereal reserves, moderating the price of food, and boosting farm employment. But this last effect-in- creasingly important because a growing proportion of poor rural people's incomes is earned from hired labor-has been subverted by Foreword xiii tractors, threshers, weedicides, mechanical rice transplanters, even combines. New technology can be progressive, but subsidies should not selectively encourage labor-replacing inputs, such as fuel or credit. Agricultural research should concentrate on achieving advances that productively demand rural labor in neglected areas (those with- out reliable irrigation) and for neglected crops (those eaten princi- pally by the poor). The Indian national crop research programs and the Hyderabad-based International Crops Research Institute for the Semi-Arid Tropics (ICRIsAT) have shown what is possible, by making progress on two crops that very poor people eat and grow labor- intensively: ragi in Karnataka and hybrid sorghum (in the main mon- soon season only) in Maharashtra. Progress is much slower for millets, winter sorghum, and root crops and for the densely populated, ill- drained rice areas of eastern India and parts of Bangladesh and Ke- rala. Two other issues face agricultural researchers. First, how can the emphasis on poverty be reconciled with the need to sustain long-run agricultural productivity, especially to reduce soil erosion on mar- ginal lands and to avoid groundwater depletion almost everywhere? Poor people, more of them all the time, need income and food now. The short-run response is to find methods and incentives to substitute employment for environment: using fertilizers in root-zone placement (mudballs) and other ways of reducing the agrochemicals required to achieve a given output; cross-bunding and micromanagement of water; contour plowing associated with vetiver grass; appropriate im- provements to terracing; and establishing research priorities, by crop and region and varietal selection, to improve the private returns to such methods. The long-run response has to involve slower population growth. In Kerala-with its disappointing agricultural performance-the im- proved health of children and the dramatic drop in mortality have been accompanied by such a rapid decline in fertility that the net re- production rate has fallen almost to the replacement level. Elsewhere in India, including the Punjab, total fertility has fallen more slowly, especially among the poor. It is verv important to focus research- and subsequent action-on the long-run effect on fertility of alterna- tive paths of agricultural progress and health provision. The thesis of this book-that fast small-farm growth is the best cure for rural poverty-raises a second issue for agricultural researchers. Simply stated, it is that the potential yield of major tropical and sub- tropical food crops must increase significantly. Rice has not done so since the IR-8 variety was released in 1963; for wheat, the rate of in- crease is only about 1 percent a year. The green revolution has there- xiv Foreword fore increasingly required farmers to narrow the gap between field performance and nearly stagnant yield potential. But such narrowing eventually becomes uneconomic or too risky, unless researchers are driving that potential upward. Field performance is itself eroded by new pest biotypes and by the depletion of micronutrients and ground- water. By exploring the effectiveness of rapid small-farm growth in re- ducing poverty-and the scope of social action to improve that effec- tiveness-this splendid book should capture the attention of policy- makers and researchers. Researchers, however, tend to do what they have previously done well. Policymakers tend to overstress fashion- able issues of aggregate economic management. It will take a major effort to focus both groups, once again, on labor-intensive technical progress for the main staple crops of the rural poor. Michael Lipton Preface After World War II the economies of the developing world were pre- dominantly agrarian. To improve living standards and eliminate rural poverty, rural productivity had to increase. This was a goal of all de- veloping countries, and in South Asia-India in particular-it be- came the central one. (Throughout this book "South Asia" includes Bangladesh, India, and Pakistan.) In the early 1950s and 1960s, agriculture was seen as tradition- bound and stagnant, contributing little to national development. De- velopment plans therefore placed heavy emphasis on industry. The Mahalonobis plan, for example-which was based on the Soviet devel- opment experience and on the blueprint for India's second plan (1956-61)-stressed capital-intensive investment in basic industries to reap the productivity gains from technology. Farming was neglected: it received few resources, and its output was seen mainly as feeding industrial growth. Such thinking continued to underpin development theory through the early 1960s. A fundamental change came with the breakthrough of high- yielding varieties of cereal (HYvs) in the mid-1960s. Norman Borlaug was awarded the Nobel Prize for developing high-yielding varieties of maize in Mexico, and new adaptive wheat and rice varieties soon followed. Pioneering research organizations, such as the Interna- tional Rice Research Institute (IRRI) and the International Center for Corn and Wheat Improvement (CIMMYT), were set up under the Consultative Group on International Agricultural Research. Their technical innovations soon found intellectual support from econo- mists. Theodore Schultz won the Nobel Prize for his theory that the economic potential of agricultural development had been largely un- exploited and that peasant agriculture could be highly responsive and productive. The bulk of poor farmers then came to be seen not as inefficient but as constrained by distortions introduced by the govern- xv xvi Preface ment; that is, by interventions to keep food prices low for politically important urban dwellers and industrial workers. The ability of the agricultural sector to become an engine of growth then became devel- opment orthodoxy. Without major productivity growth in agricul- ture, economists began to argue, efficient industrialization would be unlikely. The initial neglect of agriculture meant that by the mid-1960s pop- ulation growth was outpacing growth in agricultural productivity, and South Asia was growing more dependent on massive imports of food aid. During the late 1960s one ship a week carried grain from the Mississippi Delta ports of the United States to India, which re- quired more than a million metric tons a month. Projections of the combined future import demands of Bangladesh, India, and Pakistan became enormous, and their agrarian takeoff seemed increasingly distant. At the same time, the Indian government realized that it would have to halt the growing dependence on foreign food aid and imports if it were to maintain political sovereignty. Given the combi- nation of HYV technology and political imperatives, India embarked on the massive task of creating the research, extension, irrigation, and marketing infrastructure necessary to deliver the new HYV seeds, water, and fertilizers to farmers. By the late 1960s a rapid agrarian transformation had started in South Asia. But bureaucratic and intellectual perceptions of the situa- tion continued to lag. As late as the mid-1970s international institu- tions were forecasting that India would need 15 million to 18 mil- lion-even 20 million-tons of food imports a year in the 1980s to feed its burgeoning population. Far from draining the supplies in in- ternational markets, however, India was already starting to become self-sufficient, and by the early 1980s it had 20 million tons of food- grain surpluses. Instead of being a food deficit country, India was able to build a growing volume of food stocks. Pakistan had in the meantime also b.come self-sufficient and started to export food. And Bangladesh, despite its late start and the horror stories that suggested the need for triage, had started on its way to self-sufficiency. At the heart of these dramatic changes is the transformation of the lives of 800 million people in the rural areas of South Asian coun- tries-the "great ascent" from poverty and despair to a more materi- ally rewarding existence. If increases in agricultural productivity and output are a prerequi- site for industrial development and the great ascent, these countries have taken the first steps. But this very success has led to other prob- lems: Most of the new productivity was concentrated in a few parts of the subcontinent, which gave rise to distinct regional disparities. Three broad belts of differentiated development have emerged: Preface xvii * Northwest India and Pakistan, including the states of Gujarat, Ha- ryana, Kashmir, Punjab, western Uttar Pradesh, parts of Raja- sthan, and Pakistan. In this region productivity and output have grown faster than the population. * The central "Hindu heartland" of India and Bangladesh, including An- dhra Pradesh, Bihar, Madhya Pradesh, Maharashtra, eastern Uttar Pradesh, West Bengal, parts of Gujarat, and the eastern states of India. In this region population has grown at a higher rate than productivity. * South India, including Karnataka, Kerala, and Tamil Nadu. This region has increased its agricultural productivity but barely enough to meet the growing population demands. A core of progressive states in the Northwest has thus been generat- ing the surpluses in the subcontinent. By the mid-1970s, 65 to 75 per- cent of the total food-grain stocks of the central government of India was coming from two states, Punjab and Haryana, which have only about 5 percent of the agricultural area. At harvesttime India has to divert all its freight trains to these states to transport the millions of tons of food to central depots and then redistribute them to the deficit states. It has set up a central mechanism for state purchasing, storage, and distribution, with fair-price food shops to sell food grains at sub- sidized prices. The political implications of these changes have been dramatic. A political minority is producing food surpluses for the po- litical majority. The large deficit states in the central heartland of India generate 60 percent of the the votes and 70 percent of the mem- bers of parliament, and yet they depend on states at the periphery (Punjab and Haryana) that are often in the hands of opposition par- ties. Interstate disparities have also widened because of differences in endowments and commitment to the agricultural sector. Provinces with water, progressive agricultural policies, and programs for exten- sion services and irrigation did well. (Agriculture in India is organ- ized and funded by the states, not the central government.) In short, the states that paid attention to their rural constituencies and to agri- culture got ahead, while others lagged. Bengal, eastern Bihar, and eastern Uttar Pradesh have been left behind because their state legis- latures failed to provide the funds for irrigation, research, extension, roads, and agricultural support services. A similar dualism sprang up in the villages. The new HYV technol- ogies favored the landed over the landless, who were unable to raise credit and buy the new seeds and fertilizers. A new fragmentation of the rural economy replaced the traditional jajmani system of shared tasks and shared rewards. Landowners with large or medium-size xviii Preface holdings produced the surplus, those with smaller holdings were rela- tively self-sufficient, and the rural landless and tenants became the deficit group, dependent on the surpluses of the others. Because the unequal distribution of incomes is an initial condition, unequal rewards prevail in the markets. The real tradeoff, according to A. K. Sen, is between the value of food and the value of labor. A poor landless laborer's "entitlement" depends on his ability to market his labor (in excess supply) and his ability to buy food (in short sup- ply). With the changes in South Asia's rural economy, the entitlement a laborer received for one hour of labor dropped, so that some of the rural poor became even poorer. As a result, the idea developed that government intervention was needed to change the market tradeoffs between labor and food. Governments could correct the market imbalances by setting up special services for the poor-cheaper credit, food shops, rural works programs, and so on. But governments are no more neutral than markets. Governments in South Asia responded to the vested inter- ests of the landed classes, and they were not about to transform the terms of trade to favor the rural poor. Pakistan, for example, started few direct programs to reduce poverty because its policy formulation was dominated by elites-from the army, the bureaucracy, or the landed aristocracy. In India the poor had no money but at least they still had the vote. The middle-class farmer began to form a significant political bloc of votes by the late 1970s and early 1980s, and the government set up special programs for the rural areas. But those who benefited from these antipoverty programs were usually those who had the grass- roots power. Deciding who was "poor" and therefore eligible for aid depended on the administrators of these programs, generally the rural elite. So, rather than being a neutral agent for the distribution of gains, government became a trough where different social groups fed in relation to their economic power. Many antipoverty programs in India-such as the Drought Areas Program, the Rural Works Pro- gram, and special credit programs-delivered the greater share of their benefits to the richer, not the poorer, segments of society. As a result, the Green Revolution decades were marked by the pes- simistic view that growth was making the poor poorer and the rich richer. The failure of both markets and governments to deliver gains equitably fueled a whole series of articles in the 1970s. In response to the debate about the effectiveness of the new tech- nologies, this book follows two lines of inquiry. It first questions whether the changes from the Green Revolution have really failed to help the rural poor. Is it true that growth has been impoverishing? Have the rich benefited at the expense of the poor? Or is it more cor- Preface xix rect to say that the rich have become richer, but the poor are also bet- ter off? Second, this book looks at the other side of the coin. The ef- forts may not have succeeded as well as planned or as hoped, but what would South Asia look like without these changes? Imagine India still dependent on U.S. food aid and imports with twice the population it had twenty years ago. What would Indian politics be if the govern- ment had not set up programs to try to overcome some of the faults of the markets and had failed to tackle some of the faults in those programs? The answer is obvious. The poor would have been far worse off today, and there would have been political and economic chaos. The results of these two lines of inquiry support a series of genera- lizations: * Growth has not been impoverishing but has been essential in re- dressing poverty. * Inequality has increased, but the increase has been exaggerated, as have the problems. * The Green Revolution has had major benefits, especially in view of the alternative. D Tenancy is not as important a barrier to developmental change as has been assumed. * The real problem is not unemployment, but productive employ- ment; that is, how much a laborer can get in return for one hour of work, whether employed or not. D Nonfarm and noncrop income are just as important as land for transforming the rural economy. * The growth of the nonfarm economy depends on the vitality of the farm economy; without agricultural growth in the rural areas, redresssing poverty is an impossible task. * Rural work programs can succeed, especially when tied to a well- managed grain distribution system (the Maharashtra Employ- ment Guarantee Scheme is an excellent example). * Ancillary industries that tie rural industrialization to the develop- ment of rural areas are important but too often neglected. * Land reform, though critical, is no panacea, and even without land reform, progress is still possible in other areas. * Institutional reforms and policy changes are needed, but much has already been done to develop the appropriate institutions, such as research universities and extension agencies. * South Asian countries cannot tackle the problems of poverty without considering the future of the industrial sector. Because xx Preface agriculture cannot continue to absorb the continuing population increases, there must be rapid industrial growth, which undoubt- edly will in turn become a source of problems. The two lines of inquiry in this book converge on a larger question: has South Asia started on its great ascent? The first few steps have clearly been taken. And just as clearly, more problems will arise. In development the solution of one set of problems, seemingly simple, leads to a set of more complex problems. In addition to the foregoing problems in agriculture, four problems outside agriculture require further attention. First, there must be a policy of population control. India's popula- tion growth has slowed to 2.2 percent a year, but Pakistan and Bangla- desh are still growing at 2.6 and 3.1 percent a year respectively. In- dustrial and technical progress cannot keep pace. Even in India, agricultural productivity barely keeps up with population growth, which remains a driving mechanism for perpetuating poverty. Second, without education the poor will neither demand new pro- grams and new technologies nor have the ability to respond to those opportunities. China, Japan, and the Republic of Korea owe their suc- cess to an educated and informed population. Without an educated peasantry, long-term development in South Asia will continue to lag. Third, an industrial revolution is needed to absorb the labor from the transformation of the agrarian sector. If employment in the in- dustrial and service sectors is not forthcoming, the gains from agricul- tural growth will collapse dramatically. As 200 million to 300 million people in South Asia move into urban and urban-related activities, ways must be found to organize the industrial sector to provide pro- ductive employment, income, and higher living standards-the major challenge for the decades ahead in South Asia. Fourth, researchers and administrators need to change their per- spective. Their pursuit of the best solutions may make it difficult to carry out good solutions. There may be ten good programs that will help farmers, five that will help small farmers, and only two to help the landless. Finding one that helps only the poor and not the rich is rarely feasible, and we should not throw out the good because of a feckless search for the best. Patience is also needed. The transfor- mation from agrarian to modern industrial state takes generations. What Europe did in twenty generations and Japan did in ten the de- veloping world is trying to do in five. So far, this experiment in South Asia has had only forty years. Now the third generation is laying the foundation for another two generations beyond us. South Asia may not see the last steps of the great ascent until the middle of the 2000s. But that would be a tremendous accomplishment-in a hundred Preface xxi years to go from a traditional, feudal, agrarian economy to a modern welfare state. In addition to eschewing perfection and haste, researchers must avoid oversimplification. As the book points out, there are nuances within nuances. For example, the division between the landless and farmers is complex: some farmers spend time as wage earners, and some of the landless have access to land. Simple breakdowns and sim- ple notions do not help much. To say that poverty has either increased or decreased is an oversimplification-especially when considering economies the size of India's, with hundreds of millions of people and some 600,000 villages. Another change in perspective is that government administrators and nongovernmental organizations should not try to do everything for the poor. More is accomplished when people take command of their lives and find their own solutions. Administrators should under- stand that and allow more room for it to happen. Their job should be to enable people to help themselves-to find out what people want to do with their lives and give them the means to do it. Three additional areas of inquiry that have some bearing on the lives of the rural poor are not adequately addressed in this book. First is the importance of arid agriculture and its prospects, since many of the rural poor live in arid and semiarid areas of South Asia. Al- though experimental research has reported some initial progress in raising productivity in arid areas, big breakthroughs in application are yet to come. When they do, a further assessment of their impact on the rural poor will be needed. Second, how to provide the rural poor with greater access to credit, water, knowledge, and other ser- vices remains a question for further research. Significant work is being done, however, both to make markets operate more effectively and to develop new institutions to overcome market biases. A full treatment of this topic would require a separate undertaking. Third, there has been considerable growth in the number of programs de- signed to assist the rural poor directly, especially in India. Their cov- erage, effectiveness, and shortcomings constitute another important area of inquiry, and the considerable amount of work that is being done needs to be assessed carefully. Policymakers and administrators should know that sharecropping is not bad, that risk is not an ultimate barrier to adoption, that credit systems can be made to work, that markets generally operate better without interventions, that crop insurance is costly and generally not feasible, that price support systems have given tremendous stability, that extension has worked, and that rural works programs and dairy- ing can succeed. International agencies, which formulated many of the schemes discussed in the book, will find a wealth of project-related xxii Preface solutions, drawn not only from the recent academic literature but also from the World Bank's operational experience. Researchers will find that the theories and development notions prevalent in the 1950s need to be reformulated. This book argues for a new theoretical para- digm, perhaps a more complex paradigm to replace our earlier and more naive concepts about rural poverty. This book stems from a paper written for the 1978 edition of the World Bank's World Development Report. After subsequent research in 1979 and the first half of 1981, I compiled a 1,200-page draft, which for a variety of reasons had its publication repeatedly delayed. This manuscript has now been updated with new empirical evidence for the intervening years and has been cut in half. The original chapters on tenancy, agrarian structure, and the productivity of small farms have been published separately (Singh 1988a, b, and c) and are merely summarized here. Shankra Acharya, Montek Ahluwalia, D. C. Rao, and Martin Wolf initiated some of the early ideas for this work and reread my drafts; without their support the book would not have been written. My thanks also go to Robert McNamara and Ernest Stern, who supported my efforts to produce a book from the original research. I would also like to thank Bevin Wade, chief economist for India at the time, who insisted that I incorporate a wider research literature and stressed the importance of the book. Mark Leiserson, my former division chief in the Agriculture and Rural Employment Division of the Development and Economics Department, has been an unfailing source of moral support throughout all the vicissitudes. Even after the abolition of that department, he constantly asked how the book was coming and urged its publication. My thanks also go to my three directors-Ben- jamin King, Gregory Ingram, and Ardy Stoutjesdijk-who provided invaluable support and research assistance even when I was trans- ferred to other departments in the World Bank. I am indebted to my editor, Peter Bocock. He produced the first organized draft and the revised drafts under great personal pressure and strain. A special word of thanks goes to Tribhuwan Narain, who worked with me on the first paper for World Development Report 1978, traveled with me to South Asia for three months, visited hundreds of universities and researchers, helped ship and compile the data, and sorted through some 6,000 articles. He undertook the work on all the tables, and even though subsequently employed elsewhere in the Bank, he has never failed to come back and help update the final copy. I would also like to thank the hundreds of researchers throughout South Asia for giving so generously of their time and their latest re- search work. In addition, I am grateful to Bruce Ross-Larson and Mi- randa Elgin for helping me produce the final version under extreme Preface xxiii pressure, and to Jane Carroll for copyediting voluminous material with a fine-tooth comb. Finally, I want to thank my family for their support and humor through eight long years of my bringing this manuscript home to fin- ish-"one more time." Acronyms and Abbreviations ADB Asian Development Bank ADBP Agriculture Development Bank of Pakistan AERC Agro-Economic Research Center BARC Bangladesh Agricultural Research Council BIDS Bangladesh Institute of Development Studies CAD command area development CIMMYT International Center for Corn and Wheat Improvement CSRE Crash Scheme for Rural Employment DDP Desert Development Program DPAP Drought Prone Areas Program EGS Employment Guarantee Scheme FMS Farm Management Surveys FWP Food for Work Program GDP gross domestic product HYV high-yielding varieties IDA International Development Association IFDC International Fertilizer Development Center IFPRI International Food Policy Research Institute ILO International Labour Organisation IRDP Integrated Rural Development Program IRRI International Rice Research Institute MFLA Marginal Farmers and Landless Labor Agencies MNP Minimum Needs Program NCAER National Council of Applied Economic Research NDDB National Dairy Development Board NDP net domestic product NREP National Rural Employment Program NSS National Sample Survey NWFP Northwest Frontier Province PIREP Pilot Intensive Rural Employment Program xxiv Acronyms and Abbreviations xxv PRS personal responsibility system RIP Rural Industries Projects SFDA Small Farmers Development Agencies T&V training and visit TIP Thana Irrigation Program TVE township and village enterprise USAID U.S. Agency for International Development WAPDA Water and Power Development Authority °t PAKISTANv XlgA N, * ~~~~~Towns BANGDtiSH.j (Elr National capital /.~~~~~ ~~~ 2?0C 40 0 0 _,--\ CHINA / 9,HARYANA Mafrgr, , Hiskar-h k I;N-Whi .- < r~., AIigarh ^ NEPAL \ IKi g~~~~~~~~~iu UTA PRADESH .13, __-IBHUTAN RAJASTHAN aprr UTRP'DS si , ) Z k ~~~~~~~ ~~Luckn-w .f* iSA ^ g ,6 ~~~~~~Allaab.d. V-rn-ti flMEGAvLAYA _ - , S S , , ,r l vKo I-B G HyAaR i\, ~ GUJARAT C- J-~S Bo; rs ANGLA- 0 irS Ahnnedabad \.> Bho~~~Ap.l1 BENGAL .DESH MAHYx PADS _- U.S.S.R. s- ! -CH N PAKISTAN _ K -N * Towns and cities ® National capital NORTHWEST) Province boundaries AFGHANISTAN FRONTIER /.. ~~/ -~~-Appwx L-n International boundaries P _e or PROViNCE J (C-tn * kel--- o 200 400 Jhelun ; KILOM1ETERS Giowl . t * ~~( J D n Khur, Faraiba 2 "'Fort Sandeman l jLy[lpiu) /" * uerea PUNJAB S w \ ~ ~~ ~ ~ ~~~ ~K.J.t R.hhiy.r BALUCHISTAN Kha / ISLAMIC REP. } Khuzder '-iobebed N OF IRAN 7 ' *Lark... / SIND I * Hyderabad - R \ ~~~~~~~~~~~INDIA Arabian Sea i '- f>' ' - t._S.BANGLADESH INDIA ( W- Rangpur ® INDIA rty National capital -i ----- District boundaries *.. 2 r_ - -- Internationol boundaries jamapur, t- Bogra Jmlu <¢ Rajshahi - r4-Mymensingh 0 Syl0et < -- ;_ 1 T > - t j ; O i oo~~~~~~~~10 200 IG$ ; KishareJan; ( KILOMETERS (K.sht e se| .Jessore iFarrciPer \ J ' j INDIA j Khulna ® v ro e? MYANMAR - .rBay oif Benygal iZ . ; ii Introduction No society can surely be flourishing and happy of which the greater part of the members are poor and miserable. -ADAM SMITH Since the countries of South Asia-Bangladesh, India, and Pakistan- gained independence, the reduction of mass poverty has been a fun- damental goal of their social and economic policies. Yet in spite of nearly four decades of effort, the number of people living below the poverty line (however arbitrarily defined) has continued to increase, especially in rural areas. Rural poverty is deeply rooted in the economic and social structure of South Asia, and many observers have contended that poverty can- not be alleviated without radical changes-particularly in the unequal distribution of rural assets and the power and privilege that go with them. True, rural poverty would have been easier to eradicate, and the gains from technological and institutional improvements would have been more widely distributed, if structural reforms had been un- dertaken soon after independence. But at a minimum such reforms would have included the redistribution of landed assets-and they would have been revolutionary. The nature of the polity in South Asia was such that fundamental changes of this kind were not under- taken, and the subsequent difficulties in tackling the problems of rural poverty are the historical legacy of that missed opportunity. The chances of such redistributive reforms being undertaken today are even slimmer; but they would not be a panacea even if they could be undertaken. In areas where redistribution is most desirable there is little land available to distribute. Demographic pressures have fur- ther subdivided holdings, increased landlessness and the number of small and marginal farmers, and made the problems of poverty more intractable than ever. l 2 Introduction It has been wrong to insist that little has been or can be done to reduce poverty without structural reforms-that since the "best" was not done no "second-best" alternatives will suffice. One of the main themes of this book is that, notwithstanding the pessimism expressed in various quarters, significant progress has actually been made in reducing the proportion of those living in poverty in South Asia. The changes have been slow and in many cases the trends have been hardly discernible; but they do exist, and they are in the right direction. The trends have been most visible where per capita growth in the agricultural sector has been high. Annual rates of per capita growth in excess of 2 to 2.5 percent, when sustained for a decade or more, have had a significant impact on rural poverty. Such growth has oc- curred in relatively few areas of South Asia, however. The relative neglect of growth as an explicit strategy for attacking rural poverty has been one of the consistent threads in the evidence reviewed throughout this book. The central message is that growth does matter, and that it is the lack of growth-or rather the overwhelming stagna- tion in large parts of South Asia until recently-that is responsible for the continuing poverty observed in the subcontinent. The lesson is clear: without vigorous and continued agricultural growth, poverty cannot be alleviated. Where growth has been vigor- ous and sustained, even the poorest have benefited-directly through increased income and employment opportunities in the agricultural sector and indirectly through second-round employment opportuni- ties generated in the rural sector and greater supplies of wage goods. Without growth in per capita terms, it is the poorest who will remain the longest in poverty and continue to suffer most from the degrada- tion associated with it. Experience with various projects and programs during the past four decades clearly indicates the types of investments that have spurred agricultural growth: ,Investments in irrigation, especially in semidry and arid areas where the returns are enormous and the potential impact on growth very large, or in flood control and bunding in areas with heavy rainfall where multiple cropping is possible * Investments in new technologies associated with the Green Revo- lution, together with associated investments in essential inputs (including the timely and adequate provision of improved seeds, fertilizers, information, water, markets, and credit) * Investments in physical infrastructure-roads, transport, elec- trification, communications, and marketing-especially with the spread of technologies for high-yielding varieties (HYV) of seed, Introduction 3 which increase the marketed surpluses of small farmers and re- quire greater labor mobility and specialization on farms * Investments in rural education that enhance the productivity of farmers and laborers by enabling them to take advantage of new opportunities * Investments in research and extension that allow the dissemina- tion and adoption of existing technologies and a continuing flow of new ones to small farmers * Investments that encourage crop intensification, multiple crop- ping, and shifts to higher-value crops, including fruits and vege- tables. The positive implications of the Green Revolution have been swamped in much of the relevant literature by the criticism that it has not provided the rural poor with assets and opportunities equal to those of their richer and more powerful neighbors (who are said to have garnered all the gains from the new cereal technologies). I show that much of this criticism is misplaced. The best bet for the poor throughout South Asia, and especially in the chronically poor regions in the subcontinent, is to acquire, adapt, and adopt these technologies as rapidly as possible. It is futile to labor the point that the gains from these technologies would have been even more widespread and more beneficial had rural assets been more evenly distributed. The hard fact is that they are not. But this is a political failure, not a conse- quence of any shortcomings in the new technologies. To continue to argue against the rapid extension of these technologies and of the in- stitutions and investments needed to disseminate them as widely as possible-or to make structural reforms a precondition of their extension-is to argue against good solutions because the "best" solu- tions are not available. Thus in much of the public policy debate on this issue, the best has become the enemy of the good. None of the above should be taken as an argument against the in- trinsic desirability of reforming South Asia's agrarian structures. This book unsparingly outlines the inequities and inefficiencies in the ex- isting situation. Where reforms are possible-and I show that they are in many areas of South Asia-they must remain on the urgent agenda of policymakers. The policy debate needs to shift its focus, however, from their desirability to their political feasibility. Where appropri- ate reforms are feasible, they should certainly be carried out. In order of priority and their likely impact on the rural poor, these include: . Redistribution of land-still the most effective way to change the prospects of those without land 4 Introduction * Consolidation of holdings-shown to be both feasible and highly desirable * Increasing the security of land tenancy-where there is the politi- cal will needed to implement and monitor such programs effec- tively. In addition to programs for accelerating per capita growth and land reform in the farm sector, complementary programs in the nonfarm sector are needed if overall prospects for the rural poor are to be improved. Those with the greatest potential benefits for the rural poor include: * Dairy programs, especially the cooperative arrangements first tried out at Anand and later extended through Operation Flood and the National Dairy Development Board in India, but yet to be tried in Pakistan and Bangladesh * Small stock programs, which have yet to be fully developed and extended on a national scale anywhere in the subcontinent * Small fisheries, whether private or cooperative undertakings in freshwater or seawater, which have yet to be fully exploited but offer considerable potential benefits for the rural poor - Poultry programs, which also have yet to be tried on a cooperative basis-the only way to make them accessible to the rural poor *Agroprocessing programs, again along cooperative lines, to bring together small producers of oilseeds, fruits, and vegetables; now being tried for oilseeds on the Anand model * Development of rural manufacturing and service industries, which would offer a host of opportunities for small-scale entre- preneurship and nonfarm employment for the rural poor-notably by providing goods and services to meet the expansion of rural demand once per capita agricultural incomes begin to rise. Of the initiatives listed above, only the dairy programs have been seriously tried on a national basis. Although there has been some de- velopment of rural industry, it has been mainly the result of individ- ual efforts prompted by and dependent on the rate of agricultural growth rather than a government plan based on conscious policy choices. But these are programs that can work-and when they have been tried they have had a significant impact on the rural poor. They need to be given a more explicit role in an overall strategy for alleviat- ing rural poverty. Opportunities like these are a necessary but not a sufficient condi- tion for the alleviation of poverty, however. All the evidence suggests that although growth is absolutely essential for reducing rural poverty Introduction 5 in the long run, it cannot be accelerated immediately in most of South Asia. Moreover, even if it were accelerated, it would take time to have a significant impact on poverty. Even if all the programs listed above were to be pursued vigorously, therefore, another decade or two would be needed for their impact to become substantial in many parts of South Asia. What is to be done in the interim? Public policy, espe- cially in a democracy, cannot afford to wait for the long run to ad- dress this issue. By the early 1970s it became obvious, at least in India, that direct antipoverty programs were needed to address the problem. Four cate- gories of programs were developed for different purposes. The first group was designed to provide direct benefits to individuals. It in- cludes the Small Farmers Development Agencies (SFDA) and the Mar- ginal Farmers and Landless Labor Agencies (MFLA), which were later supplemented by the Integrated Rural Development Program (IRDP). They enabled poor households to acquire assets (mainly through sub- sidized credit) that could generate both income and self-employment. A second group included the Crash Scheme for Rural Employment (CSRE), the Pilot Intensive Rural Employment Program (PIREP), and the Food for Work Program (FwP)-later restructured and called the National Rural Employment Program (NREP)-which provided wage employment in rural areas through public works. A third group was designed for the development of ecologically disadvantaged areas and included the Drought Prone Areas Program (DPAP) and the Desert Development Program (DDP). Fourth were those designed to raise standards of living through the direct provision of social services; an example is the Minimum Needs Program (MNP), which provided clean water, health care, and family planning. The most extensive of these programs are the IRDP and the NREP. The IRDP was started in 1978-79 to cover the areas in which the SFDA and MFLA had been operating and was extended to the entire country in 1980. It mainly provides subsidies and credit for purchasing live- stock and milk animals. Its outlays have grown very rapidly from about Rs4 billion in 1980-81 to about RslO billion in 1982-83, and it covers about 3 million families a year. In fact, the IRDP has grown too rapidly and has failed to reach the poorest rural groups. Because of political pressures it caters to many who are not poor and has therefore suffered extensive leakages. Since more than 260 million rural people were estimated to be in poverty in 1979-80 and the pro- gram reaches only about 9 million of them, it is generally considered inadequate in its scale and coverage. The Food for Work Program started in 1977 and was extended in 1980 to all of India under the NREP. From 1980-81 to 1984-85 the program created about 350 million additional days of employment a 6 Introduction year and executed a wide variety of useful public works projects, in- cluding afforestation, village water tanks, drinking water wells, minor irrigation works, rural roads, soil and water conservation, and land reclamation throughout India (Dantwala 1986). The program cost about Rs5 billion in 1984-85. Dantwala (1986) estimated that the outlays for all the direct anti- poverty programs together amounted to no more than about 1.5 per- cent of the net national product of the sectors in which they are sup- posed to eliminate poverty. In view of the size of the poverty problem, he rightly concluded that these programs are too small to tackle it ad- equately. He believes, again rightly, that most of the resources going to a variety of programs should be concentrated in the NREP because food for work programs provide employment directly to the poor, avoid leakages to the bogus poor, and provide directly productive works that are badly needed in rural areas. Such programs are also less subject to political pressure to benefit the rural rich, because the latter are mainly attracted to the credits and subsidies provided under other programs, such as the IRDP. The weight of evidence from Maharashtra's Employment Guarantee Scheme (EGS), reviewed in chapter 7, confirms this view. Indeed, the experience of the EGS spurred the Indian government to expand rural public works via the NREP. Although India has taken the first steps toward experimenting and gaining experience with direct antipoverty programs (such as the NREP) on a national scale, nationwide programs have yet to be taken up in Pakistan and Bangladesh. Most of the direct antipoverty programs are less than a decade old even in India, and it is too early for a full-scale assessment of their failures and successes. Excellent reviews of individual programs are provided in Desai (1979) and India, Planning Commission (1982, 1985a). The point stressed here is that programs of this kind are an essential complement to strategies to accelerate growth, and in the final analysis they depend on rapid agricultural growth to sustain them over time. One reason why India could experiment with the NREP is that it had a surplus of foodstocks-an estimated 25 million metric tons in 1985. It has always been recognized that the poor are in dire hunger be- cause they lack the necessary purchasing power, a view formalized by Amartya Sen in his seminal work (1981), which cited the lack of "enti- tlements" on the part of the poor as the main cause of famines in South Asia. Food for work programs such as the NREP bring together poor people who need work, and the surplus food grains to provide wages in kind, with essential public works that need to be put in place. These programs create direct entitlements for the poor, enabling them to exchange their work for food-the main item in their budget. Introduction 7 It is essential to underscore the point, however, that it is only be- cause of rapid agricultural growth in states such as Punjab and Haryana that the government has the surplus stocks of food that make a scheme such as the NREP feasible on a national scale. (It would be possible for Pakistan to undertake a similar effort, again thanks to rapid agricultural growth.) Without surpluses generated in fast- growing regions, direct national antipoverty programs-whether NREP or IRDP-would have to rely on food imports or food aid. Dependence on food imports would rapidly run into balance of pay- ments problems and stimulate inflation, which is just what happened when large food for work programs were tried in the 1960s. And Bangladesh would face these constraints now if it were to extend na- tionwide its small food for work program, which depends heavily on food aid. To summarize, a workable antipoverty strategy involves a three- pronged approach consisting of: (a) policy reforms and action pro- grams to accelerate growth in both farm and nonfarm rural activities; (b) a well-funded and effectively implemented rural work program aimed at the poor, with payments in food or, better still, money; and (c) supplemental but well-targeted programs in health, education, sanitation, family planning, and potable water supply to improve the quality of life in rural areas. These three elements-growth in the longer run, guaranteed employment in the short run, and improve- ments in the quality of rural life-should remain the central focus of policies for reducing rural poverty and its degrading impact in the coming decades. Meanwhile, the number of the landless and near landless will con- tinue to grow, as will the number of the rural poor-at least in abso- lute terms. Underlying demographic trends make it certain that pov- erty will continue to increase in areas where agricultural stagnation persists. But in areas where per capita agricultural growth accelerates, poverty can certainly be expected to continue to decline. In the long run, however, the inexorable pressure of rising population on rural land can be relieved only by reducing the rate of population growth and increasing employment opportunities in the industrial sector. The former is inextricably tied to the growth of employment oppor- tunities and the incomes and access to education (especially for women) that come with them. The heart of the problem is therefore the failure to transform the economies of South Asia by adequately expanding industrial employment opportunities. The reasons for this failure require a separate and detailed analysis, but it is certain that in the long run mass poverty-rural as well as urban-can be eradi- cated only by the industrial transformation of South Asia. On this path the economies of the region are now embarked. Rural Poverty: Incidence and Trends In 1986 India, Pakistan, and Bangladesh together had 983 million people (see table 2-1). About 76 percent of them-some 741 million people-lived in rural areas, as do most of the poor; about 67 percent of the combined labor force of the three countries work in agricul- ture. Overall growth rates of gross domestic product (GDP) per capita have been small except in Pakistan, and the distribution of income suggests that even the modest increases may not have been shared equally. Agricultural output has barely exceeded population growth in Pakistan and India and has fallen behind in Bangladesh. It is conse- quently inferred that rural living standards have declined and that the incidence of rural poverty in South Asia has increased (see ILO 1977a). This inference has been reflected in a growing and disquieting liter- ature on trends in rural poverty in South Asia. Starting with the pio- neering work of Dandekar and Rath (1971), the literature is fairly ex- tensive for India, and some recent work has also been done for Pakistan and Bangladesh.' Many of the relevant studies, though not all, suggest that where agricultural growth has stagnated, the inci- dence of poverty in the rural population has definitely increased and that the absolute number of the rural poor has increased everywhere (see table 2-2 for selected aggregate estimates). More serious, some analysts have argued that even vigorous and sustained agricultural growth has been accompanied by a steady deterioration in the distri- bution of incomes-not only a relative decline in the incomes of the rural poor but also an absolute decline in real terms.2 This powerful myth, accepted by a growing number of policymak- ers, has serious implications for the poor. If the incidence of poverty increases with growth, the poor are bound to become the direct and innocent victims of policies designed to increase growth, and the con- 8 Rural Poverty: Incidence and Trends 9 Table 2-1. South Asia: Basic Economic Characteristics Item Bangladesh India Pakistan Population Total population, mid-1986 (millions) 103.0 781.0 99.0 Rural population, 1986 Number (millions) 84.5 585.5 70.9 Percent 82 75 71 Annual rate of growth, 1980-86 (percent) 2.6 2.2 3.1 GDP (current prices) Total GDP, 1986 (billions of U.S. dollars) 15.5 203.8 30.1 Per capita GDP (1986 U.S. dollars) 140 261 303 Agriculture, 1986 Agricultural GDP as a percentage of total GDP 47 32 24 Rural population per square kilometer of agricultural area 990 345 280 Labor force, 1985 (percent) 1985 estimated population of working age (15-64 years) 53 56 53 Percentage of labor force in agriculture 75 70 55 Average annual growth rates (percent) GDP, 1980-86 3.7 4.9 6.7 Agriculture, 1980-86 2.7 1.9 3.3 Labor force, 1980-85 2.8 2.0 2.0 Sources: World Bank (1988a and b). flict between equity and growth objectives consequently becomes acute. For this argument to hold, however, it must be shown both that GDP per capita has grown and that it has increased the incidence of rural poverty.3 As I shall show in this chapter, the factual basis for these suppositions is by no means sound. Patterns of Poverty Has absolute and relative rural poverty increased substantially in South Asia in spite of (or because of) agricultural growth? A careful examination of the evidence shows that much of it is based on poor data and the slippery concept of the poverty line. Where the data are reliable, no clear trends emerge. Table 2-2. South Asia: Estimates of Rural Poverty, Selected Years and Studies Percentage of Implied size of rural Definition of poverty rural population population in poverty (per capita income or expenditure) Year in poverty (millions) Bangladesh Based on a food bundle of 2,100 1963-64 88 49.9 calories and 45 grams of proteina 1966-67 62 37.9 1968-69 79 50.7 1973-74 94 64.0 Absolute poverty: Tk23.6 (monthly per capita 1963-64 40 22.2 income at constant 1963-64 prices), equivalent 1968-69 76 47.0 to 95 percent recommended calorie intakeb 1973-74 78 53.5 1975 (Ist quarter) 62 43.6 Based on food bundle of 2,122 caloriesc 1981-82 74 60.9 India Rs480 annuallyd 1968-69 75 307.7 1969-70 72 305.3 1970-71 68 295.7 Consumer expenditure level of Rsl5 1959-60 49 172.8 per 30 days at 1960-61 rural pricese 1960-61 42 152.4 1963-64 49 189.1 1965-66 51 204.7 1967-68 58 241.1 1970-71 49 216.6 1973-74 48 221.3 Pakistan Rs300 annually (at constant 1959-60 1963-64 61 23.5 prices)f 1966-67 60 24.8 1968-69 62 26.7 1969-70 60 26.5 1970-71 55 25.3 1971-72 58 27.5 Rs27.5 per capita per month (at constant 1963-64 72 27.7 1959-60 prices)g equivalent to 95 percent 1966-67 64 26.5 recommended calorie intake 1968-69 62 27.6 1969-70 68 30.0 1970-71 71 32.7 1971-72 74 35.1 a. Alamgir (1975). b. Khan in ILO (1977a), pp. 144-47. c. World Bank (1987b), p. 137. d. NCAER (1975), p. xv. e. Montek S. Ahluwalia (1978). f. Alauddin (1975). g. Naseem in ILO (1977a), pp. 41-60. 12 Patterns of Poverty India One of the most reliable estimates of the trends in rural poverty in India (where comparative data permit careful analysis) was first made by Montek Ahluwalia (1978). He used consumer expenditure data drawn from the National Sample Survey (NSS) for 1956-57 to 1973-74 to calculate the percentage of the rural population below a given poverty line for India as a whole as well as for individual states.4 Table Al reproduces results from his study. (Tables with numbers preceded by "A" are found in the statistical appendix at the back of the book.) The most important findings of his study through the early 1970s were as follows: • For India as a whole, the extent and incidence of rural poverty varied markedly over time, with no overall trend to increase or decrease, at least before the Green Revolution. * Since the population grew steadily and the incidence of poverty did not fall markedly, the absolute number of rural poor grew significantly over time. Ahluwalia calculated the increase to aver- age about 5 million people every year during that period. * The incidence of poverty in the states followed a pattern similar to that for India as a whole. Poverty fell in almost all states until the early 1960s, peaked in 1967-68 or 1968-69, and declined again thereafter. Only two states (Assam and West Bengal) showed a significant increase in poverty, while two others (An- dhra Pradesh and Tamil Nadu) showed a decline. Ahluwalia also looked at an alternative and more reliable measure of poverty-Sen's poverty index, which takes into account the per- centage of the population in poverty, the gap between the poverty line and the mean consumption of the poor, and the extent of ine- quality among the poor.5 The Sen index showed a pattern of fluctua- tion similar to that described above with "no evidence for asserting a trend increase or decrease in rural poverty over [the] period as a whole" (Ahluwalia 1978, p. 304). But that was before more recent data were available for the decade after the Green Revolution, from the late 1970s to the early 1980s. Ahluwalia included the most recent NSS data available (1956-57 to 1977-78) in his 1985 analysis and wrote: The NSS data permit the firm conclusion that there is no basis for the view that the incidence of rural poverty has increased over time, especially in the period after the green revolution. The data for 1956/57 to 1977/78 show no significant trend, and the data for the period after the green revolution show a more or less steady decline. Rural Poverty: Incidence and Trends 13 The second issue considered . . . is the relationship between rural poverty and agricultural income levels per head. Three conclusions have emerged: (1) There is fairly strong evidence of an inverse rela- tionship between agricultural income per head and the incidence of rural poverty, especially if account is taken of lagged effects; (2) there is no evidence that this relationship has weakened since the green revolution; and (3) reliance on growth alone will not bring about a large reduction in the incidence of poverty in the near fu- ture. These conclusions underscore the need for rural develop- ment programs aimed especially at the rural poor. (Ahluwalia in Mellor and Desai 1985, p. 72) Robin M. Mukherjee (1986) arrived at similar conclusions when ex- amining all-India data and further evidence from selected states. Using evidence from the NSS 38th Round (India, Ministry of Plan- ning, 1983), he found significant declines in the incidence of poverty in Haryana, Kerala, Punjab, Tamil Nadu, and even West Bengal start- ing around 1970 and continuing into the 1980s. C. H. H. Rao (1986) confirms further declines in the incidence of poverty in all states ex- cept Rajasthan between 1977-78 and 1982-83. The incidence of rural poverty in India as a whole fell from 40 percent to 32 percent between 1978 and 1983 (tables A2 and A3). The combined evidence on long-term trends suggests that the incidence of poverty fell from around 55 percent in the late 1950s to around 35 percent by the early l980s (see table 2-3 and figure 2-1). Although the debate about the reliability and usefulness of the NSS data for measuring poverty continues, they are the only data that allow consistent comparative analysis over more than three decades. Data from earlier studies have been consistently used to argue that growth has little or no impact on poverty. That conclusion predated the later advances in agricultural growth, however. More recent evi- dence should lay to rest the view that growth has in general made rural poverty worse or has failed to alleviate it. The fact is that growth as an explicit strategy for relieving poverty has never been high on the agenda of policymakers, especially in India. The new evidence should convince policymakers of the absolute importance of acceler- ating rates of growth, especially in regions where poverty is wide- spread, and of introducing special programs to alleviate poverty among those who have benefited less from this growth. Pakistan The evidence on rural poverty in Pakistan is less extensive than that for India and is based on less reliable and noncomparable data. Khan 14 Patterns of Poverty Table 2-3. Trends in Rural Poverty in India Percentage of population in poverty Sen's poverty index Year Ahluwalia Mukherjee Ahluwalia Mukherjee 1956-57 54.1 n.a. 0.23 n.a. 1958 50.2 n.a. 0.22 n.a. 1959 46.5 n.a. 0.19 n.a. 1960 44.4 n.a. 0.17 n.a. 1961 38.9 n.a. 0.14 n.a. 1962 39.4 n.a. 0.14 n.a. 1963 n.a. n.a. n.a. n.a. 1964 44.5 n.a. 0.16 n.a. 1965 46.8 45.7 0.17 0.17 1966 53.9 48.7 0.21 0.19 1967 56.6 56.0 0.24 0.24 1968 56.5 55.5 0.20 0.23 1969 51.0 49.0 0.18 0.20 1970 n.a. 45.2 n.a. 0.19 1971 47.5 n.a. n.a. 0.17 1972 41.2 46.4 n.a. n.a. 1973 43.1 46.1 0.17 0.18 1974 46.1 n.a. 0.14 0.16 1975 39.1 n.a. n.a. n.a. 1978 39.1 40.2 0.14 0.14 1983 n.a. 32.8 n.a. 0.11 n.a. Not available. Sources: Ahluwalia (in Mellor and Desai 1985), and personal communication from Robin M. Mukherjee, Economic Research Unit, Indian Statistical Institute, Calcutta. and Bose (19-68) showed that per capita incomes in the agricultural sector in East Pakistan (now Bangladesh) fell between 1949-50 and 1963-64. In West Pakistan (now Pakistan) incomes increased (though very gradually) during the same period; even in the West, however, the study found that the majority of the farming population had less than the average income, which was itself barely above the subsistence level. In a more extensive study, Naseem (1973) showed that findings based on national income data revealed substantially higher increases in rural per capita income than expenditure data drawn from house- hold surveys, especially after 1963-64. Using this latter data base, he estimated that the percentage of those in West Pakistan below a given poverty line had fallen substantially between 1963-64 and 1969-70. In a subsequent study, however, Naseem (in ILO 1977a) qualified his earlier findings. He defined three separate poverty lines by consump- Figure 2-1. Rural Poverty in India and Selected States, 1956-83 Percentage of rural population in poverty 80 70 60 O West Bengal 50 -. 4 Tamil Nadu 40 - A ' All India 30 20 IV 10 - --_lPunjab 0 I I I I I I I I I I I I I I I I II i 1956-57 1960 1965 1970 1975 1980 1983 Note: Broken line indicates extrapolated data. Sources: Tables 2-3 and A3. 15 16 Patterns of Poverty tion levels of 95, 92, and 90 percent of the "minimum calories re- quired per capita." He concluded that the proportion of the popula- tion in poverty varies considerably according to the poverty line chosen6 and that there was no definite trend in the incidence of pov- erty even when an arbitrary index was used to define it. When he de- fined poverty in "the most extreme intolerable terms" (that is, a level of expenditure that permitted the consumption of no more than 90 percent of the minimum number of required calories), he found that in 1971-72, 54 percent of rural households and 43 percent of the rural population were below these extreme levels of poverty and that the percentage of people in poverty had remained roughly the same over the years. In view of the growth in population, the implication is that the absolute number of Pakistanis living in poverty had in- creased substantially-a finding similar to that for India for the same period, but more disquieting because Pakistan had had higher rates of agricultural growth. Naseem's findings have not been universally accepted, however. Alauddin (1975) studied the same period and used the same data but reached the opposite conclusion (see table 2-2). Alauddin's calcula- tions suggested that the percentage (though not the absolute number) of persons below the poverty line had fallen in West Pakistan since 1963-64. But Alauddin's poverty line was not the same as Naseem's, and he used a different price index to make data for several years comparable. It is not possible to establish which of these studies is more credible, although Naseem's finding that "extreme poverty" had risen in West Pakistan is difficult to reconcile with World Bank data on rural wages presented in a 1977 internal report. The latter showed that real wages of casual agricultural workers remained stable be- tween 1959 and 1966, but began to rise after 1966 at a rate of about 2-5 percent a year. The real wages of permanent agricultural workers showed a similar trend between 1966 and 1973, with a fairly uniform pattern of increase in all districts.7 Reviews of more recent studies on the incidence of poverty in Paki- stan suggest that it has declined substantially, from around 20-30 percent in the mid-1970s to around 10-15 percent in the mid-1980s depending on the poverty line used (Allison 1989; Ahmad and Lud- low 1989). Because of sustained high rates of agricultural growth and substantial out-migration from rural areas (especially to the Middle East), poverty certainly appears to be far less evident in Paki- stan than in India. A conjecture about the present incidence of ru- ral poverty in Pakistan would place it close to that found in Punjab and Haryana in India-that is, about 10-12 percent of the rural population. Rural Poverty: Incidence and Trends 17 Bangladesh Work on the incidence of rural poverty in Bangladesh is also limited and has problems similar to those of the Pakistani studies, but the findings are perhaps less equivocal, at least through the 1970s.8 As noted above, Khan and Bose (1968) found that agricultural incomes in East Pakistan had actually fallen between 1949-50 and 1963-74. From more detailed data Bose (1968) concluded that per capita agri- cultural incomes fell during the entire decade of the 1950s except for a slight reversal observed in 1963-64. He observed that rural wage earnings in Bangladesh were lower in every year between 1951 and 1960 than they had been in 1949. After falling in 1962, earnings peaked in 1964 but then fell again, so that "by 1966 rural wages were at a level lower than in 1959." Alamgir (1975) seems to confirm these broad results for 1963-64 to 1973-74 (see table 2-2). No clear trend can be attributed to Alamgir's findings, however, and his study relies on two surveys which are probably not comparable. A. R. Khan (in ILO 1977a) noted that most analyses, including those for Bangladesh, make no allowance for differences in calorie require- ments for different types of workers.9 To deal with this problem, he split the poor into two groups-the absolutely poor (whose calorie in- take is only 90 percent of the "recommended requirements for their occupation") and the extremely poor (whose calorie intake is only 80 percent of the recommended requirements). Using these categories, Khan showed that there was a sharp increase in the incidence of pov- erty in Bangladesh after the early 1960s. His calculations suggested that only 5 percent of the rural population could be categorized as "extremely poor" in 1963-64, but that the proportion rose to over 40 percent in the 1970s. He further estimated that the "absolutely poor" made up about 40 percent of the population during 1963-67 and nearly 62 percent of the total in 1975. Khan also examined the evidence on real agricultural wages; although he found no trend for the entire period from the early 1950s to the 1970s, he stated categor- ically that real wages started to fall after 1964 and that this decline accelerated sharply in the early 1970s. In view of this evidence, Khan concluded that the living standards of the vast majority of the rural population in Bangladesh fell in abso- lute terms through the 1970s; real wages of agricultural laborers fell; and per capita rural incomes and output fell, especially during the first five years of the 1970s. All groups may not have shared in this fall, however; many households at the top of the scale had increases in real incomes during that period. Studies of rural poverty also showed an increase in the incidence 18 Trends in Agricultural Output and Incomes of rural poverty through the 1970s. According to Muqtada (1974) this incidence increased significantly from 44 percent in 1963-64 to over 68 percent by 1976-77 and was accompanied by continuing popula- tion growth in spite of war and natural disasters. As a result, the abso- lute number of the poor in Bangladesh grew dramatically-more than doubling according to Muqtada-in this period. No recent reliable data are available on the incidence of rural pov- erty in Bangladesh. Per capita agricultural production, however, de- clined or stagnated throughout the first half of the 1970s, and this trend was reversed in the latter half of the 1970s and the first half of the 1980s. If the relationship between the incidence of poverty and agricultural growth is similar to that confirmed elsewhere in South Asia, rural poverty must also have declined in Bangladesh in recent years. There is some recent evidence of this. A careful study carried out by the World Bank using household expenditure survey data con- cluded that the overall incidence of poverty declined between 1974 and 1982 from 82 percent to 73 percent; the incidence of extreme poverty (daily consumption of only 1,850 calories per person) in- creased from 43 to 50 percent; and there were 4 million more poor and 1.3 million more "hard core" poor in 1982 than in 1974. The study adds that as a result of continuing increases in per capita in- come and substantial increases in food for work programs, the pov- erty situation does not seem to have deteriorated further (World Bank 1987b). These findings are disturbing, for even where the incidence of pov- erty has declined and when the data are interpreted conservatively, they suggest the enormity of the problem."0 It would, however, be wrong to take them (and the numbers they imply) at face value. There are serious problems with both the quality of the data and the meth- ods used. (A few of the problems associated with deriving and inter- preting data on the incidence of poverty are discussed in appendix A.) One should therefore accept the numbers on rural poverty as merely indicative, and one should be extremely cautious in accepting the evidence on trends. Trends in Agricultural Output and Incomes Another problem with the notion that growth was impoverishing rural people in South Asia is that there was little real growth in per capita terms in most parts of South Asia during the 1970s. Increases in food-grain output per capita were confined to a few regions and slowed considerably in the latter half of the 1970s. Furthermore, agri- cultural net domestic product (NDP) per capita, a more appropriate Rural Poverty: Incidence and Trends 19 measure, stagnated or actually declined in most parts of the subconti- nent in the 1960s and 1970s. Table 2-4 gives the relevant data on food-grain production in the 1970s. Per capita food output-generally used in arguments about the impoverishing consequences of growth-did show positive trends in the 1970s for most of the Indian states (although only a few had high growth rates) and for Pakistan; the trends for Bangladesh were negative (figure 2-2). Growth rates were generally much lower in the 1960s except in East Punjab and Haryana, states which experienced an early breakthrough in the adoption of HYV wheat. But it is highly misleading to concentrate on food-grain production alone as an indi- cator of growth in the rural sector. The sector includes other products and income opportunities-and it is well known that food-grain pro- duction was increased by diverting land and resources from these other crops." A more complete picture emerges from the data on net domestic product in the agricultural sector shown in table 2-5. The figures for India show that between 1960 and 1970 agricultural NDP per head of the rural population rose on average by about 3 percent a year in six states-Gujarat, Haryana, Himachal Pradesh, Karnataka, Punjab, and Rajasthan-that accounted for 19 percent of the rural popula- tion; it rose by less than 1 percent in four other states accounting for another 31 percent of the rural population, but actually fell in the remaining seven states that made up 50 percent of the rural popula- tion.'2 This last group included Bihar and Madhya Pradesh, each of which had had a per capita NDP below the all-India average in 1960-61. By 1970-71 per capita NDP had fallen below the Indian av- erage in three more states-Assam, Maharashtra, and West Bengal. Data for the entire decade of the 1970s are not available to allow a comparable analysis, but a state-by-state breakdown of national ac- counts data for agriculture between 1970 and 1975 illustrates what happened in the first half of the decade."3 Agricultural output con- tinued to fall, and the per capita agricultural NDP of the rural popula- tion fell by 1 percent a year for the country as a whole. The states that had shown vigorous growth in the 1960s (agricultural NDP per capita growth of over 2 percent a year) actually experienced a decline; the second group, which had recorded only moderate growth, also registered an overall decline (though of a smaller magnitude). Only four states, which accounted for 18 percent of the rural population, experienced positive growth rates in agricultural NDP per capita, and only two of them-Himachal Pradesh and Maharashtra-had rates above 2 percent a year. The rest of the country, accounting for 91 percent of the rural population in 1976, experienced a relative de- Table 2-4. South Asia: Annual Growth Rates of Population and of Food-Grain Output 1960-70 1970-79 Country Rural Food-grain output Rural Food-grain output and state population Total Per capita' population Total Per capita' Bangladesh 2.8 2 Ob -0.8b 2.40 1.13 - 1.27 India 2.00 1.85 -0.15 1.88 2.74 0.86 Andhra Pradesh 1.68 0.39 -1.29 1.50 3.46 1.96 Assam 1.34 2.47 1.13 3.02 2.11 -0.91 Bihar 1.78 0.21 -1.57 1.83 2.51 0.68 Gujarat 2.29 3.16 0.87 2.00 2.07 0.07 Haryana 2.78 5.20 2.42 2.01 3.09 1.08 Karnataka 1.93 2.85 0.92 1.70 2.24 0.54 Kerala 2.22 2.01 - 0.21 1.94 - 0.14 -2.08 Madhya Pradesh 2.31 -0.42 - 2.73 2.44 0.92 -1.52 Maharashtra 2.03 -0.52 -2.55 1.52 9.15 7.63 Orissa 2.03 2.16 0.13 2.00 0.95 -1.05 Punjab 1.89 9.54 7.65 1.36 5.15 3.79 Rajasthan 2.32 -0.40 - 2.72 2.65 2.93 0.28 Tamil Nadu 1.53 0.69 -0.84 0.97 1.44 0.47 Uttar Pradesh 1.68 2.27 0.59 1.72 2.34 0.62 West Bengal 2.37 2.80 0.43 2.22 1.34 0.88 Pakistan 2.7 5.0 9.3 3.0 3.3 0.32 Notes: These are trend growth rates for comparable periods (1959-60 to 1968-69, and 1976-77 for Pakistan; 1960-61 to 1969-70, and 1978-79 for India; and 1960-65 to 1965-70, and 1978-79 for Bangladesh). a. Per capita growth rates are simply the difference between the growth rates for rural population and food-grain output. b. For rice production only. Sources: M. H. Khan (1981), p. 29 for Pakistan (log linear trend); Alagh and Sharma (1980), pp. 100-01 for India (linear and log linear estimates a corrected by output regression coefficients and standard errors); and World Bank data for Bangladesh (annual growth rates). Rural population growth rates for India and its states are from Census of India, 1971, series J, paper I of 1979 "Report of the Expert Committee on Population Projections," pp. 158-59. 22 Trends in Agricultural Output and Incomes Figure 2-2. Growth of Rural Population and Per Capita Food-Grain Output in South Asia, 1960-70 and 1970-79 nl Rural population Average annual percentage change 0 Food-grain output per capita 3.0 2.0 1.0 -1.0 -2.0 1960-70 1970-79 1960-70 1970-79 1960-70 1970-79 Bangladesh India Pakistan Source: Table 2-4. cline in per capita agricultural NDP in the first half of the 1970s. Ex- cept in the record harvest years from 1975-76 to 1977-78, agricul- tural output per head of rural population in India as a whole stood below its 1960-61 levels.'4 Even state-level data are too highly aggregated to show the diverse patterns of output. The localized nature of growth in agricultural output is confirmed by Bhalla and Alagh (1979). In a careful study of agricultural performance based on district data between 1962-65 and 1970-73, they showed that of the 282 districts studied in India, nearly a quarter (70 districts) had negative rates of growth; another 62 districts (22 percent) had rates of growth ranging between 0 and 1.5 percent a year, while 102 districts (36 percent) had growth rates varying between 1.5 and 4.5 percent. Once population growth is ac- counted for, only the remaining 48 districts with growth rates in ex- cess of 4.5 percent can be said to have experienced positive per capita growth. All of these districts-except four in Karnataka and two in Tamil Nadu-were concentrated in six states in the north and west-Rajasthan (twelve), Punjab (eleven), Western Uttar Pradesh (eight), Haryana (five), Gujarat (two), and Jammu and Kashmir (two). These districts are in the arid zones in northwestern India where the productivity base was very low to begin with or where irri- gation, particularly tubewell irrigation, has had a major impact. Rural Poverty: Incidence and Trends 23 This spatial distribution of growth suggests that, since growth has been confined to particular regions, it is in those regions that one should test the hypothesis that growth further impoverishes the poor. It also suggests a need to look carefullv at district data. Even statewide figures can be misleading since they aggregate districts that have done poorly with ones that have done well. District-level data on poverty are very difficult to obtain, however, and few studies are based on them. (Bardhan in Mellor and Desai 1985 is an ex- ception.) Although differences in the land and in irrigation facilities are im- portant determinants of these regional inequalities, there is grow- ing evidence that public policy contributed significantly to them. Subbarao (1985b) analyzed state incentive policies in India as factors explaining the long-term prospects for growth in various states. He calculated that in India total subsidies for agriculture and rural devel- opment programs rose from Rs1O,000 million in 1974-75 to about Rs25,000 million by 1980-81 at constant prices. Of the easily identifi- able state incentives, benefits from price supports accrued mainly to four states-Haryana, Punjab, Tamil Nadu, and Uttar Pradesh. Bene- fits from fertilizer subsidies (estimated at Rs9,000 million in 1983-84) also accrued to four states-Andhra Pradesh, Punjab, Tamil Nadu, and Uttar Pradesh-which accounted for half of the nation's gross ir- rigated area and claimed 55 percent of the subsidies. Subsidies to in- stitutionalized credit also went disproportionately to a few states. He concludes correctly that public incentive policies-mainly agricultural price supports, input subsidies, and subsidized credit-accrued dis- proportionately to the prosperous states that already had access to water and fertilizers. Public policies have reinforced rather than com- pensated for existing regional inequalities and have thus accentuated regional disparities. These "state-induced market price distortions," as Subbarao calls them, have aided growth in advanced states and re- tarded it in others. As a result, fewer public resources have been left for investments in poorer areas. Available data for Bangladesh show that agricultural output barely kept pace with population growth in the 1960s and that agricultural NDP per capita of the rural population declined slightly between 1969-70 and 1977-78-for all practical purposes a record of stagna- tion over two decades. Overall performance in Pakistan was similar to that in the Indian Northwest, at least in the 1950s and early 1960s, with a growth rate of 1.6 percent a year. More recent data reveal, however, that per ca- pita agricultural output of the rural population stagnated between 1969-70 and 1977-78, a period corresponding to most of the poverty studies in Pakistan. The better than average performance in the Pun- jab and Sind provinces dominated the overall Pakistan figures, how- Table 2-5. South Asia: Agricultural Growth per Capita of Rural Population (percent) 1960-70 growth rates 1970-75 growth rates Country Agricultural NDP' Agricultural NDPR 4 ~~~Country Rural population, and state Rural population Total Per capita Rural population Total Per capita 1976 Bangladesh 2.50 2.56 0.06 2.40 2.10 -0.30 72.8 India 2.00 2.24 0.24 1.88 0.88 - 1.00 481.5 Gujarat 2.29 4.00 1.71 2.00 - 10.10 - 12.10 4.4 Haryana 2.78 5.09 2.31 2.01 -0.97 -2.98 1.9 Himachal Pradesh 2.02 6.66 4.64 2.01 5.13 3.12 0.7 Karnataka 1.93 3.42 1.49 1.70 2.40 0.70 5.0 Punjab 1.89 4.60 2.71 1.36 0.53 -0.83 2.3 Rajasthan 2.32 5.18 2.86 2.65 -5.00 -7.65 5.0 Regional aggregate 2.19 4.89 2.70 2.03 -2.24 -4.27 19.3 Andhra Pradesh 1.68 2.35 0.67 1.50 0.13 - 1.37 7.9 Jammu and Kashmir 2.39 3.37 0.98 2.47 0.42 -2.05 0.9 Kerala 2.22 2.74 0.52 1.94 -1.76 -3.70 4.1 Uttar Pradesh 1.68 2.58 0.90 1.72 -3.10 -4.82 17.2 Others 2.52 3.02 0.50 n.a. n.a. n.a. n.a. Regional aggregate 1.79 2.56 0.77 1.73 1.78 -3.51 15.5 Assam 1.34 0.99 -0.35 3.02 1.70 -1.32 3.3 Bihar 1.78 1.38 -0.40 1.83 -0.34 -2.17 11.5 Madhya Pradesh 2.31 -0.03 - 2.34 2.44 - 2.77 -5.21 8.2 Maharashtra 2.03 -2.07 -4.10 1.52 7.10 5.58 7.8 Orissa 2.03 1.12 -0.91 2.00 2.14 0.14 4.6 Tamil Nadu 1.53 1.42 -0.10 0.97 -3.87 -4.84 6.3 West Bengal 2.37 1.67 -0.70 2.22 -0.43 -2.65 7.7 Regional aggregate 1.95 0.56 -1.39 - 1.87 0.40 - 1.47 49.4 o Pakistan 3.34 4.90 1.56 2.28 2.36 0.08 52.6 n.a. Not available. Notes: Ordinary least square regressions are used to calculate all growth rates. a. Net domestic product. b. Millions. Sources: Data from the World Bank, Reserve Bank of India, and National Accounts Statistics of India. 26 Agricultural Growth and Poverty ever, and there is evidence to suggest that the Northwest Frontier Province (NWFP) and Baluchistan may have fallen considerably be- hind. Agricultural Growth and Poverty For India, where relatively reliable production data are available, Ahluwalia (1978 and in Mellor and Desai 1985) shows that there is clear evidence of an overall inverse relationship between the inci- dence of poverty and indices of food production, agricultural pro- duction, and net domestic product per capita of the rural population. He found that this relationship held in at least seven of the fourteen states that he examined. These states-Andhra Pradesh, Bihar, Karnataka, Madhya Pradesh, Maharashtra, Tamil Nadu, and Uttar Pradesh-accounted for 56 percent of the rural population and about three-quarters of the rural population in poverty."5 Ahluwalia's first results for Punjab and Haryana, based on data from the 1960s and early 1970s, are of special interest because other studies have purported to show an increase in rural poverty in these states despite the success of the Green Revolution in raising agricul- tural output.16 His results showed that despite a dramatic growth in agricultural output per rural person, there was no evidence of any significant increase in the incidence of poverty and no discernible negative relationship between the incidence of poverty and the index of agricultural production per capita. So although there was no case for an increasing incidence of poverty as other studies have main- tained, there was also no sign that real consumption had increased among the poorest 25 percent of the population. There are three possible explanations. One frequently used is that the real consumption levels of the poorest 25 percent dropped be- cause the labor-displacing technologies that accompanied growth in the region had reduced employment opportunities (and perhaps real wages) for agricultural laborers and the landless. This explanation fails, however, to square with the evidence of heavy migration into Punjab and Haryana from Bihar and eastern Uttar Pradesh in re- sponse to rapidly rising labor demand. Ahluwalia puts forward a dif- ferent hypothesis. He suggests that the poorest quarter of the rural population in the high-growth states contains a rising proportion of immigrants. Although real consumption levels have indeed remained stagnant, this group "consists increasingly of individuals whose con- sumption is higher than it would have been if they had not mi- grated."'7 The implication is that the benefits of agricultural growth have in fact "trickled down," not only to poorer Punjabis (who have presumably moved into higher expenditure groups) but also to the Rural Poverty: Incidence and Trends 27 poor from other states who have migrated to areas of relatively rapid growth in search of employment. The ample evidence of the dramatic expansion of demand for rural labor in high-growth states, of increas- ing in-migration (seasonal and permanent) from other states, and of increases in average real wages in rural areas makes this second expla- nation at least as plausible as the first."8 A third explanation is that most studies do not cover a long enough period and in fact fail to cover the evidence from more recent years. Ahluwalia (in Mellor and Desai 1985) re-examined his results on the incidence of poverty and its relationship to agricultural growth using data from 1956-57 to 1977-78. The evidence is now clearer. There is a significant and consistent negative relationship between agricul- tural growth and the percentage of rural population below poverty (table 2-3). This relationship is also borne out by C. H. H. Rao (1986), who compares the incidence of poverty across regions using data for 1977-78 and 1983-84. Almost everywhere in India-in the lagging eastern and central regions as well as in the south-the rural popula- tion below poverty has declined significantly. As mentioned earlier, similar evidence is provided by Robin M. Mukherjee (1986), who extends Ahluwalia's results to include data from the 32nd and 38th Rounds of the NSS for 1978 and 1983 (tables A2 and A3). They all show a clear trend of decline in the incidence of poverty for all the states studied and for India as a whole. A continuing problem with most poverty studies in South Asia is that they cover very short time spans. Some are based on data from the 1960s only, a period of low growth rates; others focus on even shorter periods. Few cover periods exceeding two to three decades. A few studies, carried out mainly by anthropologists or historians, have used longer time frames of five decades or more. Their findings are in sharp contrast to the more shortsighted views of economists and report significant amelioration in the conditions of the poor. Attwood (1979), for example, analyzed economic change over fifty years (1920-70) in a village in western India. He concluded that there was considerable upward as well as downward mobility; landholding had become less concentrated; some of the poor had gotten richer, while the rich had not gotten richer, at least in terms of their share of land-indeed some had gotten poorer; and "no less than 25% of the landless in 1920 had moved upward into the landed category" while "44% of those who were landed in 1920 had lost their lands by 1970" (p. 500). Bruce (1982) covered an even longer period, 1800-1979, in his study of villages in Bellary district in Karnataka. He concluded that agrarian change had led to a more equitable distribution of resources and more widely dispersed economic opportunities. In particular, 28 Agricultural Growth and Poverty technological change between 1945 and 1980 had greatly benefited agricultural laborers, and real wages for agricultural labor reached historic highs while employment opportunities nearly doubled. Ex- tensive evidence is available on the progress of the Green Revolution in the Punjab at the state level (see Gill 1976 and Blyn 1983). At the local level, Leaf (1983, p. 268), reports that in one village between 1965 and 1978 "the gains have gone at least as much to the poorer villagers as to the wealthier." Thus, the evidence for "impoverishing growth" in South Asia seems shaky, at least on the basis of direct evidence on poverty and its relationship to growth. Before the issue can be resolved, however, more time series data at the district and household levels will need detailed analysis. As Bardhan (1981) has pointed out, to trace the spe- cific impacts of growth we need a much larger number of micro stud- ies, even down to the district level. Without them, it is impossible to settle the controversy. More generally, growth in agricultural output per capita can be reconciled with a rising incidence of poverty only if the distribution of the benefits of growth is uneven or growth has ceased. The avail- able evidence on income inequalities, at least at the aggregate level, does not support the view that growth has increased poverty. Ahluwalia (1978) has tested for trends in inequality over time by calculating Gini coefficients of the distribution of rural consumption expenditures (in nominal terms) for India as a whole and for individ- ual states between 1957-58 and 1973-74 (see table A4).'9 He found that far from increasing, relative inequality appeared to be diminish- ing for India as a whole and for seven of the fourteen states that he examined. None of the states that showed a significant trend toward increasing output per capita of rural population had experienced an increase in relative inequality (which would have occurred if growth had been accompanied by absolute impoverishment, as some analysts contend). In particular, Ahluwalia found that the Gini coefficients for Punjab and Haryana showed a significant decline in inequality over two decades. The same general conclusions can be drawn from data on trends in income distribution in India from other sources (table 2-6). Some authors have asserted that inequalities have also increased in Pakistan and Bangladesh. On the basis of the available evidence (table 2-6) this assertion cannot be supported, at least for Pakistan. A study by Guisinger and Hicks (1978) clearly shows that rural income ine- quality did not increase in Pakistan between 1963 and 1971 and that the income shares of the poorest 20-40 percent of households showed no tendency to shrink over this period. De Kruijk (1986b) shows that inequality in Pakistan is due to inequality within provinces Rural Poverty: Incidence and Trends 29 Table 2-6. South Asia: Income Distribution in Rural Households (share of income in selected years) Lowest Lowest Middle Highest Highest Gini Country and year 20% 40% 40-60% 20% 5% coefficient Bangladesh 1963-64 7.8 19.2 15.5 43.1 17.2 0.35 1966-67 8.1 19.9 16.0 41.7 16.1 0.33 1968-69 10.4 23.6 16.6 37.0 13.1 0.27 1973-74 6.8 18.3 11.2 41.7 n.a. 0.35 India 1960 3.9 13.8 14.9 49.6 23.8 0.45 1962 5.9 16.2 14.5 48.1 n.a. 0.41 1964-65 7.5 19.3 15.4 44.6 21.9 0.37 1966-67 4.9 13.2 12.7 53.3 n.a. 0.46 1967-68 4.7 13.0 13.0 53.1 24.6 0.48 1968-69 5.3 14.8 14.0 50.6 n.a. 0.43 1969-70 6.3 16.6 14.7 47.1 n.a. 0.39 1970-71 6.4 17.1 15.1 46.6 n.a. 0.38 1975-76 7.3 17.1 15.1 46.4 n.a. 0.39 Pakistan 1963-64 6.8 18.0 16.5 43.0 17.0 0.36 1966-67 8.5 20.5 16.0 41.4 16.5 0.33 1968-69 8.8 21.6 16.8 30.0 14.6 0.30 1969-70 8.6 21.5 16.9 39.0 14.8 0.30 1970-71 9.2 21.9 16.9 38.6 15.0 0.29 n.a. Not available. Source: Calculated from data in Jain (1975). rather than between provinces, as is widely believed. Although no di- rect evidence on trends in inequality is available for Bangladesh, sup- plementary evidence makes it at least possible that the picture there was not dissimilar from that in Pakistan. Agricultural Wage Rates Since a majority of the poor belong either to rural labor households or to small cultivating households that depend on wage employment for a large share of their income, trends in real wage rates in rural areas and their relationship to agricultural growth yield additional ev- idence relevant to this controversy (table 2-7). The evidence from Bangladesh is clear. A. R. Khan's study (in ILO 1977a), based on earlier data reported by S. R. Bose (1974b), shows a general decline in real average rural wages.20 Although no clear trends exist, these studies have used a long time series from 1949 to 1975 and have concluded that real wages fell sharply up to the early Table 2-7. Indexes of Money and Real Wage Rates for Male Agricultural Laborers, 1956-57 to 1971-72 North India West India South India East India Punjab and Uttar Madhya Andhra Year Haryana Pradesh Gujarat Pradesh Maharashtra Pradesh Karnataka Kerala Tamil Nadu Assam Bihar Orissa West Bengal Pakistan Bangladesh Index of money wage rates (1956-57=100) 1957-58 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 1958-59 112 129 110 11 101 122 114 107 98 97 111 97 101 n.a. n.a. 1959-60 110 249 131 125 92 121 126 114 102 101 100 95 106 n.a. n.a. 1960-61 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. ii.a. n.a. n.a. n.a. n.a. 1961-62 126 141 122 121 115 138 128 133 108 102 104 127 114 n.a. n.a. 1962-63 127 140 124 121 117 136 128 155 112 100 102 146 110 n.a. n.a. 1963-64 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 1964-65 144 178 139 129 144 159 160 170 140 105 135 192 129 n.a. n.a. 1965-66 182 221 144 142 146 168 142 195 142 123 160 187 155 n.a. n.a. 1966-67 180 250 158 163 175 189 145 230 159 138 n.a. 211 177 n.a. n.a. 1967-68 229 284 178 n.a. 189 213 156 282 170 n.a. n.a. 176 191 n.a. n.a. 1968-69 276 290 181 179 195 217 156 301 183 162 196 n.a. 175 n.a. n.a. 1969-70 303 303 185 190 217 233 n.a. 308 187 167 n.a. n.a. 185 n.a. n.a. 1970-71 304 315 213 194 212 261 188 337 194 178 n.a. 212 183 n.a. n.a. 1971-72 305 326 233 210 211 239 198 338 217 180 n.a. 226 202 n.a. n.a. Index of real wage rates (1956-57=100) 1957-58 n.a: n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 90 1958-59 101 114 110 104 99 119 112 103 97 89 98 98 92 n.a. 87 1959-60 106 122 137 126 87 111 114 107 96 96 98 91 96 n.a. 87 1960-61 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 102 1961-62 121 143 117 119 114 124 114 118 94 95 100 119 109 n.a. 99 1962-63 121 144 116 113 107 122 111 136 105 87 94 115 90 n.a. 106 1963-64 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a, n.a. n.a. n.a. n.a. 110 120 1964-65 103 110 102 93 93 116 98 121 99 74 89 128 95 100 100 1965-66 131 135 103 92 90 112 74 123 97 78 89 110 91 105 86 1966-67 102 120 101 81 98 110 73 134 90 67 n.a. 105 90 113 86 1967-68 118 122 109 n.a. 101 122 74 155 97 n.a. n.a. 80 79 113 92 1968-69 142 165 111 93 108 118 76 146 101 72 104 n.a. 88 116 100 1969-70 153 156 107 93 116 121 n.a. 144 96 83 n.a. n.a. 93 125 101 1971-72 140 172 125 102 99 119 92 151 113 78 n.a. 96 96 130 72 n.a. Not available. Sources: India: Jose (1974); Lal (1976); Yelamanchili (1981). Bangladesh: A. R. Khan (in ILO 1977a), reporting data from Swadesh R. Bose (1974b) deflated by the cost of living index; the series of data is very close to that in Clay (1976), p. 424. Pakistan: FAO Production Yearbook 1974, deflated by cost of living index of agricultural laborers. Table 2-8. Bangladesh: Trends in Real Wages by Sector, 1969-70 to 1985-86 (index of real wage rates; 1973-74 = 100) 1985-86 Sector 1969-70 1973-74 1981-82 1984-85 1985-86 (taka per day) Rural Agriculture (unskilled without food) 113 100 88 104 120 29.5 Fishery (unskilled) 166 100 157 122 132 25.4 14i Urbana Cotton textile 186 100 107 106 116 28.7 Jute textile 113 100 77 67 83 28.3 Engineering 194 100 97 102 130 36.4 Vegetable oil 179 100 92 81 84 20.9 Small-scale industry 131 100 103 141 139 31.5 Construction 123 100 101 91 104 33.3 Notes: Nominal wages deflated by rural and urban cost of living indexes. a. Average wage rates for Chittagong, Dhaka, Khulna, and Rajshalii. Source: Bangladesh Bureau of Statistics in World Bank (1987b), table 6.4, p. 141. Rural Poverty: Incidence and Trends 33 1950s, rose steadily for about a decade thereafter, but began to fall again after 1964. Despite some fluctuations, the recent direction of change has been generally downward, reaching a nadir between 1973 and 1975-years of near famine in Bangladesh. The real wage rates in Bangladesh seem to have shown a persistent decline over most of the last three decades. A. R. Khan and others (1981) estimate that real wage rates for agricultural laborers declined nearly 17 percent between 1949 and 1970 and fell by almost 40 per- cent between 1949 and 1980. The most recent data, however, show that between 1981-82 and 1985-86 wages of unskilled agricultural laborers may have increased by some 36 percent. The-steady decline in real wages for agricultural workers between 1970 and 1982 seems to have been halted-even reversed-in Bangladesh (table 2-8.) These wage indices are consistent with Khan's earlier finding (in ILO 1977a) of a growing incidence of poverty in Bangladesh in the 1970s. Although agricultural output and the associated demand for labor were virtually flat, demographic conditions were worsening. Conse- quently, the living standards of a substantial part of the rural popula- tion (especially those who depended primarily on wage labor) must have fallen sharply during the 1970s. Notwithstanding the caveats noted earlier about the data and methodologies used to measure the incidence of poverty, this particular conclusion is clear-and every re- cent observer of rural Bangladesh seemed to concur with it. By contrast, real agricultural wages in Pakistan appear to have risen. Guisinger and Hicks (1978) found a pronounced trend for real wages to increase, particularly in the period 1966-73. Even the wages of casual laborers (mostly the landless or very small farmers) rose in real terms at a rate of 2-5 percent a year after 1966; by 1973 they were about 60 percent higher than their 1952 level. The real wages of permanent workers showed a similar trend, and the broad pattern of increase was fairly uniform across districts.21 This is consistent with other evidence of increasing rural productivity and incomes in Pakistan in the 1970s and 1980s. These developments were brought about by the Green Revolution and rising demand for labor, together with labor shortages during harvest periods in spite of increased mechanization. Extensive out-migration from rural areas in Pakistan to the Middle East and elsewhere has produced trends toward new labor shortages and higher rural wage rates, together with higher lev- els of remittances to rural areas. The evidence from India is extensive and mixed, as would be expected given the diversity of agrarian conditions and agricul- tural performance. The findings of some early studies on trends in money and real wages in agriculture have been very thoroughly reviewed by Bardhan (1977a and b) and are summarized below.22 34 Agricultural Wage Rates Although some studies showed a declining trend in real wages be- tween the mid-1950s and the mid-1960s (before high-yielding varie- ties had been adopted even in the Punjab, the heartland of the Green Revolution), analysis over a longer period does not reveal any under- lying trends in most states. In the fifteen years from 1956-57 to 1971-72, real wages rose in the first five years, declined by varying amounts between 1961-62 and 1967-68 in most states except Ke- rala,23 and then rose again except in the states where both agricul- tural output and wages generally stagnated-Assam, Bihar, Madhya Pradesh, Orissa, and West Bengal. Lal (1976) analyzed relatively recent and reliable NSS evidence on wage levels and found that real agricultural wages rose in most Indian states (other than those in the east) between 1956-57 and 1970-71 (except for West Bengal, for which there were no data for 1970-71). From 1956-57 to 1964-65, before the Green Revolution, real wages rose in only seven states and were constant or fell in six others; they subsequently rose generally, however, most notably where the impact of the Green Revolution was beginning to be felt-in Gujarat, Haryana, Punjab, and western Uttar Pradesh. This is confirmed by the rising disparity in real wage rates between agriculturally growing and agriculturally stagnant states and regions in India (Jose 1974; Krishnaji 1971). Lal thus surmises that earlier studies were wrong to conclude that agricultural growth was unlikely to raise the real wages of agricultural laborers and that it was accompanied by their further impoverishment. He shows that the percentage of total income accru- ing to the poorest three deciles of the population increased in all states where real wages rose between 1956-57 and 1970-71, years for which comparable NSS data were available for analysis.24 The most careful review of wage data for India is by Yelamanchili (1981), who analyzed state-level data. His results, in the form of indi- ces of money and real wages of agricultural workers, are given in table 2-8. These data support later conclusions that real wages have risen in states where agricultural growth has been rapid. A recent study by Parthasarathy (1987) also supports the view that real wages for agricultural laborers have increased in several states- Madhya Pradesh, Kerala, and even Bihar-especially where the per- centage of wage laborers in the rural population has declined (table 2-9). His evidence on declining real wages in Punjab and Haryana is, however, refuted by other evidence. He does show that because of large-scale in-migration the share of wage labor in the rural popu- lation has increased in these states-a finding evident to any visitor. Available evidence suggests that indices of real agricultural wages have generally increased in periods and areas in which agricultural growth has been rapid enough to outstrip population pressures. The Rural Poverty: Incidence and Trends 35 data do not support the view that agricultural growth has failed to raise the standard of living of the rural poor and may even have been accompanied by their further impoverishment.25 But where agricul- tural output has lagged behind or barely kept pace with population growth, it is not surprising that rural poverty has increased, in some instances dramatically, in the predominantly agrarian economies of South Asia. To summarize, agricultural growth does benefit even the poorest, and some (albeit small) benefits have trickled down to them, directly or indirectly, through rising real wage rates.2" But not enough! For the poor continue to grow in number, and the areas in which signifi- cant benefits have accrued to the rural poor account for only a small share of the total rural population of the subcontinent. This is a direct consequence of the general stagnation in the agricultural sector. The proper conclusion to draw, therefore, is that it is stagnation, not growth, that has brought impoverishment. Without general and rapid increases in ag- ricultural output per capita there is no chance of further redressing rural poverty in the subcontinent, irrespective of whether or not rad- ical institutional changes are brought about. Who Are the Rural Poor? Census data in South Asia do not make it easy to break down income or expenditure data by rural occupational classes and thus classify the poor by some easily recognizable features. One characteristic of the rural poor in South Asia-the smallness of their landholdings, if any-does, however, command general agreement among those who have studied the subject. Most of the rural poor, as Minhas (1974) points out, are made up of agricultural laborers without land, agricul- tural laborers with very little land, other rural laborers without land, and small landholders who usually have less than 5 acres. Landholding Characteristics The earliest study to confirm that most of the rural poor in India have little or no land and rely mainly on wage labor was done by Dandekar and Rath (1971). Using NSS survey data for 1956-57, they found that nearly 60 percent of all agricultural labor households had no land and 57 percent had less than the minimum level of per capita con- sumption. Agricultural labor households constituted 40 percent of all households in the category they called "rock bottom physical exis- tence." Three-quarters of the households that were cultivating be- tween 1 acre and 1.5 acres of land lived below poverty levels.27 The incidence of poverty remained high even among those with larger Table 2-9. India: Exponential Growth Rates of Money and Real Wages of Male Agricultural Laborers for Selected Villages, 1974-75 to 1984-85 (percent) Sowing Harvesting Field labor Money Real Money Real Money Real State District Village wage wage wage wage wage wage Bihar Patna Mahadevpur 6.4a 0.3 0,5b - 6.2a n.a. n.a. Bihar Muzaffarpur Narasingpur 11.7a 5*2b 15.7b 9.6 n.a. n.a. Bihar Ranchi Gaitalsood 7.0a 0.9 6.5b 0.4 n.a. n.a. Haryana Karnal Uggarkheri 8.3a 1.7 5.1a - 1.3 n.a. n.a. Madhya Pradesh Hoshangabad Sangkherakatan 10.la 5 lb 10.0a 5.0a n.a. n.a. Madhya Pradesh Surguja Basdei 11.5a 6,4b 9.5a 4,5b n.a. n.a. Madhya Pradesh Morena Bijaipur 10.3a 4,3b 8.1a 3.2 n.a. n.a. Madhya Pradesh Satna Kotar 9.7a 4,7b 9.Ia 4, b n.a. n.a. Kerala Kozhikode Koduvally 12.4a 6.5a n.a. n.a. n.a. n.a. Kerala Palghat Elapully 6.7a 3,7a n.a. n.a. n.a. n.a. Punjab Ludhiana Pakhowal 6.1a -0.4 3.6 -2.7 n.a. n.a. Rajasthan Kota Dhoti 6.6a 0.9 8.4a 2.5 n.a. n.a. Tamil Nadu Thanjavur Alangudi n.a. n.a. 1.0 -4.0 n.a. n.a. Tamil Nadu Tirunelveli Malayankulan n.a. n.a. 4.3a - 1.9 n.a. n.a. Uttar Pradesh Varanasi Awajapur 10.4a 3.8 8,1a 1.6 n.a. n.a. Uttar Pradesh Varanasi Keshavpur 10.Oa 354 6 ga 0.5 n.a. n.a. Andhra Pradesh Krishna Ghantasala n.a. n.a. n.a. n.a. 10.1a 4.8a Andhra Pradesh Guntur Tadikonda n.a. n.a. n.a. n.a. 12.7a 7,3a Andhra Pradesh Rangareddy Arutla n.a. n.a. n.a. n.a. 13.2a 7.8a Karnataka Bangalore Harisandra n.a. n.a. n.a. n.a. 9.7b 8.0 Karnataka Tumkur Gaddahalli n.a. n.a. n.a. n.a. 6.5a 0.4 n.a. Not available. <,, a. Significant at the I percent level. ,- b. Significant at the 5 percent level. Source: Parthasarathy (1987), table V, p. 12. 38 Who Are the Rural Poor? amounts of land; nearly 60 percent of households that cultivated landholdings of between 3.5 and 5 acres lived in poverty in 1963-64. Minhas (1974) reached similar conclusions. He stated, for example, that between 75 and 85 percent of all households that were operating holdings of fewer than 5 acres lay below a poverty line that he calcu- lated for 1960-61. There is little evidence that conditions have changed substantially in the 1980s. Other studies also state that the rural poor consist almost exclusively of landless laborers and small operators. In a study of the per capita income distribution of different rural groups, Bhatty (in Srinivasan and Bardhan 1974) found that for any given poverty line the incidence of poverty was far higher for-agricultural laborers than for cultivators, with nonagricultural workers falling in between. The more accurate Sen poverty index revealed, however, that cultivators at the lower end of the poverty scale were far worse off than agricul- tural laborers.28 Thus the small cultivator (owner and tenant) and the landless are not readily distinguishable by their standards of living. The poorest groups in Bangladesh belong to the same categories. A. R. Khan (in ILO 1977a) noted that agricultural households that were entirely dependent on wage income fell in the category of "ex- tremely poor"; even if they had significant nonwage income, they generally fell in the category of "absolutely poor." He also noted that nearly a sixth of the agriculturally active population and about a quarter of the rural population as a whole depended on wages as the major source of their income.29 It is clear from his study and that of Alamgir (1978) that the rural poor are agricultural laborers and those with small owned or tenanted holdings of fewer than 5 acres. The situation in Pakistan is no different, but, as Naseem (1971, 1979) points out, the real dichotomy may be not between large and small farms (the differences in size are getting smaller) but between those who own land and those who do not. He states that a large num- ber of small and marginal farmers were forced to sell and leave their land in the last decade and to eke out their existence by grazing live- stock on communal land. These, along with small tenant farmers and agricultural laborers, constitute the core of the rural poor. He esti- mates that whereas 60 percent of owner-cultivators and nearly 67 per- cent of tenants fell below the subsistence level, only 38 percent of owners who are also tenants belonged in this category. It is the me- dium-size owners (those with holdings of 5-25 acres) who have gained the most and have further enlarged their holdings by leasing- in land from smallholders or evicting their own smaller tenants. Agri- cultural labor households and small tenants were the worst off, with Rural Poverty: Incidence and Trends 39 per capita incomes about half that of small farming households (those with holdings of less than 7.5 acres) and about one-tenth that of large farming households (those with holdings of over 12.5 acres).30 A study by Albrecht (1976) gives figures of a similar order of magni- tude. Other Socioeconomic Characteristics The poor lack not only food and land, but also access to many other basic necessities-health, education, shelter, and security, for ex- ample. Most of the rural poor share economic, demographic, and so- cial characteristics that can help us to identify them (at least in a broadly qualitative sense) and to understand their economic plight. The assets and employment of the rural poor are examined in chapter 3. Other socioeconomic factors that set apart the poorest in the rural population of South Asia are outlined briefly below.3" In comparison with better-off groups, the poor (and especially the poor- est) have: * Higher birth rates, but not necessarily larger families because they also have higher infant and child mortality (mainly because of malnutrition)32 *Poorer health (as indicated by higher morbidity rates, lower weight to height ratios, higher incidence of protein-calorie mal- nutrition, and lower access to health care)33 and lower life expec- tancy *Larger households (but landless households are smaller than those with land, other things being equal)34 - A higher proportion of infants and small children (an especially vulnerable group) in their households3" *A larger proportion of young people (below 14 years)36 • A larger proportion of women of childbearing age and a higher number of pregnancies per adult woman. As a consequence of these demographic and health characteristics, the very poor: * Tend to have few earners and much higher dependency ratios, so that the ratio of workers to consumers, or of earners to nonearners or. part-time earners, is low37 * Have lower average labor force participation rates but higher age-specific participation rates among those (especially women) of working age38 40 Conclusions * Have lower literacy rates, less education, fewer skills, and hence lower overall endowments of human capital (again, especially true of women)39 * Are more likely to be members of nuclear families-rather than joint or extended households, which are common among the not- so-poor-and so lack the support of extended families in times of adversity * Are less able to migrate in search of better opportunities in rural or urban areas, and if they do migrate, are more likely to do so temporarily40 * Are especially vulnerable to adverse contingencies-chronic ill- ness, bad harvest, crop failure, death in the family, seasonal loss of employment- which often strike simultaneously and cause ex- treme deprivation41 * Are particularly vulnerable during certain seasons of the year, especially the wet, slack seasons (May through August in South Asia) when labor demand and wages are low, food prices are high, the harvest has come and gone, income flows are few, and illness and malnutrition often reach their peaks42 * Are more likely to belong to lower castes and as social outcasts are more likely to suffer both economic deprivation and social prejudice. Members of higher castes are more likely to be land- owners, moneylenders, merchants, and traders while lower castes are more often sharecroppers, agricultural laborers, and landless peasants. These characteristics and their implications for policy will barely be alluded to in the rest of this book. A full examination of their inci- dence and consequences would require another set of studies. None- theless they should be borne in mind as background for the more nar- rowly economic characteristics discussed here. Poverty in South Asia has many dimensions, and no single policy approach can be expected to solve all the problems that beset the rural poor. Conclusions Despite disagreement among the experts about the conclusions to be drawn from some of the available evidence, the following general points about rural poverty in South Asia seem reasonably clear: * However poverty is measured, a large percentage of rural house- holds are poor and their absolute number has been increasing in the subcontinent. Rural Poverty: Incidence and Trends 41 •Where agricultural output per capita has been stagnant or declin- ing, the living conditions of the rural poor have deteriorated (par- ticularly those of the small landholders and agricultural laborers who make up the lowest two to three income deciles of the rural population). These conditions have prevailed in Bangladesh and in the majority of the Indian states, especially those in eastern India-Assam, Bihar, Madhya Pradesh, Orissa, eastern Uttar Pradesh, and West Bengal-since the early 1960s. Together, these areas account for most of the rural population of South Asia. *Where agricultural output per capita has been growing relatively rapidly (over 2 percent a year), this growth has benefited all classes of the rural population. Even the poorest agricultural labor households have gained, directly or indirectly, although the extent of the benefit accruing especially to the poorest group is uncertain. These conditions have prevailed in only a few regions of the subcontinent, including the Indian and Pakistani Punjabs and the Indian states of Gujarat, Haryana, Himachal Pradesh, and Rajasthan, which account for a small minority of those living in rural South Asia. * The rural poor consist predominantly of agricultural labor households (mainly the landless and some who cultivate very small plots of land) and small landholders with cultivated hold- ings of less than 5 acres. This group is largely made up of ten- ants, but includes some who own their land. * The very poor also exhibit socioeconomic characteristics that set them apart from other groups and increase both the severity of their problems and the intensity of their need for greater earn- ings. Notes 1. For India see Bardhan (1970), Raj (1976), Minhas (1974), Lal (1976), Rao (1977a and b), and articles in Srinivasan and Bardhan (1974) and ILO (1977a). For a comprehensive treatment of the available evidence on India, see Ahluwalia (1978). For Pakistan, see S. M. Naseem (1973, 1979, and in ILO 1977a), T M. Khan and Bose (1968), and Khandker (1973). For Bangladesh see Bose (1968, 1974b), ILO (1977a), Alamgir (1975, 1978), Alauddin (1975), and Chaudhry and Chaudhry (1974). 2. The most persistent advocates of this view have been Griffin and Khan (in ILO 1977a), Griffin and Ghose (1979), Naseem (in ILO 1977a), and Alamgir (1978). Data purporting to support this argument are given in ILO (1977a). 3. If output growth has been outstripped by population growth, then, other things being equal, one would expect poverty to increase. 42 Notes 4. The poverty line was estimated at Rsl5 per person for thirty days at 1960-61 rural prices. Any line is necessarily arbitrary in all work of this kind, but Ahluwalia (1978, pp. 6-7) states that this particular figure has an "estab- lished pedigree" in Indian literature. It is supposed to correspond to an ex- penditure level at which food consumption on the average provided a "norm of 2,250 calories per day," and represents "an extremely low level of living." 5. The Sen index is defined as I = x [c* - c(I - g)] where x = percentage of population below poverty line, c* = the poverty line, c = mean consump- tion of the poor, and g = the Gini coefficient of consumption of the poor. See A. K. Sen (1973) for an explanation of the superiority of this measure both to the usual head count of those below an arbitrary poverty line and to stan- dard measures of relative inequality (such as the Gini coefficient by itself). 6. This is true of all studies that specify more than one poverty line. Be- cause each poverty line is somewhat arbitrary, calculations of the number of poor from such studies are also arbitrary. See appendix A for the implica- tions of this arbitrary choice and of the use of calorie intake to define a pov- erty line. 7. Naseem also examined the evidence on real wages and found that the evidence for Punjab did not provide a firm basis for the conclusion that real wages had increased significantly during the period under consideration (1960-73). This is contrary to his own data (based on three observations over thirteen years), which show increases in real wages of 2.1 percent a year be- tween 1960 and 1966 and 2.8 percent between 1966 and 1973. He finally compromised by saying "the evidence does not rule out the possibility of some increase in the real wages" (in ILO 1977a, p. 58). 8. See Khan and Bose (1968), Bose (1974b), Salimullah and Islam (1976), and A. R. Khan (1972). 9. Khan (in ILO 1977a) argues that most agricultural workers work as hard during the peak season as some rickshaw pullers whose calorie requirements are as high as 5,000 calories per day. Men engaged in exceptionally heavy work, such as harvesting without mechanical help, have reached levels of 5,000 calories per day over a few days. These include smallholders and land- less laborers, who are among the poorest classes in Bangladesh. But having made this seemingly telling point, he proceeds to define the "extremely poor" as those consuming only 1,720 calories. Presumably using the higher calorie figures would make everyone in Bangladesh either poor or undernourished or both. 10. If Ahluwalia's estimate of the percentage of the 1973-74 rural popula- tion in poverty were applied to the rural population in 1981, an estimated 252 million people in rural areas of India would be living below the poverty line he defines. A similar exercise using Khan's conservative percentage for 1973-74 gives a 1981 figure of 71 million in poverty for Bangladesh; using Alauddin's (1975) equally conservative Pakistani percentage for 1971-72 gives 46 million in poverty in that country in 1981. These figures cannot be added because they are not based on comparable poverty lines. 11. There has been a significant decline in area and output of minor cere- Rural Poverty: Incidence and Trends 43 als (such as jowar, ragi, barley, and gram), pulses, oilseeds, and jute in India. See Bhalla and Alagh (1979). 12. There are serious problems in using these data, and alternative meth- ods give different rates. Growth rates have been calculated by first deflating the agricultural NDP, with the use of state-specific agricultural price deflators, and then estimating an exponential trend with the use of ordinary least squares. 13. These results for 1960-70 are consistent with the findings of Katyal and Sood (1976). They have done no comparable analysis for later periods. 14. In 1978-79 the worst drought in recent history caused output to de- cline by 15-50 percent for the kharif season in some states. 15. Ahluwalia's state-level results, based on data for 1953-54 to 1973-74, have been questioned by Griffin and Ghose (1979), who argue that if the ref- erence period for the analysis is restricted to 1960-61 to 1973-74, there is no significant inverse relationship between the incidence of poverty and the rate of agricultural growth. The dropping of the first two observations to ar- rive at this conclusion seems arbitrary. 16. Bardhan (1971) reported an increase in the incidence of poverty for these two states between 1960-61 and 1967-78, while Rajaraman (1975) re- ported a similar result for Punjab between 1960-61 and 1970-71 on the basis of small samples. Ahluwalia (1978) points out that although the conclusions of these studies stand as long as their particular starting and finishing points are used, they are misleading about underlying trends. One of the general problems in evaluating this type of evidence is that the choice of different starting and end points leads different studies to different conclusions based on the same data, when the data in fact show only fluctuations rather than broad trends. 17. For this argument to hold, the in-migration would have to be perma- nent to allow migrants to be included in the NSS sample over time. 18. See especially Lal (1976), Johl (1975), Kahlon and Singh (1973), and Sidhu and Grewal (1981). 19. Distributions of real consumption were not used because they required fractile-specific price indices that were unavailable. It is therefore possible that the data are somewhat biased. Further, the distribution of consumption tends to be less unequal than the distribution of income, so that disparities in welfare are underestimated by the coefficients used. The data do, however, provide a basis for gauging broad trends. 20. Since wages vary by season, region, and type of task performed, an unweighted index of average wages is misleading. The data needed to assign appropriate weights to different wage series would be very difficult to obtain, however. 21. Naseem (in ILO 1977a), using rural-labor-specific price indices pro- vided by Chaudhry and Chaudhry (1974) to deflate money wages, calculates annual increases in real wages of 2.6 and 2.8 percent for permanent and ca- sual laborers respectively in the Punjab between 1966 and 1973. He nonethe- less erroneously concludes that "the evidence for Punjab, the most prosper- 44 Notes ous province, does not provide a firm basis from which to conclude that real wages have increased significantly during the period under consideration. It should, however, be noted that the evidence does not rule out the possibility of some increase in real wages in the Punjab" (p. 58; italics added). 22. The major studies reviewed by him include his own work (1970, 1973a) based on the NSS data; Jose (1974) and Krishnaji (1971) based on the agricul- tural wage data collected by the Ministry of Food and Agriculture and pub- lished in Agricultural Wages in India Bulletin; and Bardhan (1976) based on wage data available in various farm management surveys. 23. According to Bardhan (1970) real wage rates rose in Kerala but de- clined in the Punjab between 1956-57 and 1964-65, in spite of higher growth in agricultural output in the Punjab. In a subsequent study covering a longer period (1956-57 to 1970-71), Bardhan (in Srinivasan and Bardhan 1974) found the initial decline in real wage rates in Punjab and Haryana to be temporary, with real wages rising after the mid-1960s. But he found that real wages rose faster in Kerala in spite of its higher proportion of landless labor and slower rates of growth. He attributed this to the bargaining power of Kerala's agrarian trade unions. 24. A similar conclusion was stated earlier by Herdt and Baker (1972). The 1970-71 25th NSS Round was geared toward the "weaker sections" so that it included both small cultivators and noncultivating wage earners, and the 1956-57 12th NSs Round dealt with agricultural labor households. These sur- veys are thus especially useful for assessing changes in the incomes of the poorest groups. 25. See Griffin (1974), Byres (1972), Farmer (1977), and Frankel (1971) among others for relevant bibliographies of what might be termed "the new orthodoxy of rural poverty." 26. 1. G. Patel is perhaps too sanguine, but he has put the matter suc- cinctly: "While those glib of tongue may repeat the cliche of the rich getting richer and the poor getting poorer, and while the first part of this statement is almost universally true, I doubt if there are many countries where the sec- ond part of the charge is even remotely valid. The poor may seem poorer to eyes accustomed to plenty and this is perhaps as it should be. But impa- tience with what could be should not lead to the denigration of the good that has already taken place." (Eastern Economist, March 26, 1982.) 27. Cultivating households with less than half an acre were classed as agri- cultural labor households, both because they derived their main income from labor and because their level and pattern of expenditure were similar to those of labor households. 28. The lower end of the poverty scale refers to the lowest two income dec- iles of the population; for these two decile groups, annual per capita incomes were Rsl80 and Rs240 for cultivators, compared with Rs360 or Rs420 for agricultural laborers in 1968-69. 29. In 1964-65 they included landless wage earners (10.5 percent), land- less wage earners who were also sharecroppers (1.3 percent), domestic ser- vants who were also cultivators (3.6 percent), wage earners owning some land Rural Poverty: Incidence and Trends 45 (6.2 percent), and wage earners-cum-sharecroppers owning some land (3.8 percent) (A. R. Khan in ILO 1977a, p. 150). 30. Naseem (in ILO 1977a) cites Eckert's (1972) study in the Punjab, which shows 1970-71 per capita incomes as follows: large farmers, Rsl,102; small farmers, Rs318; tenants, Rs200; agricultural laborers, Rsl92 and Rsl73 for permanent and casual labor respectively. The surveys of the Punjab Board of Economic Inquiry give the following: large farmers, Rs933; small farmers, Rs633; tenants, Rs343; and agricultural laborers, Rsl80. 31. It is beyond the scope of the present study to present detailed evidence on these other characteristics. For more detailed data, see Visaria (1979, 1980) and World Bank (1980). Lipton (1983a) has paid special attention to the question of whether there are any discontinuities that separate the "poor" from the "poorest" in terms of the socioeconomic characteristics cited here. 32. In Karnataka, for example, laborers and tenants had 67 percent higher levels of infant and child mortality rates than owner-cultivators farm- ing more than three acres (see Lipton 1983a). Also see Visaria (1980) for evi- dence from West India and Chen (1973) for Bangladesh. 33. Morbidity rates among the lowest income groups were 12.8 percent higher than the average, even in Sri Lanka where the poor have access to extensive health care systems. In India as many as 75 percent of all hospitals and 89 percent of all hospital beds were in urban areas having only 20 per- cent of the population (Grawe, Krishnamurti, and Baah-Dwomoh 1979). 34. The poorest have slightly larger households than the poor, and the poor have larger households than the nonpoor, because household size and per capita income are positively correlated. But household size and assets (mainly land) are negatively correlated in general. Thus land available per household exaggerates per capita poverty, while income per household un- derstates it. See Lipton (1983b) for a thorough statement. 35. For all of India in 1972-73, the ratio of children per rural household drops from 0.5-0.54 for the four lowest classes of monthly expenditure per person to 0.29-0.33 for the three highest classes in the NSS samples. The per- centage of children in the rural population drops from around 50 percent for the class with monthly outlays of RsO-34 per person to 31 percent for the Rsl5O+ class in Maharashtra (see Lipton 1983a). 36. In India over a third of all household members are between the ages of 5 and 14 in the poorest decile, compared with a quarter or less in the rich- est decile (Lipton 1983a). 37. Dependency ratios fall dramatically as per capita income increases but rise as size of landholding increases. 38. This is so despite the fact that women from poorer households have very limited opportunities for work outside their homes owing to social sanc- tions (especially in Muslim households and areas) and are heavily burdened with child care, family health care, and pregnancy. Their poverty drives them into higher participation rates than those of better-off women and reduces the amount of time spent caring for the health and informal education of their children. In the poorest households, greater labor force participation 46 Notes results from attempts to overcome the high burden of dependency (Lipton 1983b). 39. Literacy rates and education are generally positively associated with mean family per capita expenditures. In Gujarat, for example, only 22 per- cent of males and 8 percent of females in the lowest income decile partici- pated in primary education compared with 54 percent and 51 percent respec- tively for the highest income decile (Visaria 1977a). 40. See Lipton (1983a and c), Visaria (1980), and Connell and others (1976). Long-term rural-urban migration tends to be highly selective by age and sex, with a high proportion of men between the ages of 15 and 30. The proportion of women in migration streams has been rising rapidly, but they are not the poorest. 41. The illness of adult workers results in losses, and crop failures lead to the loss of on-farm wage employment. The impact of multiple contingencies on the poor is especially severe because their incomes are already low (and hence vulnerable) and they have few or no assets. These interlocking contin- gencies can destroy the whole economic status and well-being of poor fami- lies. The descent into poverty often starts with a "bad year" on top of illness and debt. (See Lipton 1983c, Alamgir 1980, R. H. Chaudhry 1978, and Chen 1973.) 42. Physical weakness, malnutrition, sickness, and death all peak in the wet season, months before the kharif harvest. Women and children are espe- cially vulnerable to hardship in this period when public services are least likely to be effective. (See Chambers 1979, Longhurst and Payne 1979, R. H. Chaudhry 1978, and especially Pacey, Longhurst, and Chambers 1981.) The Agrarian Structure To evaluate the prospects and problems of small farmers and the landless in South Asia, one has to understand the agrarian structure within which they seek their livelihood and the ways this structure in- fluences the distribution of the benefits of economic progress:' * The agrarian structure determines the ownership of rural assets and, in turn, the way rural people derive their livelihood, whether from land, from other assets, or from labor. * It shapes the nature and pace of technological change and the ways in which the benefits (and costs) of these changes are distrib- uted. * It determines the hierarchy of rural social and political relation- ships, which closely follow control over land. In general, the term "agrarian structure" encompasses all the institu- tions relating to land, labor, and the productive assets by which rural people earn a living, together with the network of economic, social, and political relationships linking different rural groups. This chap- ter focuses on two of the central features of South Asia's agrarian structure: the access to land, as owner or operator, and the access to employment in agricultural and nonagricultural rural sectors. In rural South Asia, as in other developing areas, disparities in landholdings produce disparities in incomes, and control of land usu- ally coincides with control of local institutions. Access to other rural factors of production is largely determined by the size of land- holdings. Variables such as the quality of land, the nature of tenur- ial arrangements, and the intensity with which land and technology are used in different areas or by different groups may offset (or strengthen) the effects of the size of holdings. Nonetheless, the size distribution of landholdings remains the single most important deter- minant of agrarian relationships. 47 48 Two Theories of Agrarian Change This chapter begins with a brief discussion of alternative explana- tions (more accurately, theories) that purport to explain the trends in the patterns of landholding and the changing agrarian structure in South Asia. Two Theories of Agrarian Change In recent years the rural economy of South Asia has been character- ized by high population growth rates in all areas and rapid changes in production technology in many regions. These factors form the starting points for two alternative but not mutually exclusive explana- tions of the patterns of landholding and landlessness in the subconti- nent: the Malthusian or population pressure theory and the Marxian or class-polarization theory. The population pressure theory is based on the observation that the combination of rising rural population and a limited land base re- duces the average size of owned holdings as they are apportioned among heirs, and thus adds to the problem of rural poverty. If the average rate of population increase is the same among large and small landholders, families with initially large holdings will be better off than others, but the size of all holdings will fall in absolute terms; there will be no change in the pattern of concentration. If wealthier families with larger holdings have more surviving sons whose off- farm incomes can be invested in additional land, or if they divide their land less often than other families, they may be able to compensate for the process of partitioning or actually to enlarge their holdings. If the rich wish to buy land, the poorest may be obliged to sell: parti- tioning by the smallest holders tends to produce holdings too small to withstand crises such as crop failures or the death of draft animals. Lack of new land prevents expansion of the cultivable area as the labor force grows and forces more people to rely on wage employ- ment. The class-polarization theory observes that technological changes lead to the commercialization of agriculture and to increases in the use of purchased inputs, in the proportion of total output sold for cash, and in the use of hired labor. As commercialization proceeds, the control of essential resources, including land, becomes more con- centrated. Small farmers (especially tenants) facing competition from commercialized agriculture run increasingly into debt. In an eco- nomic system based on private property and with a high incidence of tenancy, the demand for credit leads to mortgages, foreclosures, and the alienation of land from small cultivators to big landlords and moneylenders. Disparities in incomes and in the ownership of land and other assets increase as small cultivators are driven into the classes The Agrarian Structure 49 of sharecroppers and landless laborers, while the land they formerly owned becomes concentrated in fewer hands. Much of the literature on agrarian change in South Asia subscribes to one or the other of these theories.2 Attwood (1979, pp. 496-97) demonstrates that their relative merits cannot easily be tested on the basis of available data, since most of the testable hypotheses are com- mon to both theories. Thus both predict that * The proportion of landless villagers will increase. * This increase will be due mainly to downward mobility among cultivators, especially smallholders.3 * Most families will move downward, very few upward. * Big landholders will make a disproportionately large share of the land purchases while smallholders will make a disproportionately large share of the sales. * The mean or median size of holdings will decline. But there are three predictions on which they differ: * The Marxian theory predicts that ownership holdings are bound to become more concentrated, while the Malthusian theory sug- gests that although holdings of all sizes are likely to shrink, there may be little or no change in the overall degree of concentration. * The Marxian theory suggests that the main cause of downward mobility is land sales by smaller to larger owners, while the Mal- thusian theory suggests that the main cause is the partitioning of holdings among multiple heirs (sales may, but need not necessar- ily, take place). * The Marxian theory suggests that big holdings will get bigger and small ones smaller, so that their heterogeneity (standard devia- tion from the mean) will increase, while the Malthusian theory suggests that, as partitioning will reduce most holdings to a frac- tion of their original size, heterogeneity will decrease. These last two hypotheses cannot be tested with aggregate data; it is necessary to trace what happened to particular landowning house- holds over time. One useful analysis of this type is Attwood's (1979) study of the landholdings of the same households and their heirs over fifty years (1920-70) in a village in Maharashtra. He found both up- ward and downward mobility: in terms of size of landholdings, some of the poor had become richer and some of the rich had become poorer. Land had not become concentrated in only a few hands. Nearly 25 percent of the landless in 1920 had land in 1970 (a change predicted by neither theory), while 44 percent of the originally landed had lost their land. Those with the smallest holdings in 1920 50 Two Theories of Agrarian Change had enlarged them, while the largest landowners of 1920 had lost most of their holdings. Contrary to the Marxian theory, most land purchases were by smallholders and most sales were by large owners. Consistent with the population pressure theory, repeated partitioning was an important cause of downward mobility. This leaves the hypothesis that the distribution of owned holdings will (or will not) become more concentrated. The available evidence on trends in ownership in India and Pakistan suggests that the distri- bution has become less concentrated over time, a trend that is not con- sistent with the Marxian perspective. Vyas (1976) showed that, over a forty-year period in Gujarat, mar- ket transactions in land were few and most changes in holdings were made in response to land reform legislation. Landless households ac- quired land much more frequently than small households lost their land. Evidence of the landless acquiring small plots of land is also given for Gujarat by Dantwala and Shah (1971) and for Haryana by Sheila Bhalla (1977).4 Most of the additions to the ranks of agricul- tural laborers came from small-farm and artisan households. Vyas noted the strong possibility that many workers reported themselves as agricultural laborers not so much because they were landless as be- cause they lacked enough land to provide a living. He found no evi- dence of large-scale displacement of small farmers by larger farmers.5 As Vyas points out in other studies (1979, 1987), a decline in the concentration of owned holdings can be caused by market processes, whereby marginal and small farmers may have purchased land from larger holders; institutional processes, whereby larger holdings may have been broken up by land reform and allocated to smaller holders; and demographic processes, whereby holdings have been subdivided in response to increases in population and the size of rural households. The evidence suggests that all three processes have been at work in India, and that by and large land markets have worked in favor of small and marginal farmers rather than against them. Although there was some evidence in the 1950s of larger holders purchasing land from smaller owners, this process came to a halt during the 1960s. Fear of the imposition of ceilings on landholdings, opportunities for profitable intensification of land use, and tenancy legislation that made it easier for tenants to acquire land in a number of states all made it less desirable and more difficult for larger owners to enlarge their holdings. Even where technological change has been rapid, the "proletariani- zation" thesis cannot be accepted, although the superficial evidence- an increase in the "agricultural laborer" category of the rural work force and a downward shift in the structure of farm holdings-seems to support it. Sheila Bhalla's studies (1976, 1977) of changing agrar- The Agrarian Structure 51 ian structure and relations in Haryana during the 1960s, when the Green Revolution was in full spate, found that "the more-than- doubling of the number of male agricultural laborers from 1961 to 1971 . . . cannot be attributed to the loss of land previously owned or leased in by households which are now landless labor households," and that "no household cultivating under 15 acres today, and no land- less labor household, lost any land by sale to pay off debt, or sale for any other reason. None lost land through foreclosure of mortgages and none leased out land previously self-cultivated."6 The same ana- lyst also found that the decline in the average size of small holdings in Haryana was caused by the subdivision of plots among families as farming became more profitable.7 Other microeconomic evidence, however, suggests that big farmers in Punjab and Haryana may have been acquiring more land for culti- vation, though it is not certain from whom.8 Where technologies have changed rapidly-East and West Punjab, Haryana, and Tamil Nadu- the processes predicted under both theories have accelerated, while operational landholdings have definitely become more concentrated. Further displacement of tenants, a sharp increase in the use of hired labor, and, among larger holders, diversification of assets away from land have all been observed, especially in irrigated areas. These trends lend some credibility to the Marxian perspective. Similar changes may be expected to accompany agricultural innovations in other areas. Government policies, however, notably in India, where they have been of special help to somewhat better-off peasants, may have limited the process of polarization. It thus appears that both population pressures and technological change have helped to alter the agrarian structure, but that each of these determinants explains only certain aspects of the transforma- tion of the pattern of South Asian landholdings. But what changes have there been among the landless? The relevant evidence and data sources have been examined in Singh (1988a). The main findings from that study are summarized here. The Structure of Ownership Holdings Land accounts for two-thirds of the value of all assets in rural areas. Those who control land, control the major source of income as well as other rural institutions. Initial disparities in landownership beget disparities in rural incomes, and access to land in the sense of who operates it and under what conditions is often the main determinant of the distribution of rural incomes. Data on landownership are notoriously unreliable, are seldom com- parable over time or across countries, and reveal an enormous diver- 52 The Structure of Ownership Holdings sity in the subcontinent. Nonetheless they permit some valid compari- sons. India The data from India (tables 3-1 and 3-2) are the most reliable. They reveal that inequality in the distribution of landownership was fairly high at independence but actually declined in the 1960s. In spite of this, ownership remained extremely uneven-about 78 percent of all rural households owned no land or fewer than 2 hectares and only 25 percent of the area, while 3 percent of all households owned more than 8 hectares and nearly 30 percent of the area. The number and proportion of marginal owned holdings (those less than an acre) in- creased significantly and their average size declined, whereas the number of small owned holdings (1.1 to 4.9 acres) increased, their proportion declined, and their average size remained almost con- stant. The importance of larger owned holdings decreased both in number and in total area held with few exceptions, and the absolute number as well as proportion of households not owning any land (about 10 percent of all rural households) declined significantly. Fig- ure 3-1 gives the percentage share of various size groups in the total number and area of owned holdings. The medium-to-large owned holdings (those more than 4 hectares) accounted for 10 percent of the total number and 54 percent of the total area of owned holdings in 1971, while the shares of this size group in 1982 declined to 7 and 48 percent respectively. The proportion of marginal holdings in- creased from 63 percent of the total number in 1971 to 67 percent in 1982, and from 10 percent of the total area owned in 1971 to 12 percent in 1982. This evidence contradicts the general impression that landownership is becoming more unequal and the broader gen- eralization that landlessness has been increasing via the "proletariani- zation of the rural masses" through loss of landownership. There is no evidence of a direct or simple relationship between the inequal- ity of landownership and agricultural performance in the past two decades. Bangladesh For Bangladesh there are few reliable data on landownership. The most recent data are from the Land Occupancy Survey (1979) and the 1983-84 Agricultural Census. They reveal that ownership is highly skewed, more so than in India, with 90 percent of all rural households owning zero or fewer than 5 acres of land and only 51 percent of the area, while 2.5 percent own more than 10 acres and Table 3-1. India: Cumulative Distribution of Rural Households and Landownership, 1961-62 and 1971-72 (percent) North India West India South India East India All India 1961-62 1971-72 1961-62 1971-72 1961-62 1971-72 1961462 1971-72 1961-62 1971-72 Size of holding (hectares) N A N A N A N A N A N A N A N A N A N A 0 4 0 4 0 13 0 11 0 18 0 14 0 12 0 8 0 12 0 9 0 0-1 58 11 64 16 44 7 44 9 70 10 70 12 67 14 71 22 60 9 63 13 0-2 78 30 81 49 57 19 60 38 81 21 82 28 84 37 88 54 75 24 78 26 0-4 94 65 93 73 80 46 78 59 91 43 92 49 95 64 96 77 90 52 90 62 0-10 99 85 99 92 93 70 94 83 98 73 98 76 100 89 100 95 97 79 98 85 All holdings 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 Number of owned holdings (millions) 16.6 17.4 17.0 18.8 19.6 20.8 19.3 20.8 72.5 77.8 Total area owned (million hectares) 24.2 23.2 56.2 58.8 26.8 27.1 21.0 19.0 128.2 128.1 N: Cumulative percentage of total number of owned holdings. A: Cumulative percentage of total area owned. Sources: National Sample Survey (NSS) 17th (1961-62) and 26th (1971-72) Rounds. The percentages have been rounded to the nearest whole number. Table 3-2. India: Percentage Change in Number of Owned Holdings and Area Owned by Size of Holdings Average Percentage change, Average Pe-rcentage cha-nge, Average Size of holding area (acres), 1961-62 over 1953-54 area (acres), 1971-72 over 1961-62 area (acres), (acres) 1953-54 Number Area 1961-62 Number Area 1971-72 Marginal (<1) 0.27 53.5 21.5 0.21 51.2 -3.0 0.14 Small (1-4.9) 2.62 28.8 28.0 2.60 22.0 21.7 2.60 Medium (5-14.9) 8.45 16.7 15.2 8.44 4.3 2.5 8.29 Big (15-49.9) 24.80 4.8 2.3 24.20 -10.2 -11.5 23.87 Large (50+) 88.71 -37.7 -34.0 80.96 -20.0 -27.0 73.87 Total 6.25 31.0 4.1 4.97 26.6 - 2.1 3.84 Sources: NSS Report 36, 8th Round (1953-54); NSS Report 144, 17th Round (1961-62); and NSS Report 215, 26th Round (1971-72). Figure 3-1. India: Percentage Distribution of Owned Holdings and Area Owned by Size of Holding, 1961-62, 1971-72, and 1982 Marginal Small Semimedium Medium Large Number Area Number Area Number Area Number Area Number Area State owned owned owned owned owned Haryana 1971-72 1 1982 Punjab JI4JI~~~~ ~I~~jI~ 3930 1961-62 C 1 38 1971-72 25 8 8 13 2 2 24 1982 1 E 1 2 19 West Bengal 69 26~~~2 1961-62 178E 18 i 17 263 1971-72 27 13 30 29~~~~~2 1982 _ L 2 All India 1961-62 5 1 13 21 39 I 28 1971-72 6710 15 12 2 8 I i 2 3 1982 1 23 L ANote: Marginal holdings are less than 1.01 hectares; small, 1.01-2.02 hectares; semimedium, 2.03-4.04 hectares; medium, 4.05-10.12 hectares; and large, 10.13+ hectares. Sources: National Sample Surveys and Sarvekshana 5 (3 and 4). 55 56 The Structure of Ownership Holdings 25 percent of the area. Nearly a third of all rural households owned no land other than their house lots (table 3-3). Supplementary evi- dence suggests that ownership may be becoming more concentrated through a slow process of land alienation, with smaller owners being forced to sell their land to larger owners in recent years. West Bengal and Bangladesh Because agrarian conditions in Bangladesh and West Bengal (India) are similar, it is worth comparing the structure of landownership in the two areas (table A5 presents the data). If anything, landownership was marginally less egalitarian in Bangladesh in 1977 than in West Bengal in 1971. This difference may have become more pronounced in recent years. Although both have a heavy concentration of owner- ship in very small holdings, the number and area under larger hold- ings is declining in West Bengal, while it has actually increased in Ban- gladesh. In both regions, over 50 percent of all holdings are smaller than 1 acre, although the percentage of households owning no land at all seems larger in Bangladesh. Given the similarity between the two areas and the relatively greater reliability of the data for West Bengal (especially with regard to changes over time), it might be reasonable to use the latter as a rough guide to the kinds of changes that may have occurred in Bangladesh in the past decade. In West Bengal the average size of holdings fell for all size groups; the number of marginal (less than 1 acre) and small (1-5 acres) holdings rose, as did their share of the total owned area; the number of medium (5-10 acres) and large (10 acres and more) holdings fell, as did their share of total area; and the concentra- tion of landholdings consequently fell, as did the percentage of rural households owning no land. Similar trends would, in principle, be likely to have continued in Bangladesh in the past decade, with certain exceptions. Land reform in West Bengal had relatively greater success, and the devastating civil war, floods, and famine in Bangladesh are likely to have increased the proportion of those owning no land and the concentration of owner- ship, as the Land Occupancy Survey data seem to suggest. Recent data reported by Hossain (1988) show that both the number and share of households owning more than 5 acres increased between 1979 and 1983-84. Pakistan Data on landownership in Pakistan are also less reliable than in India. The information that is available shows that ownership was heavily The Agrarian Structure 57 concentrated (especially in Sind) at independence and remained so for more than two decades: over two-thirds of the owners owned fewer than 5 acres and accounted for over 18 percent of the area, while less than 1 percent owned 25 acres or more and accounted for over 12 percent of the area. This concentration was considerably re- duced by the 1976 land reforms, but the concentration of land in the hands of large (25-50 acres) and very large (more than 50 acres) own- ers remains a salient feature that sets Pakistan apart from the rest of the subcontinent (tables 3-4 and 3-5). Yet this inequality has not been a barrier to high rates of growth. East and West Punjab Contrary to accepted notions, there are many basic similarities in the ownership patterns in West Punjab (Pakistan) and East Punjab (India).9 As Singh (1988a) shows, nearly two-thirds of all owners or landowning households in both areas owned fewer than 5 acres; the number and total area of these holdings has been increasing while their average size has declined or remained constant; the number of larger owned holdings has declined, but their total area has decreased less or increased. Thus the average size of medium (5-25 acres), large (25-50 acres), and very large (more than 50 acres) holdings owned by households (East Punjab) or individuals (West Punjab) has in- creased, with the largest increase occurring in the medium-size groups. The main difference between the two regions lies in the average size of farms. Very large farms still exist in Pakistan, where in West Punjab in 1976 more than 10,000 individuals owned more than 150 acres each; in the Indian Punjab fewer than 8,000 households owned holdings of more than 50 acres each.'° Ownership holdings of fewer than 5.0-6.6 acres make up about two-thirds of all holdings in both Punjabs, but in West Punjab their average size is 2.5 times larger. Dif- ferences in the importance of large owners and the average size of small owned holdings have major consequences for agrarian struc- ture and for the policies followed in the two areas, as will be shown later. Conclusions According to Singh (1988a), the broad evidence on landownership in rural India shows that the basic patterns of ownership after indepen- dence persisted for four decades, despite movements in the direction of a more even distribution because of population pressure. Never- theless, the number of marginal and small owners as well as the total Table 3-3. Bangladesh: Percentage Distribution of Total Owned Land in Rural Areas by Size of Holding, 1977 and 1978 Percentage of absolute Number of households Total area owned' change for group w ~~~~~Size of holding °° (acres) 1977 1978 1977 1978 Number Area 0 11.07 14.69 - - 34.8 - 0-1 47.44 44.68 9.30 8.33 -4.4 33.3 1-2 16.43 15.21 14.43 12.78 -6.0 -4.8 2-3 8.91 8.69 13.18 12.28 -1.0 0.2 3-4 5.27 5.16 11.13 10.29 -0.5 -0.5 4-5 3.29 3.08 9.00 7.93 -4.6 -5.2 5-6 2.09 2.11 6.90 6.61 2.4 3.1 6-7 1.43 1.44 5.69 5.40 2.4 2.0 7-8 1.02 0.92 4.65 3.98 -7.5 -7.9 8-9 0.69 0.79 3.60 3.86 15.9 15.3 9-10 0.42 0.56 2.46 3.06 34.0 33.7 10-11 0.34 0.51 2.15 3.05 52.5 52.7 11-12 0.29 0.32 2.03 2.14 14.7 13.1 12-13 0.16 0.30 1.18 2.17 100.0 97.6 13-14 0.22 0.22 1.82 1.69 0.0 0.2 14-15 0.13 0.16 1.19 1.37 25.0 23.6 15+ 0.80 1.16 11.29 15.07 0.0 43.7 All 100.00 100.00 100.00 100.00 1.5 7.6 Total (millions) Number 11.85 12.03 - - 1.5 - Acres - - 19.35 20.81 - 7.6 0-5 acres Including landless 10.95 11.01 10.54 10.74 0.6 1.9 Excluding landless 9.64 9.24 10.54 10.74 -4.2 1.9 5+ acres 0.90 1.02 8.81 10.07 13.3 14.3 Overall Gini coefficient of concentration 0.6577 0.6842 -Not applicable. a. Including the house lot. Source: Jannuzi and Peach (1980), appendix tables D.I and E.l. Table 3-4. Pakistan: Landownership by Size of Owned Holding, 1950s and 1970s 1950s 1972 1976 Size of holding Size of holding Size of holding (acres) N A (acres) N A (acres) N A 0-5 64.4 15.3 0-5 66.5 17.6 0.0-6.25 70.8 24.9 5-25 93.1 47.0 5-25 94.5 56.7 6.25-12.5 88.3 46.2 25-100 98.8 68.8 25-100 99.0 82.8 12.5-25.0 95.9 64.3 100-500 99.9 84.6 100-150 99.4 89.1 25.0-50.0 98.5 77.5 500+ 100.0 100.0 150+ 100.0 100.0 50.0+ 100.0 100.0 Number of owners (millions) 5.07 10.06 10.3 Total area owned (million acres) 48.64 67.64 60.13 N: Cumulative percentage of total number of owners. A: Cumulative percentage of total area owned. Sources: Pakistan Land Reform Commission data in M. H. Khan (1981), landownership tables 3.1-3.5; 1972 Agricultural Census, annex. Table 3-5. Pakistan: Distribution of Owners, Area Owned, and Average Size of Owned Holdings in Punjab and Sind, 1971 and 1976 Punjab Sind 1971 1976 1971 1976 Size of owned holding Average size Average size Average size Average size (acres) N A (acres) N A (acres) N A (acres) N A (acres) 0-6.25 60.4 20.6 1.98 69.1 26.1 1.98 36.7 5.6 2.94 40.3 8.2 3.58 6.25-12.5 25.7 25.0 5.65 19.6 24.3 6.32 23.2 10.5 8.74 23.9 12.5 9.16 12.5-25.0 9.9 20.7 12.16 7.8 18.7 12.93 19.7 19.4 18.93 17.6 18.0 17.96 c 25.0-50.0 2.8 12.9 27.12 2.3 12.7 28.76 11.6 20.3 33.64 10.1 19.3 33.61 50+ 1.2 20.8 98.12 1.2 18.2 80.89 8.6 44.0 98.84 8.0 42.0 93.69 All 100.0 100.0 5.8a 100.0 100.0 5.25a 100.0 100.0 19.28a 100.0 100.0 17.62a Number of owned holdings (millions) 6.27 7.45 0.59 0.66 Total area owned (million acres) 36.36 39.12 11.45 11.68 - Not applicable. N: Percentage of total number of owned holdings. A: Percentage of total area owned. a. Average number of acres for all size groups. Source: Pakistan Land Reform Commission data in M. H. Khan (1981), tables 3.3 and 3.4. 62 The Structure of Operational Holdings area they owned and the average size of their holdings appeared to increase, while the importance of large farms decreased. Meanwhile, the number and proportion of rural households owning no land declined significantly in many parts of India. It is generally agreed that the subdivision of landholdings and the sale and purchase of land have been more important in bringing about structural redistri- bution than the redistribution of land through tenancy reforms and legal ceilings on landholdings (C. H. H. Rao 1976). This state- ment is quite contrary to the uninformed rhetoric that suggests a growing concentration of landownership and a slow proletarianiza- tion of the small peasantry by large owners who have bought them off. The Structure of Operational Holdings The distribution of landownership provides a useful guide to the ownership of assets in rural areas, but it does not tell how these assets are used. Since land rental markets are a prominent feature of rural life in the subcontinent, one must look beyond the distribution of owned land to the distribution of operational holdings (those cultivated by either owners or tenants) to determine who has access to land and consequently who can benefit from strategies to make land more pro- ductive. India Table 3-6 shows the cumulative size distributions of operational hold- ings by region in India for 1961-62 and 1971-72 (including house- holds that operate no land). For the latter year several features stand out. * Between one-quarter and one-third of all rural households in the four regions operate no land. (Not all of these households de- pend wholly on wage employment, but many do.) • Another quarter to one-third of all rural households operate marginal holdings of less than 1 hectare, and a further 15-20 percent of all rural households (20-25 percent of cultivating households) operate small holdings of between 1 and 2 hectares. Holdings are especially small in the east and south. • There are very few medium to large holdings: only 4-8 percent of all rural households (5-12 percent of cultivating households) operate holdings larger than 4 hectares, except in western India where 22 percent (30 percent of cultivating households) operate holdings larger than this. The Agrarian Structure 63 * Marginal and small farmers account for 80 percent of all the rural households but only a third of the cultivated land. By contrast, the 10 percent of rural households with cultivated hold- ings larger than 4 hectares account for nearly 47 percent of the operated area. One might think that the size distribution of operational holdings would be less skewed than that of owned holdings, on the assumption that larger owners generally rent out land to smaller ones. The need to use land intensively, together with the diseconomies of scale associ- ated with the management of labor, would seem to favor relatively small holdings and promote transfers of land in this direction. Thus it is commonly believed that the agrarian structure consists of large landlords at one end and small tenants at the other, with owner- cultivators lying in between. The reality is far more complex, however. The percentage of rural households that operate no land is much larger (24-35 percent, de- pending on the region; see table 3-6) than the percentage owning no land (4-18 percent; see table 3-1). Further, although quite a large proportion of rural households (12-30 percent) own very small hold- ings (less than 0.5 acre), a much smaller proportion (only 3-11 per- cent) operate holdings of this size. It seems likely that because owners of very small holdings find them uneconomic to run, they rent them out to larger operators and earn their livelihood in some way other than farming." Thus, landlords and tenants cannot be divided into mutually exclu- sive social classes-a point that is often missed. Many of those with relatively large operational holdings are tenants of small owners. The analysis by Vyas (1979) of changes in the distribution of operational holdings tends to support this interpretation. His study reveals that the distribution of operational holdings in the 1960s was less skewed than that of owned holdings in only five states-Assam, Bihar, Orissa, Uttar Pradesh, and West Bengal-all in eastern India. It was more skewed in four states-Haryana, Karnataka, Punjab, and Tamil Nadu-all of which except Tamil Nadu experienced high rates of ag- ricultural growth during the same period. No clear conclusion emerges for the remaining states. What have been the changes in the distribution of operational hold- ings over time? Three trends are evident from the all-India data: * Although operational land is still very unequally distributed, ine- quality has been reduced somewhat over time. * The number of small and very small holdings increased during the 1950s, 1960s, and 1970s as a result of subdivision and the Table 3-6. India: Cumulative Distribution of Rural Households and Operated Areas, 1961-62 and 1971-72 North India West India South India East India All India 1961-62 1971-72 1961-62 1971-72 1961-62 1971-72 1961-62 1971-72 1961-62 1971-72 Size of holding (hectares) N A N A N A N A N A N A N A N A N A N A 0 23 0 26 0 2 1 0 25 0 33 0 35 0 29 0 24 0 27 0 28 0 0-1 56 10 61 13 42 3 43 3 67 9 67 11 63 13 68 20 56 7 61 9 0-2 76 29 80 42 56 10 59 18 79 21 82 30 82 38 87 55 74 21 83 30 0-4 90 56 93 68 74 25 78 37 91 42 92 59 94 66 96 80 86 41 90 53 0-12 99 91 100 94 94 66 96 75 89 76 99 84 100 93 100 97 98 77 99 84 All 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 Number of opera- tional holdings (millions) 11.6 12.7 11.9 14.1 13.7 14.9 13.5 15.3 50.8 57.1 Total area operated (million hectares) 25.4 22.7 57.9 58.1 28.0 26.4 22.1 18.5 133.4 125.7 N: Cumulative percentage of total number of operational holdings. A: Cumulative percentage of total area operated. Sources: NSS 17th (1961-62) and 26th (1971-72) Rounds; Dandekar and Rath (1971); Sirohi, Ram, and Singh (1976); and Reserve Bank of India (1976). Since 1971-72 statewide distributions have been unavailable. TIhese data were compiled by applying the changes in the distribution of operational holdings of various rates given in Sirohi, Ram, and Singh to the 1961-62 distributions from the NSS and Dandekar and Rath. The numbers of households were grouped by various regions using the Reserve Bank of India data. These data were then consolidated to get the all-India distribution. The Agrarian Structure 65 growth of population, but the average size of marginal holdings (fewer than 2.5 acres) also increased. * Larger holdings (more than 15 acres) fell proportionately (in both number and area), but their average size did not decline.)2 Figure 3-2 gives more recent data and shows that the area operated under marginal holdings rose from 9 percent in 1971 to 12 percent by 1982. The area under large holdings dropped significantly from 29 to 18 percent from 1961 to 1982. The view that small operators are losing their land because large ones buy them out is widely held among proponents of the class- polarization theory of changing agrarian structure and may be diffi- cult to dislodge. It provides a convenient explanation for the rise in the number and proportion of agricultural laborers in the rural work force-an increase too large to be accounted for solely by the growing number of people in agricultural labor households. Nevertheless, the evidence presented so far on the changes in the size distribution of holdings in India is not consistent with this perception. Changes in Punjab and Haryana have run contrary to the general findings noted above that marginal and small holdings have in- creased, large holdings have decreased, and inequality has declined somewhat over time (see Singh 1988a, table A.14). In this region, both the numbers and area of small owned and operational holdings de- clined in the 1960s, and inequalities as measured by Gini coefficients increased (see Sanyal 1977b; Sirohi, Ram, and Singh 1976). The num- ber of very small holdings of fewer than 2.5 acres rose (while their combined area declined), and the number of holdings in all other size classes decreased as did their average area, the decline being greatest in the large size groups, according to the National Sample Survey (Nss) data. A tendency toward an increase in both the number and the share of area accounted for by medium-size holdings (5-15 acres) in Punjab is supported by the NSS data as analyzed by Sanyal (1977b). Bangladesh Data on the size distribution of operational holdings in Bangladesh do not permit systematic comparisons between the size distributions of owned and operational holdings for the same year, nor do they show what proportion of rural households operated no land. The size distributions of operational holdings in Bangladesh in 1960-61 and 1967-68 are shown in table 3-7. By 1967-68 nearly 57 percent of these holdings were smaller than 2.5 acres, accounting for 21 percent of the cultivated area. Farms of less than 1 acre accounted for 25 percent of all holdings and less than 5 percent of the total culti- Figure 3-2. India: Percentage Distribution of Operated Holdings and Area Operated by Size of Holding, 1961-62,1971-72, and 1981-82 Marginal Small Semimedium Medium Large Number Area Number Area Number Area Number Area Number Area State operated operated operated operated operated Haryana 1971-72 3~43j jjj ~ jj 1981-82 tL 42 Punjab 1961-62 23 2 1 25 i3 29 1971 72 9 3 2 198182 i 14 2 West Bengal 1961-62 6114 3 3 1971-72 _ 4 l 23 X 13 E2..LJ 1981-82 16 1 1L All India 1961 62 23 12 2 1 14 21 1971972 22 18 2330 1981-82 Note: Marginal holdings are less than 1.01 hectares; small, 1.01-2.02 hectares; semimedium, 2.03-4.04 hectares; medium, 4.05-10.12 hectares; and large, 10.13+ hectares. Sources: National Sample Surveys and Sarvekshana 5 (3 and 4). 66 The Agrarian Structure 67 vated area. By comparison with 1960-61, the distribution of opera- tional holdings had become a little less unequal. Evidence for later years suggests a dramatic reversal. Early tabula- tions of the 1977 Agricultural Census show that the size distribution of operational holdings shifted quite dramatically against farms of fewer than 2.5 acres and in favor of medium-size farms (table 3-7)- reversing earlier trends during the 1960s, when both the number and total area of small farms had actually increased while that of larger farms had decreased. Combined with an increase in the average size of larger farms and a growth in their number, this evidence suggests that many small farmers either sold out their farms and became land- less or stopped operating. Nearly 0.82 million holdings have dropped out of this category-their operators probably joining the class of the landless in Bangladesh. This trend in the increase in the number of small holdings has continued into the 1980s; as a result the average size of small holdings (less than 1 acre) has declined in spite of an in- crease in the total operated area in such holdings. The data on operational holdings from the 1977 Agricultural Cen- sus together with the 1977 Land Occupancy Survey data on owner- ship holdings indicate that the distribution of operational holdings is more skewed than that of owned holdings. The implication is that, in Bangladesh as in India, very small owners are probably net lessors of land.'3 The gaps in the data are so wide and the available information so generally unreliable that any number of conjectures is possible. It seems likely, however, that the share of land operated in small hold- ings has declined dramatically and that land markets have worked against smallholders, pushing them into the rural proletariat. This is supported by fragmentary evidence on rural land sales and purchases in the early 1970s (see Singh 1988a, table A.17). West Bengal and Bangladesh Although the eastern parts of India, particularly West Bengal in the 1960s, experienced changes similar to those occurring in Bangladesh, later changes in Bangladesh were quite dramatic. As Singh (1988a, table A. 18) shows, although both West Bengal and Bangladesh in the beginning of the 1960s had fairly similar distributions of operational holdings, by 1977 the number of very small holdings in Bangladesh had decreased while those in West Bengal continued to increase. Fur- thermore, the area and number of medium-size holdings in West Bengal continued to decrease, whereas those in Bangladesh in- creased. These trends seem to have been reversed in the early 1980s in Bangladesh. Table 3-7. Bangladesh: Size Distribution of Operational Holdings, 1960-61, 1967-68, and 1977 Percentage change, Percentage change, Average size 1960-61 1967-68 1977 1960-67 1967-77 (acres) Farm size (acres) N A N A N A Number Area Number Area 1960 1967 1977 <0.5 13.1 0.9 12.3 1.2 5.5 0.5 5 22 -60 -60 0.2 0.3 0.3 0.5-1.0 11.2 2.3 12.7 3.1 10.4 2.1 26 33 -25 -31 0.7 0.8 0.7 1.0-2.5 27.3 13.0 31.6 17.1 33.9 16.1 30 30 -3 -4 1.7 1.7 1.7 2.5-5.0 26.3 26.4 26.3 30.0 29.2 29.2 12 13 0 - 1 3.6 3.6 3.5 5.0-7.5 11.4 19.3 9.2 17.8 11.6 19.7 -9 -9 15 13 6.0 6.1 6.0 7.5-12.5 7.2 19.1 5.2 15.5 6.6 17.4 - 18 - 19 15 15 9.4 9.3 9.3 0 12.5-25.0 3.1 14.1 2.2 11.0 2.5 11.6 -21 -23 3 7 16.0 15.6 16.3 25.0+ 0.4 4.8 0.5 4.5 0.4 3.4 40 7 1 1 33.9 28.2 32.1 All 100.0 100.0 100.0 100.0 100.0 100.0 12 - 1 - 10 2 3.5 3.1 3.5 Number of opera- tional holdings (millions) 6.14 6.87 6.25 Total area owned (million acres) 21.7 21.5 21.96 N: Percentage of total number of operational holdings. A: Percentage of total area operated. Notes: Data are for farms and not households. Columns may not add to total because of rounding. Sources: Pakistan Ministry of Agriculture and Works (1960a), table 3; Bangladesh, Bureau of Statistics (1972), table 1, and (1977). The Agrarian Structure 69 Notwithstanding changes in the 1970s, the average size of opera- tional holdings continued to be much larger in Bangladesh than in West Bengal: by 37 percent (in 1960-61) and 84 percent (Bangladesh in 1977 compared with West Bengal in 1970-71). This runs contrary to the general view that Bangladesh has the smallest holdings in the subcontinent; although it experienced severe demographic pressures, the concentration of landholdings did not become more pronounced. West Bengal, however, had a much greater long-term decline in the average size of holding. It is one of the most densely populated and longest settled regions in the subcontinent, but it did not experience the social and economic dislocations that in the 1970s made Bangladesh's agrarian structure more inequitable. Pakistan The only data on the size distribution of operational holdings in Paki- stan come from the 1960 and 1972 agricultural censuses. These have been carefully examined by M. H. Khan (1981), some of whose main findings are reproduced in table 3-8 (see also Singh 1988a, tables A2.19 and A2.20). The data show that in 1972: - In sharp contrast to Bangladesh and most of India, only a third of all operational holdings were smaller than 5 acres. Together they accounted for 5 percent of the area in 1972, and their num- ber and area had declined rapidly nationwide since 1960. The de- cline was especially noticeable in the Punjab, where all other size groups increased both in number of holdings and in percentage of total cultivated area. * The number and combined area of farms larger than 50 acres increased in Pakistan as a whole, but not in Sind, where they fell slightly. Medium-size farms-12.5 to 50 acres-increased in number and combined area, most noticeably in the Punjab: this category also declined in Sind. • The average size of holdings fell in all size groups except the 5-12.5 acre group. Holdings of this size, small by Pakistan's standards, dramatically increased in number and area covered, especially in Sind.'4 The conclusion is clear: especially in the Punjab, the size distribution of operational holdings in Pakistan has become more skewed, favor- ing medium (12.5-50 acres) and large (more than 50 acres) farms at the expense of small ones. A comparison of the distribution of ownership (table 3-5) with that of operational holdings (table 3-8) shows that though the Punjab has 70 The Structure of Operational Holdings Table 3-8. Pakistan: Distribution of Operational Holdings and Area by Farm Size, 1960 and 1972 Farm size Pakistan Punjab Sind (acres) Year N A N A N A <5.0 1960 49.5 9.4 51.6 10.9 25.6 5.5 1972 28.2 5.2 26.1 4.8 19.0 4.5 5.1-12.5 1960 27.6 22.3 26.9 24.9 38.9 22.5 1972 39.9 25.2 39.0 24.6 51.7 34.5 12.5-25.0 1960 15.0 25.6 14.7 28.5 23.3 28.6 1972 21.1 26.6 23.1 28.8 21.9 29.2 25.1-50.0 1960 5.9 19.4 5.4 20.2 9.0 21.1 1972 7.6 18.8 8.8 21.3 5.2 13.2 50.1-150.0 1960 1.8 13.4 1.3 10.6 2.1 14.2 1972 2.7 15.1 2.7 14.7 1.7 10.7 150.1+ 1960 0.3 10.0 0.2 4.9 0.3 8.2 1972 0.4 9.1 0.3 5.8 0.4 7.0 Number of opera- tional holdings 1960 4.86 3.33 0.68 (millions) 1972 3.76 2.38 0.75 Total area operated 1960 48.93 29.21 9.71 (million acres) 1972 49.06 31.03 9.46 N: Percentage of total number of operational holdings. A: Percentage of total area operated. Source: Agricultural Census data in M. H. Khan (1981), table 3.17. the large farms, it is Sind that has the large landlords. In the Punjab, a large proportion of marginal and small owners (fewer than 12.5 acres) rent their land to owners of medium and large holdings. In Sind, by contrast, the owners of large and very large holdings rent land to small and sometimes medium-size sharecroppers. Medium and larger operational holdings predominate in the Punjab, and their operators are more likely to be owners who have rented more land to enlarge their holdings. In Sind, small operational holdings are mostly farmed by sharecroppers (haris), who rent from larger owners. East and West Punjab The Indian and Pakistani Punjabs share one basic similarity (see Singh 1988a, table A.2 1): a fall in the percentage of operational hold- ings of fewer than 5 acres, combined with an increase in their average size. In contrast, whereas West Punjab shows a rise in the share of medium-size holdings (12.5-50 acres) in the total number and area, East Punjab shows a decrease. As with ownership holdings, however, The Agrarian Structure 71 operational holdings are typically much larger in Pakistan than in India. In West Punjab nearly 21 percent of the operated area is in holdings between 25 and 50 acres, with another 15 percent in even larger holdings; in East Punjab, by contrast, nearly 80 percent of the operated area is in holdings of fewer than 25 acres. Thus the total area of operated holdings of fewer than 5 acres increased in Punjab- Haryana (India) but declined in West Punjab (Pakistan), and the num- ber and combined area of holdings larger than 25 acres rose substan- tially in the latter but fell quite dramatically in the former. Although both regions saw a dramatic shift toward medium-size holdings, this group consisted of farms of 20-50 acres in West Punjab, but of 5-15 acres in East Punjab. In addition, West Punjab had nearly 65,000 farms larger than 50 acres compared with 11,000 in East Punjab and Haryana. These differences in the size of holdings are an impor- tant factor in the patterns of technological change that have charac- terized the rapid growth in these two regions. These trends have been accentuated in the 1980s and have important implications for both the type of mechanization and rural employment in the two regions. Overstatement of Land Inequality The data on the concentration of landholdings in South Asia do not present a true picture of the extent of inequality in the ownership and use of land assets in rural areas. The unequal distributions of both owned and operational holdings are ameliorated by two facts. First, the distribution of holdings per household (or per farm) overstates the inequality because households with smaller holdings also tend to have fewer members. In Bangladesh, for example, households own- ing less than 1 acre of land had only 4.7 members on the average com- pared with 11.2 members per family for those owning more than 15 acres. This negative relationship between size of holding and number of household members who are dependent on that holding pertains throughout the subcontinent. As a consequence, the distribution of per capita holdings is much more equitable than the data on holdings per household suggest. Second, the size of holding is not even a close proxy for the income potential of the land held, because land quality and productivity vary significantly. Smaller holdings are generally farmed more intensely and have higher percentages of area under irrigation almost every- where in South Asia. The inequality in income potential between small and large holdings is therefore reduced within a given region. Furthermore, given the enormous diversity in land quality, irrigation, and cropping intensities in different regions, comparisons by size of 72 Landlessness holding across regions are inevitably misleading. More arid areas and those with poor land and low productivity (such as Sind and Raja- sthan) have larger holdings on the average than areas with assured irrigation (the Punjabs, Tamil Nadu) or adequate rainfall (West Bengal). Thus the raw data on the size distribution of holdings both within and across regions greatly overstate the real inequality in income po- tential between small and large holdings. A proper comparison would require adjustments by household size and land quality. Available data are inadequate to allow such adjustments to be made, however, at national or state levels. If adjustments were made to bring out the real differences in the net returns per hectare on farms of different sizes in different regions, the resulting distribution of the per capita income streams from different size holdings would be much more evenly distributed. Although these facts are widely recognized, un- critical use of the data on landholdings continues unabated. Landlessness Concern for the landless has been growing in recent years, as evi- denced by a rash of publications (ILO 1977a; Esman 1978). In fact, so much concern has been expressed that a certain cynicism has set in.'5 The main problem is that when the rural poor are identified with small farmers, many policies and programs currently being un- dertaken by both national governments and international agencies may be irrelevant or even counterproductive. In particular, it is widely believed that the landless constitute a majority of the poor; that their numbers and proportions are increasing dramatically; that they are being bypassed by programs that benefit mainly the landed; that the benefits of growth do not trickle down to them; that few if any programs have proved successful in increasing their incomes; and that apart from a radical program of land redistribution, little can be done to improve their prospects. Some observers go so far as to argue that growth has actually increased the poverty of the landless. At least three definitions for the landless in rural areas are tenable: those who own no land, those who operate no land, and those whose major source of income is wage employment. Each definition includes different but not mutually exclusive subsets of the rural population (typically these subsets overlap) and has different implications for the control over rural and other assets and for how incomes are derived. The three are often hopelessly confused. Data on landlessness per se are meaningless without adequate differentiation of these categories. But the data are not up to the task. Evidence from Indian data, which are the most complete, shows that landlessness in terms of those who own no land is large (about The Agrarian Structure 73 29 percent) and has declined over time; landlessness in terms of those who operate no land has increased in both relative and absolute terms; but landlessness in terms of those who neither own nor operate any land is very small (around 7 percent) and has declined in both relative and absolute terms (tables 3-9 and 3-10). Landlessness in terms of those not operating any land has increased significantly in many agriculturally dynamic states (Gujarat, Haryana, Punjab). There is no cause for alarm, however, because this increase is the result of a larger proportion of the rural population moving away from agricultural occupations and into a growing and dynamic nonfarm sector. In some agriculturally stagnant states (West Bengal and Bihar) this type of landlessness has actually declined! These data on landlessness suggest the ambiguity of the concept because they fail to distinguish between dynamic and stagnant processes that have the same outcome-fewer people owning or operating land! Much that has been written on landlessness in Bangladesh is replete with confusion on definitions and data sources. No consistent set of data exists to verify trends in landlessness, although some have been inferred from a large set of disparate and noncomparable sources. What little is firmly known suggests that (a) in terms of nonownership of land a third of all rural households are landless, (b) about 30 per- cent of all rural households fall in the category of "landless labor," and (c) this category has been growing in both relative and absolute terms at a pace that may have accelerated in recent years (table 3-1 1). Table 3-9. Rural India: Agrarian Profile and Landlessness (household number in millions) Owning land Not owning land Raw total Year Aumber Percent Number Percent Number Percent Operating land 1960-61 51.8 71.5 1.6 2.2 53.4 73.7 1970-71 54.7 69.8 2.2 2.8 56.9 72.6 1980-81a 64.4 67.0 3.3 3.5 67.7 70.5 Not operating land 1960-61 12.2 16.8 6.8 9.4 19.0 26.3 1970-71 16.1 20.5 5.4 6.9 21.5 27.4 1980-81a 23.6 24.6 4.8 4.9 28.4 29.5 Total 1960-61 64.0 88.3 8.5 11.7 72.5 100.0 1970-71 70.8 90.3 7.6 9.7 78.4 100.0 1980-81a 88.0 91.6 8.1 8.4 96.1 100.0 a. Estimated by extrapolating past trends. Sources: NSS 16th Round (1960-61); NSS 25th Round (1970-71). Table 3-10. India: Percentage of Rural Households Not Owning and Not Operating Land, Selected States, 1954-55, 1961-62, and 1971-72 Owning but not Neither owning Not owning land Not operating land operatinej nor operatinge Leasing in land 8th 17th 26th 8th 17th 26th 17th 26th 17th 26th 26th Round, Round, Round, Round,b Round, Round, Round, Round, Round, Round, Round, State 1954-55 1961-62 1971-72 1954-55 1961-62 1971-72 1961-62 1971-72 1961-62 1971-72 1971-72 Andhra Pradesh 30.12 6.84 6.95 42.80 37.95 36.05 32.03 29.68 5.92 6.37 83.58 4' Bihar 16.56 8.63 4.34 23.84 21.71 20.65 15.28 17.52 6.43 3.13 83.56 Gujarat n.a. 14.74 13.44 n.a. 25.41 33.75 11.78 25.47 13.63 8.28 55.98 Haryana - - 11.89 - - 48.00 - 41.05 - 6.94j 56 Punjab 36.86 12.33 7.14 38.92 39.09 58.61 30.51 52.90 8.58 5.71J .83 West Bengal 20.54 12.56 9.78 24.30 33.88 30.94 24.21 23.09 9.67 7.85 86.65 - Not applicable. n.a. Not available. a. Estimates are not available for NSS 8th Round. b. Percentage of nonagricultural holdings deemed comparable by S. K. Sanyal (1977b) to NSS 17th and 26th Rounds. Source: Sanyal (1977b). Table 3-11. Bangladesh: Landlessness Landless laborers as percentage of Landless laborers Source Year total cultivators (millions) Cited in A. R. Khan (1972) Agricultural Census 1951 14.3 1.51 Agricultural Census 1961 17.5 2.47 Master Survey of Agriculture 1963-64 17.8 2.71 Master Survey of Agriculture 1964-65 17.5 2.75 Master Survey of Agriculture 1967-68 19.8 3.4 Agricultural Agricultural Households laborers plus laborers with no sharecroppers as as percentage land other percentage of of total than rural households cultivators house lot Cited in Adnan and others (1978) Agricultural Census 1951 n.a. 15.3 n.a. Agricultural Census 1961 n.a. 17.5 n.a. IRDPa (14 villages) 1973-74 n.a. 38 BIDSb (8 villages) 1974 26 n.a. 38 Land Occupancy Survey (400 villages) 1977 n.a. n.a. 32 n.a. Not available. a. Integrated Rural Development Program. b. Bangladesh Institute of Development Studies. 76 Landlessness Table 3-12. Pakistan: Rural Households, Landownership, Tenancy, and Estimates of Landlessness, 1961 and 1972 Indicator 1961 1972 (1) Rural population (millions) 35.8 47.4 (2) Rural household size (persons) 5.5 5.8 (3) Rural households (millions) 6.5 8.2 (4) Number of landowners (million persons) 5.1 10.1 (5) Total area owned (million acres) 48.6 67.6 (6) Number of farm households (millions) 4.9 4.0 Percentage operating as Owner cultivators 41.1 41.7 Owner-tenant cultivators 17.2 23.8 Tenant cultivators 41.7 34.4 (7) Area operated by farm households (million acres) 48.9 49.1 Percentage operated by Owner cultivators 38.3 39.5 Owner-tenant cultivators 22.5 30.9 Tenant cultivators 39.2 29.6 (8) Percentage of farm households cultivating area other than that owned 58.9 58.2 (9) Percentage of area operated by farm households that is not owned by the operating household 50.5 46.2 (10) Estimated number of rural households not operating land (millions) 1.6 4.2 (11) Estimated landless agricultural labor households (millions) 0.6 1.6a a. Estimated using 1961 proportion: (11) / [(11) + (10)]. Sources: (1) World Bank data; (2) Afzal (1974), p. 47; (3) computed from (1) and (2); (4) through (9) M. H. Khan (1981); (10) by mere subtraction, (3) - (6); (11) M. H. Khan (1981) citing figure in 1961 census of approximately 2.0 million tenant "farmer- workers," confirmed by (6), whichjustified estimating 565,435 landless laborers belong- ing to separate households. Nothing can be inferred directly about landlessness-its magnitude or trends-in Pakistan because data are practically nonexistent. De- spite this serious lacuna in the agrarian data base, data can be "con- structed" from a number of sources. They show that the proportion of rural households not operating any land nearly doubled in the 1960s, as did landless agricultural labor households as a percentage of all rural households. More recent data from Pakistan are very diffi- cult to obtain and are not easily comparable with data from the 1970s. De Kruijk (1986b) used a variety of sources to construct a distribution of rural households by activity and status in 1979-80. He found that about two-thirds of all households were engaged in agricultural work; The Agrarian Structure 77 of these only 18 percent were landless, while the rest (45 percent) were either owners, owner-tenants, or pure tenants. The proportion of landless households (18 percent) is slightly lower than that esti- mated indirectly in table 3-12 for 1972. Trends in landlessness in and of themselves do not signify much. They need to be examined along with other evidence on the changing agrarian structure. One should concentrate directly on the conditions of cultivation for those who operate land-the size, productivity, and tenurial status of holdings-and on the conditions of wage employ- ment for those who depend primarily on wage labor in rural areas for their incomes. For this purpose, a closer look at the occupational distribution of rural labor households is more helpful. Rural Wage Labor The households that depend primarily on wage employment in rural areas-more specifically, in agriculture (agricultural labor house- holds)-include not only the so-called landless but also many mar- ginal and small farmers who operate owned land as well as small tenants who lease-in land and supplement their farm incomes with wage employment. Data on rural labor are sparse, except for India, but they all show that the proportion as well as the absolute number of rural house- holds that have to depend primarily on wage employment are increas- ing (tables 3-13 to 3-16). In 1985 a third of the rural population in Pakistan and Bangladesh and more than two-fifths in India were esti- mated to belong in the category of rural labor. A quarter of this rural labor force in India, a third in Pakistan, and nearly half in Bangladesh were also working on their own farms-that is, they were operating land as farmers or tenants. In India nearly two-fifths of the rural labor force are casual or agricultural laborers; in Bangladesh 25 per- cent and in Pakistan only 6 percent (figure 3-3). But not all are de- pendent on agriculture. A quarter of the rural labor force in Pakistan and about 13 percent in India work in nonagricultural occupations (table 3-13 and Singh 1988a). In India in 1974-75 more than half of all rural labor households (about 16 percent of all rural households) were landless-in the sense of not cultivating any land, whether owned or leased-and were de- pendent entirely on manual wage labor. Although the proportion re- mained fairly constant, the absolute number of such households increased (table 3-14). In Bangladesh nearly a third of all rural households are in this category, and to the extent that the available data allow temporal comparisons, proportions and absolute numbers seem to have declined (table 3-15). 78 Rural Wage Labor Figure 3-3. Composition of Rural Labor Force in South Asia Bangladesh, 1974 [ ...... ... India, 1972-73 Pakistan, 1974 0 25 50 75 100 Percent O Working on own farm J Working as casual wage earners or agricultural laborers 0 Working as helpers 0 Working in nonagricultural sector or seeking work Source: Table 3-13. Although the proportion of rural households not operating any land seems to be high and increasing in Pakistan, many of these households engage in nonfarm occupations, and the share of those mainly dependent on agricultural labor is much smaller (table 3-16). Several additional characteristics of the rural labor force in South Asia stand out clearly. Both small and marginal farmers depend heav- ily on wage labor and participate in rural labor markets, but agricul- ture is not the only source of their income since a large variety of nonfarm productive activities also hire wage labor. Much of the em- ployment, however, is casual rather than permanent (Parthasarathy 1987). Rural laborers face declining wage opportunities in many parts of India and Bangladesh, but not in Pakistan or areas of rapid growth in India. They are heavily indebted and their indebtedness may be increasing. Rural labor is not at all organized in the subcontinent, with the possible exception of Kerala (Singh 1988a). Table 3-13. South Asia: Composition of Rural Labor Force by Usual Activity Bangladesh India Pakistan /974 1972-73 1974 estimate Percent 1985 Percent 1985 Percent 1985 Percent of rural estimate of rural estimate of rural estimate of rural Item Millions labor (millions) Millions labor (millions) Millions labor (millions) Millions labor Rural population 65.0 86.0 455.0 573.0 50.0 69.0 728.0 Rural labor force 16.0 25a 26.5 200.0 44a 252.0 15.0 30a 20.7 299.0 42a Working on own farm 7.5 48 12.7 50.6 25 63.0 5.7 36 7.5 83.3 28 Working as casual wage earners or agricultural laborers 4.0 25 6.6 78.0 39 98.3 0.9 6 1.2 106.1 35 Working as helpers 4.3 27 7.2 45.6 23 58.0 4.6 30 6.2 71.4 24 Working in nonagricultural sector and/or seeking work __b 25.1 13 32.8 4.3 28 5.6 38.4 13 Notes: In Bangladesh only the agricultural labor force was considered. Those working on their own farms include part-owners, and pure sharecroppers are also included here. In India the available breakdown was for farm and nonfarm work combined. It was assumed that 74 percent of the labor was for agricultural activities; the reduced numbers are shown and the last category increased accordingly. For Pakistan the 1972 Agricultural Census gives higher estimates for the labor force than the Labor Force Survey used here. Estimates for 1985 assume there is no change in the labor structure from that in the 1970s. a. Percentage of total rural population. b. Included in other categories. Sources: Bangladesh, Ministry of Agricultuire and Forests (1978), p. 64: World Bank data: India, Planning Commission, "Draft Five-Year Plans, 1978-83" (New Delhi, 1978), vol. 2, table 1; Pakistan, Ministry of Finance, Planning and Economic Affairs, "Labor Force Survey, 1974-75," pp. xix, 67, and 81. Table 3-14. India: Changes in Rural Labor, 1964-65 and 1974-75 Rural labor households Agricultural labor households Rural With Without With Without households As percentage All land land As percentage All land land Region (millions) of rural households (millions) (millions) (millions) of rural households (millions) (millions) (millions) 1964-65 North 16.4 15.2 2.5 1.1 1.4 13.4 2.2 1.0 1.2 West 17.5 22.3 3.9 1.4 2.5 18.9 3.3 1.3 2.0 South 18.3 33.3 6.1 2.4 3.7 30.0 5.3 2.0 3.3 East 18.2 29.7 5.4 2.8 2.5 24.2 4.4 2.4 2.0 All India 70.4 25.4 17.9 7.7 10.1 26.6 15.2 6.7 8.5 1974-75 North 19.6 18.9 3.7 1.7 2.0 14.8 2.9 1.4 1.5 West 19.7 25.4 5.0 2.3 2.7 21.3 4.2 2.0 2.2 South 21.1 40.8 8.6 3.9 4.7 34.6 7.3 3.3 4.0 East 21.7 35.0 7.6 4.2 3.4 29.5 6.4 3.6 2.8 All India 82.1 30.3 24.9 12.1 12.8 25.3 20.8 10.3 10.5 Percentage increase North 20 48 55 43 32 40 25 West 13 28 64 8 27 54 10 South 15 41 63 27 38 65 21 East 19 41 50 36 45 50 40 All India 17 39 57 27 37 54 24 Note: Agricultural labor households draw over half their incomes as wages for agricultural work; they constitute 85 percent of the rural labor house- holds (which draw wages for labor in rural areas) and 22 percent of all rural households. Source: India, Ministry of Labour (1978). The Agrarian Structure 81 Table 3-15. Bangladesh: Rural Labor Households Percentage change, Item 1973-74 1977 1978 1973-74 to 1978 Rural population (millions) 67.6 76.0 78.0 15.4 Rural households (millions) 11.1 11.8 12.0 19.8 Households not owninga and not operating any land (millions) 3.7 3.3 3.1 -17.2 As a percentage of all rural households 33.3 27.9 23.3 Estimated rural labor households (millions) 4.4 4.7 5.3 20.4 As a percentage of all rural households 39.6 39.8 40.0 Estimated agricultural labor households (millions) 2.8 3.4 4.0 70.0 As a percentage of all rural households 25.2 28.8 29.2 Average household size (persons) 6. lb 6.4b 6.5b 1.5 a. Those who owned no land other than house lots. b. Estimated simply by dividing the rural population by number of rural house- holds. Sources: Government of Bangladesh and World Bank population estimates; Jannuzi and Peach (1977a); Bangladesh, Agricultural Census, 1960; Robinson (1969); Jabbar in Bangladesh, Ministry of Agriculture and Forests (1978), and World Bank data. Because these households often constitute a majority of the "poor- est of the poor," special concern must be given to increasing employ- ment opportunities for them in rural areas in the short run and in the economy as a whole in the long run. Most of the changes in the agrarian structure can be attributed to population pressures on land-the Malthusian imperative. But changing technologies and their interaction with rural institutions have also contributed to and in some cases accelerated these changes in the agrarian structure. Rural Unemployment Despite the serious limitations of the available data, it seems reason- able to conclude from the previous section that more and more peo- ple in rural South Asia are earning wages rather than cultivating land they hold as owners or tenants. During recent years, however, the growth of rural employment opportunities has failed to keep pace with the number seeking jobs. Estimates of the imbalance between 82 Rural Unemployment Table 3-16. Pakistan: Rural Laborers Percentage change, Iterm 1961 1972 1961-72 (1) Rural population (millions) 42.9 65.3 52.2 Punjab 25.6 37.8 47.7 Sind 8.5 14.2 67.1 (2) Rural household size (persons) 5.5 5.8 5.5 (3) Rural households (millions)a 7.8 11.3 44.9 (4) Farm households (millions) 4.9 4.0 -18.4 (5) Rural households not operating land (millions)b 2.9 7.3 150.7 As a percentage of all rural households 37.2 64.6 (6) Agricultural laborers (millions) 7.6 9.7 27.6 Punjab 4.8 5.8 20.8 Sind 1.6 2.4 50.0 As a percentage of all rural households 17.7 14.8 (7) Agricultural labor households (millions)c 1.4 1.7 19.5 a. Derived by dividing population by household size. b. Residually derived, (3)-(4). c. (6) 1 (2). Sources: M. H. Khan (1979b); Afzal (1974). work and workers in rural areas vary, depending on how employment and unemployment are defined and how the available quantitative es- timates are interpreted.'6 This section is concerned with the broad magnitude of the unemployment problem (and hence the scale of the efforts that might be needed to tackle it), the forms unemployment takes, and the characteristics of the unemployed. This information is needed to determine which measures might be appropriate to raise employment levels. As A. K. Sen (1975) has pointed out, more has been written on unem- ployment in India than in any other country. Nevertheless, the esti- mates of rural unemployment in India vary widely, as do the figures for Pakistan and Bangladesh-largely because of differences in con- cepts and in measurement and estimation procedures."7 A great deal of the confusion surrounding the figures can be avoided if some of these conceptual issues are clarified at the start. (Singh 1988a, annex 3, attempts to clarify some of the common concepts and definitions used in measuring rural unemployment. It should be referred to in interpreting the data that follow.) Given the extent of self- The Agrarian Structure 83 employment in rural areas (in farm or nonfarm enterprises), and the fact that much of the work consists of household tasks shared among members, what it means to be "employed" or "unemployed" or "seek- ing employment" must be interpreted with care. Most poor people- and those who rely mainly on rural wage employment fall in this category-try to engage in whatever gainful work they can find, often on a temporary or casual basis. The amount of work they do, the time they work, and the income they earn are all difficult to define and measure. India The available evidence from India shows that open, or chronic, un- employment is very low-about 1 percent if measured on the basis of the "usual activity" of the respondents in the rural labor force- which suggests that nearly everyone finds some work during the year. When measured on the basis of whether or not respondents were un- employed during the week or on the day of the survey, unemploy- ment is not exceptionally high (about 4-8 percent), but the numbers are large-about 13 million on any given day are without jobs. There are enormous regional differences, with most of the high unemploy- ment rates (measured in person days) concentrated in six states in India-Bihar and West Bengal in the east, Andhra Pradesh, Kerala, and Tamil Nadu in the south, and Maharashtra in the west (tables 3-17, 3-18, and 3-19). Unemployment is a highly seasonal phenomenon, as might be ex- pected, and the extent of unemployment in the slack period also var- ies by state. In general, states that have experienced high growth have both low unemployment rates and less variation between peak and slack seasons. Much of the unemployment is heavily concentrated in landless households, and casual laborers account for between half to three- quarters of the total person days of unemployment in rural areas in some states. Women are a particularly vulnerable group with unem- ployment rates two to three times higher than among men. There is general agreement that of the three types of employment-self-em- ployment, regular wage or salaried work, and casual work-it is ca- sual work that offers the most uncertainty. There is considerable evi- dence that unemployment is highest among casual workers in the rural labor force (Vaidyanathan 1986b; Visaria 1980). In India the proportion of casual laborers in the rural labor force has recently been increasing (Dantwala 1987b). Moreover, casual labor is becom- ing the norm for wage employment in rural areas. In practically all Table 3-17. India: Percentage Distribution of Rural Labor Force by Usual Activity, Current Activity, and Sex Activity in Activity on Usual activity, current week, current day, 1972-73 1977-78 1977-78 Status Male Female Total Male Female Total Male Female Total Self-employed + helpers in agriculture 35.0 19.4 54.4 34.5 14.7 49.2 35.5 13.4 48.9 Self-employed + helpers in nonagriculture 7.0 3.4 10.4 7.7 2.7 10.4 7.8 2.4 10.2 °° Regular salaried farm employee 4.2 0.7 4.9 2.9 0.4 3.3 3.0 0.4 3.4 Regular salaried nonfarm employee 3.5 0.7 4.2 4.6 0.7 5.3 4.9 0.7 5.6 All casual wage earners 14.1 11.1 25.2 17.2 10.3 27.5 15.3 8.5 23.8 Unemployed 0.7 0.1 0.8 2.7 1.3 4.0 5.4 2.7 8.1 Total percentage 64.8 35.2 100.0 69.9 30.1 100.0 71.9 28.1 100.0 Millions 128.7 71.0 199.7 116.3 50.2 166.5 113.4 44.2 157.6 Note: The 1972-73 data are based on a rural labor force that was defined to include those 5 years old or ol(der, whereas the 1977-78 sample included only those 15-19 years old. Source: NSS data in Sarvekshana (1977, 1979), the journal of the NSS. The Agrarian Structure 85 Table 3-18. India: States by Level and Variability of Unemployment Rates, 1972-73 Unemployment rate Group and states (percent of laborforce) Index of variationa High unemployment, high variation Andhra Pradesh 11.23 87 Jammu and Kashmir 8.58 595 Orissa 10.18 105 Tamil Nadu 10.45 134 West Bengal 10.66 109 High unemployment, low variation Bihar 10.02 31 Karnataka 8.55 24 Kerala 23.49 18 Maharashtra 9.43 46 Low unemployment, high variation Assam 1.89 352 Madhya Pradesh 3.41 128 Rajasthan 3.25 426 Low unemployment, low variation Gujarat 5.43 30 Haryana 2.93 52 Punjab 3.94 48 Uttar Pradesh 3.38 30 a. The index of variation = [(maximum - minimum) / minimum] x 100. Source: NSS 27th Round. states in India two-thirds to three-fourths of all wage laborers were in casual employment in 1982-83-a substantial increase over the preceding decade (Parthasarathy 1987). Bangladesh Data from Bangladesh are inadequate, too, but indications are that although open unemployment is low-about 10 percent-a large pro- portion of the population works long hours for small returns in un- certain, seasonal, and casual jobs. More than half of those who were self-employed or wage earners worked less than 250 days, with 10-15 percent working less than 150 days a year (see I. Ahmed 1978; M. H: Khan 1975; R. I. Rahman 1978; Singh 1988a; and World Bank 1987b). Pakistan As with Bangladesh, available data are less reliable than for India, but they show very low rates of rural unemployment-about 2 percent 86 Rural Unemployment Table 3-19. India: Unemployed as Percentage of Total Males 5 Years Old and Older, According to Activity on Current Day in Rural Areas State 1972-73 1983 Andhra Pradesh 6.90 5.59 Assam 1.46 1.88 Bihar 5.67 4.24 Gujarat 3.48 3.02 Haryana 2.25 3.41 Jammu and Kashmir 5.24 8.36 Karnataka 4.65 4.69 Kerala 12.84 13.39 Madhya Pradesh 2.19 1.56 Maharashtra 5.60 3.99 Orissa 6.31 5.09 Punjab 2.89 4.08 Rajasthan 3.25 3.09 Tamil Nadu 7.19 12.00 Uttar Pradesh 1.95 2.16 West Bengal 7.23 8.80 All India 4.75 4.79 Source: NSS data in Parthasarathy (1987), table IV. of the labor force-mostly among landless casual laborers. This is consistent with rising real wage rates and actual shortages of labor in recent years produced by migration to the Middle East from rural areas. (See Robinson and Abbasi 1979; Eckert 1972; Anderson and Leiserson 1978; and Singh 1988a.) Conclusions In aggregate terms, rates of unemployment in South Asia are low by traditional standards, but a large proportion of the population works in marginal activities with low productivity. Whether or not this is called "disguised unemployment" is less important than the fact that a high proportion of those gainfully employed are available for addi- tional work and meanwhile work long hours with little remuneration. The problem is one of inadequate growth in labor demand; its solu- tion lies in increasing the opportunities for productive employment of the wage labor force, a group that includes many cultivators who own or lease small holdings. The final section of this chapter briefly outlines past and current trends in the occupational distribution of these opportunities. The Agrarian Structure 87 The Occupational Distribution of Rural Wage Employment There are three possibilities for raising the incomes of poor rural households that depend primarily on wage earnings. First, on-farm employment in crop production may increase as a result of multiple cropping, increased output, and shifts to labor-intensive and higher- value crops such as vegetables and fruits. Second, noncrop agri- cultural activities (such as dairying, poultry farming, fisheries, and forestry) may offer new income-producing opportunities. Third, nonagricultural rural activities (for example, small-scale manufactur- ing, processing, transport, repair, marketing and other services, or rural public works programs) may provide additional jobs and wage earnings. The extent to which increased crop production raises the demand for wage labor depends on the size distribution of holdings, the crop- ping patterns adopted, and, most critically, farm technologies. In some areas of rapid agricultural growth (the Indian and Pakistani Punjabs, for example), much of the potential stimulus to employment has been offset by farm mechanization (see chapter 5). Greater use of irrigation and higher cropping intensities should on balance in- crease the demand for farm labor (although changing from persian wheel to tubewell irrigation may not), but how much of this demand will be met by family labor rather than hired workers is not clear. In general, crop production in the 1990s is unlikely to generate enough new demand for rural labor to absorb the two main components of potential supply-the currently unemployed and underemployed in the rural labor force, together with the anticipated increments to that labor force over the years. Consequently, the rapid expansion of noncrop agricultural and nonagricultural rural activities is crucial for the prospects of the poorest in rural areas."8 This section briefly out- lines the extent to which these activities are already contributing to the incomes of rural labor households; subsequent chapters examine the prospects for growth in these activities. Noncrop and Nonfarm Employment Noncrop and nonfarm sources of employment and incomes are usu- ally found in small-scale, highly dispersed, informal enterprises that are often overlooked in surveys and censuses. Data on the extent and nature of these enterprises can therefore be hard to come by. Ander- son and Leiserson (1980) evaluated the available evidence from some fifteen developing countries and found that nonfarm activities in rural areas are a primary source of employment and earnings for ap- 88 The Occupational Distribution of Rural Wage Employment proximately one-quarter of the rural labor force in most developing countries (one-third if the labor force in rural towns is included).19 They are a significant source of secondary earnings for small farmers and the landless in the slack season when farm and nonfarm employ- ment move in opposite directions. The total contribution of nonfarm activities to employment in rural areas is growing (see also Chuta and Liedholm 1979, p. 18); they include a wide variety of activities in man- ufacturing (20-30 percent), services (20-35 percent), commerce (15-30 percent), construction (5-15 percent), and transport and pro- cessing (5-15 percent).20 They are generally carried out on a very small scale, rely on simple labor-intensive methods, require little capi- tal and skills (other than those provided indigenously via apprentice- ship), and are widely dispersed spatially (at least until they become concentrated in rural towns as infrastructure improves and markets grow). Most of the demand for these products and services is gener- ated locally in rural areas and is linked closely to the growth of agri- cultural and rural incomes. India In India 39 percent of nonfarm employment is estimated to be in manufacturing, 14 percent in construction, 14 percent in trade and commerce, 24 percent in services, and 9 percent in transport and "other" categories (Anderson and Leiserson 1980, p. 245). The lim- ited available evidence shows that noncrop incomes account for nearly half of the incomes of all rural households. The proportions are smaller but still substantial for cultivating households and depend on whether high-yielding varieties are planted. As table 3-20 shows, the nonagricultural sector accounts for about 15 percent of the rural labor force in India and Bangladesh and about 30 percent in Paki- stan.21 Table 3-21 presents data on rural per capita incomes in India, com- piled from an extensive series of surveys of 4,100 rural households by the National Council of Applied Economic Research (NCAER) be- tween 1969 and 1971. The figures show that the proportion of in- come derived from crops rises with the size of holding, and that non- agricultural wages and nonfarm earnings provide nearly half of the incomes of landless and marginal farmers (those with holdings of less than I hectare) and between a fifth and a quarter of the incomes of small farmers (those with holdings of up to 5 hectares).22 All these groups have very low per capita incomes; only their noncrop and nonfarm earnings permit the very poorest to match the near- destitution incomes of the small farmers. The Agrarian Structure 89 Table 3-20. South Asia: Rural Labor Force in the Agricultural and Nonagricultural Sectors Rural labor force (millions) Nonagricultural In agricultural In nonagricul- sector as per- Country Year Total sector tural sector centage of total Bangladesh 1951 96.8 82.8 14.0 14 1961 96.6 84.7 11.9 14 India 1961 86.0 70.5 15.5 18 1971 82.5 70.3 12.2 15 Pakistan 1951 86.8 63.1 23.7 27 1961 84.5 58.3 26.2 31 Source: Country census data. The sources of income for the landless are of special interest. It is often thought that they depend heavily on agricultural wage employ- ment for their livelihood. For this reason, much work has gone into estimating on-farm employment elasticities (see Singh 1988a and chapter 7) and into efforts to analyze how these elasticities are trans- lated into jobs and income for the landless. But the NCAER data reveal that only a quarter of the incomes of the landless came from agricul- tural wages. The rest came from noncrop and nonfarm activities, the largest proportion being in construction, manufacturing, rural ser- vices, and trades. Vyas and Mathai (1978) have noted that comparisons of India's rural occupational structure over time show hardly any change in the proportion of the labor force working in agriculture; the figures for 1957, 1961, and 1971 were 72.1, 71.2, and 72.1 percent respectively. There has thus been little occupational diversification in the rural economy. In fact, the proportion of the rural work force employed in industrial activities actually fell (from 18 to 16 percent between 1961 and 1971). Vyas and Mathai attribute these findings to the weak- ness of the linkages between agricultural and nonagricultural activi- ties in the rural economy (see chapter 7 for a discussion of these link- ages and nonfarm activities generally). Nevertheless, nonagricultural activities are important sources of rural employment in India, accounting for 15-20 percent of the labor force in nearly all states (see Singh 1988a, table A.44). If noncrop ag- riculture (livestock and dairying, poultry, forestry, and fishing) is in- cluded, this proportion rises to 20-25 percent. (Singh 1988a, table A.45, provides data from selected Indian states on the sources of rural per capita income by size of holding.) Even if noncrop income represents a small proportion of the income of the total rural labor Table 3-21. India: Rural Per Capita Incomes by Farm Size and Source of Income, 1970-71 Percentage of income derived from Size of Estimated Annual per Ancillary Agricultural Nonagricultural Nonfarm and cultivated land percentage capita income Crops activities wages wages other sources (acres) of households (rupees) (A) (B) (C) (D) (E) Landlessa 28 540 - 26 26 5 43 0-1.0 9 544 33 7 16 3 41 1.0-2.5 10 457 50 10 14 2 24 2.5-5.0 15 566 67 7 7 1 18 5.0-7.5 9 672 78 5 3 1 13 7.5-10.0 8 822 82 4 2 1 11 10.0-15.0 8 905 86 4 1 ... 9 15.0-25.0 8 1,040 92 2 1 ... 5 25.0 + 4 1,409 98 ... ... 0 1 All size groups 100 677 59 9 9 2 21 (Number of households in sample survey = 4,100) - Not applicable. ... Negligible. Notes: Columns (A) and (B) are gross values of output from crops and other farm activities less operating expenses. Column (C) includes cash and wages in kind, as does column (D), where workers are laborers and not salaried. Column (E) includes salaries, remittances, property rental income, and incomes from employment and enterprises in rural areas. Ancillary activities include livestock, poultry, dairying, forestry, and on-farm processing, while nonfarm sources include incomes from rural manufacture, handicrafts, repair and construction, transport, trade, and commerce. a. All noncultivating households. Source: NCAER, Assets and Rural Indebtedness Survey. The Agrarian Structure 91 force, it is a significant source of earnings for small farmers and the landless throughout India-in rice and wheat zones, arid and irri- gated areas, and high and low growth states." Data from the National Sample Survey (NSs) 25th Round, 1970-71, showed that a significant proportion of poor households (35 percent of small cultivators and 43 percent of noncultivating wage earners) supplemented their incomes in noncrop agricultural activities (India, Ministry of Planning 1976a; Visaria and Visaria 1973). Evidence from farm management surveys in India indicates that off-farm employ- ment accounted for 50 percent of small farmers' labor and somewhat less of their incomes in the 1950s (Bharadwaj 1974b). By the early 1970s, noncrop activities had assumed a greater significance; many small farmers earned much more from occupations such as dairying than they did from work associated with food crop cultivation. In Tamil Nadu, for instance, male agricultural laborers' average annual earnings from nonagricultural occupations increased from 10 per- cent to 38 percent of their total incomes in the 1960s and 1970s (ILO 1977b). Microeconomic studies of small farmers and agricultural laborers carried out by Agro-Economic Research Centers confirm this trend; they also suggest that members of agricultural labor households who engage in noncrop activities have marginally higher incomes than those who cultivate very small holdings (less than half a hectare), whether owned or tenanted.24 Pakistan and Bangladesh The data for Pakistan and Bangladesh are sparse and two decades old, but they confirm the importance of nonfarm sources of income in rural areas, especially for those with small holdings (tables 3-22 and 3-23). More recent evidence suggests that a rising share of the rural labor force is engaged in nonfarm work and that an increasing share of rural income comes from nonfarm sources. In 1970-71, 19 percent of rural employment in the Pakistani Punjab was in nonfarm activities; 48 percent of farm families in Pakistan relied on income from nonfarm activities, which contributed 23 percent of their total income. Nearly 70 percent of small farmers and the landless under- took nonfarm work, which accounted for nearly 39 percent of their income (Anderson and Leiserson 1980). In the Pakistani Punjab, Haider (1977b) has documented the rapid shift into nonfarm employment produced by the Green Revolution. New jobs have emerged in distribution, repair and maintenance ser- vices (especially for farm machinery), rice milling, small engineering 92 The Occupational Distribution of Rural Wage Employment Table 3-22. Bangladesh: Farm and Other Income by Size of Holding, 1965-66 Average size of Annual household incomes (rupees) Size of holding in Percentage Nonfarm holding size group of households Other as a percentage (acres) (acres) in the sample Farm sources of total 0-3 1.82 46 584 362 38 4-10 4.80 48 858 163 16 10+ 13.51 6 2,307 1,220 34 Note: The main source of other income of small cultivators is wage labor on and off the farm, while for the medium cultivators it is petty trade and in some cases remit- tances from family members working elsewhere. The bulk of the nonfarm income of large cultivators is from trade, nonfarm work, and additional remittances. Source: Rajshahi University survey data in T. M. Khan and Bose (1968). Table 3-23. Pakistan: Contribution of Nonfarm Income to Total Income of Farm Families by Size of Holding, 1968 Size of holding Nonfarm as a percentage (acres) of income 0-6.25 39 6.25-12.49 22 12.5-18.74 22 18.75-24.99 8 25+ 6 All size groups 23 Source: Steele (1975). workshops, transport, construction, marketing agencies, and financial institutions. Meanwhile, however, some of the traditional activities of village artisans-blacksmiths, carpenters, weavers, potters, cobblers, traders, millers-have declined in importance or have taken different forms on farms. Many artisans stay in villages but commute to work elsewhere. Another feature of the rural economy in Pakistan has been exten- sive migration to cities and the Middle East. Most of the migrants are from landowning groups, but artisans represent a significant percent- age of the total.25 Small towns have grown rapidly as new nonagricul- tural occupations have flourished and attracted rural migrants. These migrants have become a major source of remittances to the rural economy, thus raising both rural incomes and the demand for nonfarm work. Farm labor is becoming scarce as higher returns in nonagricultural occupations raise rural wage rates.26 With the exten- The Agrarian Structure 93 sive development of rural infrastructure, new roads and electrifica- tion have made it easier to set up small engineering shops and manu- facturing facilities. D. A. Khan (1978) estimates that between 1960 and 1972 incremental employment in nonfarm occupations exceeded that in on-farm employment (which had been considerable owing to the Green Revolution).27 In Bangladesh, both the 1974 census and the 1973-74 Household Survey showed that nearly 30 percent of all rural households could be classified in nonagricultural occupations. Most of this activity con- sisted of rural trade and services, organized around small rural towns that serve as marketing centers. Small-scale and cottage industries in rural areas accounted for less than 4 percent of the total labor force in Bangladesh. A survey by Ali (1980) documented the extent to which farmers in Bangladesh were taking up nonfarm employment to obtain additional income. Interviews with more than 2,140 farm families in 1980 showed that 84 percent of all farm operators reported off-farm em- ployment of more than 100 days a year, but that those with holdings of more than 5 acres did not work off the farm. The percentage re- porting off-farm work of 100 days or more had risen steadily, from 8.4 percent in 1951 to 30 percent in 1960 to 51 percent in 1970 to 84 percent in 1980-a tenfold increase in three decades! Only 10 per- cent of those engaged in part-time farming produced enough on their land to meet their yearly food needs. More than 53 percent worked on their holdings for more than nine months (270 days), another 23 percent did so for six to nine months. Conclusions This chapter has dealt with agrarian structures in South Asia. On the basis of the limited evidence available, I have shown that large num- bers of those living in rural areas in South Asia are unable to support themselves by farming because their landholdings are very small or nonexistent. The employment opportunities for these people who make up the rural wage labor force fall short of the demand for jobs, and those who do find work as farm laborers seem increasingly to suf- fer from underemployment and the erosion of the real value of their earnings. Earnings from nonfarm agricultural activities and from nonagricultural rural jobs are of special importance for marginal farmers and the landless. Any dramatic expansion of these income sources in recent years has been confined, however, to areas in which agricultural growth has been high. This link between agricultural growth and the growth of nonfarm employment opportunities is criti- Table 3-24. South Asia: Estimates of Small-Farm and Landless Households and Population, 1980 (households amid population in millions) Bangladesh India Pakistan South Asia House- Family Popu- House- Family Popu- House- Family Popu- House- Popu- Item holds size lation holds size lation holds size lation holds lation Landless rural labor Noncultivators with wage income more than half of total 3.0 4.6 13.8 15.0 4.4 66.0 2.0 4.5 9.0 20.0 88.8 Percentage of all rural 21.0 - 16.7 16.0 - 12.6 19.8 - 15.4 17.0 13.3 Near landless Cultivating fewer than 0.4 hectares 1.1 5.1 5.6 14.0 5.0 70.0 0.2 4.9 1.0 15.3 76.6 Percentage of all rural 7.7 - 6.8 15.0 - 13.3 1.9 - 1.7 13.0 11.5 Marginal farmers Cultivating 0.4-1.0 hectares 2.3 6.2 14.3 16.9 5.3 89.6 0.5 5.0 2.6 19.7 106.5 Percentage of all rural 16.1 - 17.2 18.1 - 17.1 5.0 - 4.4 17.0 16.0 Small farmers Cultivating 1.0-2.0 hectares 2.0 7.1 14.2 15.9 6.1 97.0 0.7 5.3 3.8 18.6 119.0 Percentage of all rural 14.0 - 17.1 17.0 - 18.5 6.9 - 6.5 15.8 17.3 Subtotal 8.4 - 47.9 61.8 - 122.6 3.4 - 16.4 73.6 386.9 Percentage of all rural 58.8 - 57.8 66.1 - 61.5 33.6 - 28.0 62.8 58.1 All rural households 14.3 5.8 82.9 93.6 5.6 524.2 10.1 5.8 58.6 118.0 665.7 - Not applicable. Source: Based on 1970-80 changes in the population and on 1970 percentages. The Agrarian Structure 95 cal to any evaluation of employment prospects for the rural poor in the future. (For some final numbers on the agrarian structures in South Asia see table 3-24 and figure 3-4). Although it is widely known that unequal access to land-owned or operated-is a major source of income inequalities in rural areas of South Asia, what is less well understood is the role of wage employ- ment, nonfarm enterprises, and ancillary activities in mitigating these inequalities. In a recent study of income inequalities in rural Pakistan, Ercelawn (1984) has shown this role to be significant. Poverty is borne primarily by those with little or no access to land; one solution is to redistribute land. But as I shall show later this is not the only solution to alleviating poverty. Another is to increase sources of nonfarm em- ployment through agricultural growth. In chapters 2 and 3, I have tried to sketch the economic environ- ment in which small farmers and the landless in South Asia seek their livelihood. I have discussed the dimensions of rural poverty in gen- eral terms and have outlined some of the characteristics of the rural poor. In particular, I have examined two factors which both define these people and determine their economic status-their access to Figure 3-4. Small-Farm and Landless Households as a Percentage of Rural Households in South Asia, 1980 100 75 66.1% 5 8. 8 % /:...::: 50 ....::.........:.,,'.. - Small farmers Marginal farmers Near landless ~ S~i.~e Landless rural labor Bangladesh India Pakistan Note: The household categories are defined in Table 3-24. Source: Table 3-24. 96 Notes land and to jobs. Clearly, any coherent policy to alleviate the condi- tions outlined in chapter 2 must base itself on new approaches to the problems summarized here-the small landholdings that characterize most rural South Asian households and the lack of adequate income- earning opportunities for those who lack even the minimal security of an economically viable plot of land. Chapters 4 and 5 focus on the prospects and problems of raising the productivity of small farmers in the crucially important food- grain sector and look at small farmers' access to credit and markets. Chapters 6 and 7 turn to the earnings potential of non-food-grain ag- riculture (especially livestock and dairying) and of nonagricultural rural activities. These possibilities for loosening the stranglehold of rural poverty will be of interest to academic researchers and practical policymakers alike. Despite the difficulties, the prospects of the rural poor need not be wholly bleak, as I hope to show in what follows. Notes 1. See Singh (1988a) for a more detailed analysis of the issues outlined in this chapter. 2. The Marxian perspective seems to dominate the literature on rural South Asia. Among others see Alamgir (1978); A. R. Khan (in ILO 1977a); Rudra (1978a and b); Alavi in Stevens, Alavi, and Bertocci (1976); Frankel (1971); Patnaik (1987); and Sau (1973). For the Malthusian perspective, see Vyas (1979); and Stokes (1978). 3. Attwood (1979) defines a family's upward mobility as an expansion in its landholding and downward mobility as a contraction of that holding. 4. Between 1961 and 1971 some 38,000 formerly landless households in two relatively poor regions in Haryana purchased land. 5. The fact that an overwhelming number of small farmers have been able to retain their holdings does not mean that they may not have become poorer. As Vyas (1979) has pointed out, further impoverishment can occur with or without proletarianization. As family size increases, previous levels of produc- tion may become more and more inadequate, forcing households to rely more and more on agricultural labor and other supplementary occupations. Under these circumstances, and in the absence of productivity increases or easy access to off-farm employment, per capita incomes will fall. 6. Sheila Bhalla (1977), p. A-4. "Only 5.3 percent of today's landless agri- cultural labor households cultivated any land at all in 1962. All of them were pure tenants" (p. A-12). 7. Nearly 75 percent of those who "lost" land had in fact transferred it to other family members (Sheila Bhalla 1977). 8. See Rudra, Majid, and Talib's (1969) study of 261 farms of more than 20 acres in 1968-69. It is hard to reconcile this with the National Sample Sur- vey (Nss) data, but since Rudra, Majid, and Talib collected their data from The Agrarian Structure 97 block development officers at the local level, one tends to give their findings greater credence. The official NSS data reveal the de jure structure; it is doubtful whether the de facto nature of landownership will ever be known. 9. The Indian data are for both East Punjab and Haryana and have been adjusted to exclude the category of households owning no land (a category not given in the Pakistan data). The figures are still not strictly comparable because the Pakistan data are for individual owners while the Indian data are for owning households. 10. Although there must be large holdings hidden under bogus transfers in both regions, the extent of this concealment is unlikely to differ between them. 11. Small owners who rent out their land do not necessarily join the ranks of the landless or agricultural wage earners. The category "nonoperating owners" covers people engaged in occupations other than agriculture, includ- ing workers in cities. 12. These broad conclusions are supported by the analyses by Raj (1976) and H. Singh (1976) and by the analysis of state data by Sirohi, Ram, and Singh (1976), except that the latter find a dramatic decline in the average size of marginal holdings between 1960-61 and 1970-71, even in Punjab. 13. According to the Land Occupancy Survey, 72 percent of all house- holds that owned land were in the 0.1-2 acre category and accounted for 24 percent of the owned land in 1977; according to the 1976 Pilot Census, 48 percent of all cultivators were in the 0.1-2.5 acre category and accounted for only 18 percent of the area cultivated. 14. The five-acre cutoff may represent a subsistence holding below which farmers in Pakistan cannot make an adequate living. Those with holdings smaller than this may rent them out or sell them to take up other occupations. 15. "The very recent fad about and display of interest in the landless is less due to a charitable concern on the part of the established officialdom and academia than due to a very real bout of enlightened self-interest arising out of the threat of disintegration to the established order in the face of the grow- ing trend of landlessness and agricultural stagnation" (Adnan, Khan, and others 1978, p. 19). 16. The conceptual and measurement problems in estimating unemploy- ment are discussed in some detail in Krishna (1976); A. Sen (1975); and Raj (1976). 17. The Indian debate on these issues can be found in Krishna (1973a and b, 1976); Dantwala (1973); Visaria (1972); and Raj (1976). There are only a few studies available for the other two countries: for Bangladesh, see Ahmed (1978); Clay and Khan (1977); and Clay in Bangladesh, Ministry of Agriculture and Forests (1978); and for Pakistan, see Eckert and others (1973). 18. Migration represents another option for the rural poor, but its poten- tial as a solution to the oversupply of labor in rural areas is strictly limited- notably by constraints on the mobility of potential migrants and on the ab- sorptive capacity of receiving areas and employers. Rural-urban migration is a major subject of study in its own right and is 98 Notes not discussed comprehensively in this book for reasons of space. There have been a number of useful studies on the dynamics of migration in South Asia; the interested reader might begin by examining Connell and others (1976). 19. It is estimated that 20-24 percent of the rural labor force in India is employed in nonfarm activities, around 50 percent in Taiwan, 30-40 per- cent in the Philippines, 20-25 percent in the Republic of Korea, 30-40 per- cent in West Malaysia, and around 30 percent in Indonesia. Nonagricultural rural activities include manufacturing, repair, processing, construction, handicrafts and artisanal work, transport, trade and commerce-many of which are, of course, ancillary to agricultural activities and needs. 20. Manufacturing includes a wide variety of activities in food processing (including oil pressing, milling, and husking), handicrafts, leather work, tex- tiles, wood products (including sawmilling, furniture, and carpentry), and metalworking (including blacksmithing, welding, fabrication and assembly work for machine tools and equipment); in Pakistan, East Punjab, and Tamil Nadu it includes small-scale manufacture and repairs of diesel and electric tubewells. Construction includes that of dwellings, farm buildings, roads, and civil works. In commerce, retail trade accounts for three-quarters of total em- ployment, the remainder being made up of marketing and financial services; other services include petty trading, small business, and repair and artisanal work. 21. This proportion has declined by 3 percentage points in India and in- creased by 4 points in Pakistan in the intercensal decade, but this could be due to differences in what is included in the nonagricultural category. 22. These data partly overstate the importance of nonfarm incomes since they include pensions, remittances, and rental incomes. Rent is, of course, an unlikely source of income for the landless or those with small holdings if they are also cultivating land, but it could be a significant part of nonfarm income for those with larger holdings. Remittances, however, could be a sig- nificant source of income for the poorest groups; the data needed to establish their importance are unfortunately not available. 23. Biases resulting from the inclusion of remittances, pensions, and other transfer incomes in nonfarm incomes should be kept in mind, however. 24. See AERC (1974; 1975b and c; 1976a, b, c, and d), for example. But agricultural laborers are also worse off in some states; see Chauhan and oth- ers (1973) and Kaul and Kahlon (1971). 25. Over half the annual migrants from rural Punjab are derived from land-owning households of small or large farmers. An additional 10 percent come from farm tenant households. Fifteen percent are classified as un- skilled laborers, while 17.5 percent bear with them the diverse skills of their respective trades (Eckert and Khan 1977). 26. Even village artisans have incomes that exceed those of permanent hired laborers by 20 percent to as much as 115 percent for such well-paid tradesmen as blacksmiths, barbers, and weavers (see Eckert 1972; D. A. Khan 1978). Between 1964-65 and 1976-77, in-kind and wage payments for farm work by permanent laborers rose by 149 percent in nominal terms; nonfarm occupations matched these increases. The Agrarian Structure 99 27. D. A. Khan (1978, pp. 935-36) estimates that on-farm employment on small farms increased by some 0.2 million work years and on medium and large farms by 0.99 million work years. He reports that nonfarm employment increased by 1.5 million work years. These estimates are, however, based on the use of simple fixed input-output ratios of labor coefficients to crop aver- ages; indirect employment is assumed to increase by 1.23 work years for every work year of farm employment-an estimate based on an Indian study. 4 Prerequisites for Sustai'ned Gains in Output on Small Farms Despite the unequal distribution of rural land in the subcontinent, large numbers of households obtain their livelihood from farming on small but viable holdings. Raising the productivity of these small-farm households must represent a central element in any strategy to reduce rural poverty in South Asia. In the subsequent chapters I will identify some of the opportunities that exist for implementing such a strategy. This chapter examines the characteristics of the small-farm sector, summarizes the available evidence on farm size as a factor in farm productivity and efficiency, and reviews the prospects for raising the output of traditional food grains through improved cultivation prac- tices on small farms.' The Small-Farm Sector Misconceptions about small farmers abound. It is sometimes sug- gested, for example, that they are inefficient and relatively unproduc- tive, grow only food crops for subsistence and do not produce for the market, do not use capital or credit extensively or effectively, and are inherently less willing to take risks than other farmers. These gener- alizations are often untrue and misleading. The evidence from South Asia shows that small farmers are by no means pure subsistence farmers-if subsistence farmers are defined as those who grow only food, mainly for home consumption; rely very little on markets for the sale of their output; and use only owned fac- tors of production, of which family labor is the most important (Wharton 1963; Nakajima 1970). In fact, although small farmers do indeed devote much of their cultivated land to food crops, they also plant a significant proportion of it with cash and other crops; and al- though they consume much of their food output on the farm, they 100 Prerequisites for Sustained Gains in Output on Small Farms 101 also market a significant share of it locally (even though they may later have to buy some of it back). Small farm operations are further characterized by higher irriga- tion and cropping intensities than on large farms, and as a result they use more inputs, especially (family) labor, per hectare and have higher total productivity per hectare. The last-the inverse relation- ship between farm size and productivity per hectare-is well estab- lished in traditional agriculture. There is some evidence, however, that in wheat-producing areas where the Green Revolution has had a significant impact, this relationship is being reversed (see Singh 1988b). The increased intensity in the use of cash inputs on larger farms seems to have neutralized the advantage previously held by small farms in terms of productivity. This is due not to lower effi- ciency on small farms, but to the better access of large farmers to these yield-increasing inputs. Imperfections in factor markets are the main sources of productiv- ity differences among farms, and there is nothing inherently superior or inferior about small family farms. Because these imperfections along with the economies of scale introduced by mechanization have favored larger farms, they appear to be more productive in this era of high-yielding varieties (HYv). With proper access to the relevant in- puts, some differences in productivity and efficiency may disappear, but real economies of scale will persist because of the indivisibilities of capital. Tables 4-1, 4-2, and 4-3 summarize the evidence sur- rounding the productivity-efficiency debate reviewed in Singh (1988b). The presumed higher productivity of small farms was one of the arguments in favor of redistribution of landholdings. The evidence suggests that redistributive land reform cannot be predicated solely on the assumed "productive superiority" of small farms. Other con- siderations of equity, employment, and social issues provide a much firmer basis for such a policy. Two other characteristics set small farmers apart. They are not more risk averse but are more vulnerable to the consequences of risk than are larger, better endowed farmers, and they represent farm- household units in which interdependent decisions are made.that af- fect both the farm and the household. Both these characteristics re- duce the number of choices available to small farmers and subdue their response to economic incentives. In evaluating prospects for increasing output and productivity, one must bear in mind that small farmers are not inefficient, tradition- bound, unresponsive to economic opportunities, or unwilling to adopt new ideas and technologies. But they do have limited means (Text continues on page 108.) Table 4-1. South Asia: Relationship between Farm Size and Productivity, the Pre-HYV Period Data Inverse Years Study Region Data source relationship Remarks 1954-59 A. K. Sen (1962) All India Various districts FMS Holds By and large, productivity per acre decreases with the size of holding. 1955-56 D. Mazumdar North India Two villages in Uttar FMS Holds Small farms may not be (1965) Pradesh viable. 1955-56 C. H. H. Rao West India Three districts, Maha- FMs/Au- Holds Size, not wage differen- (1966) rashtra thor tials, is important in ex- plaining relationship. 1954-57 A. P. Rao (1967) North India Disaggregated data from FMS May not hold Productivity, input use, ir- 3 villages in Punjab and rigation per acre do Uttar Pradesh (249 obser- not vary by size. vations) A. Rudra (1968) North India Disaggregated data: 17 dis- AERC Does not hold In correlating gross area tricts, 20 villages in Pun- with output, does not jab, Haryana, and Uttar hold; in correlating Pradesh net area with output, holds for all crops, not iindividual crops. 1959-60 S. K. Sanyal All India State-level data, all states NSS Probably holds Large farms have lower (1969) land use, fewer at- tached workers, less in- vestable surplus, and lower use of inputs per acre. 1957-64 U. Rani (1971) West, South, Farm-level data from NSS May not hold Statistically weak basis for and East Madhya Pradesh, Kerala, inverse relationship. India Andhra Pradesh, and Oris- sa (1,431 observations) 1954-68 G. R. Saini (1971) All India Disaggregated data from FMS Holds Statistically significant in- 9 states, 11 districts, verse relationship. and 25 cases 1955-57 N. Bhattacharya North India Disaggregated farm-level FMS May not hold Intravillage differences; and G. R. Saini data from Uttar Pradesh relationship holds for (1972) and Punjab Muzaffarnagar, Uttar Pradesh, and not for Ferozepur, Punjab. 2 1954-57 K. Bharadwaj All India Aggregate data from 6 FMS Appears to Not always statistically (1974a) states, 9 districts, 25 cases hold significant. 1956-71 G. K. Chaddha North India Twelve districts in Punjab FMs/Agri- Holds In all districts. (1978) cultural census 1955-57 A. K. Ghose North and Farm-level data from FMS Holds Relationship exists inde- 1968-73 (1979a) East India Hooghly, West Bengal, pendent of production and Ferozepur, Punjab relationship. 1962-70 V. V. Rao and T. South India Pooled data from Kerala, FMS Does not hold Especially with higher Chotigeat (1981) Tamil Nadu, and Andhra capital and cropping Pradesh intensities. Note: AERC = Agro-Economic Research Center; FMS = Farm Management Surveys; NSS = National Sample Survey (India). Table 4-2. Relationship between Farm Size and Efficiency: Evidence from South Asia Year of Study data Region and data description Remarks Pre-HYV Period T W. Schultz (1964) Traditional farming is efficient. M. Paglin (1965) - Large farmers are less efficient. W. D. Hopper (1965) 1954 Data for 43 farms in Uttar Pradesh Traditional farming is efficient. A. M. Khusro (1964) 1954-57 FMS data from Andhra Pradesh, Bombay, Efficiency does not decrease by farm size. Madhya Pradesh, Madras, Punjab, Uttar Pradesh, West Bengal G. S. Sahota (1968) 1955-57 FMS data from Madras, Uttar Pradesh, West No significant differences in efficiency Bengal across farm sizes. G. R. Saini (1969) 1955-57 FMS data from Punjab and Uttar Pradesh Farmers are efficient but no efficiency dif- ferences across size of farms. P. A. Yotopoulos, L. J. Lau, and 1955-57 FMS data from Madhya Pradesh, Madras, Small farmers are technically more effi- K. Somel (1970) Punjab, Uttar Pradesh, West Bengal cient, but there are no differences in price efficiency by size. L. J. Lau and P. A. Yotopoulos 1955-57 FMS data from Madhya Pradesh, Madras, Small farmers exhibit higher level of both (1971) Punjab, Uttar Pradesh, West Bengal price and technical efficiency. P. A. Yotopoulos and L. J. Lau 1955-65 FMs data from Madhya Pradesh, Madras, Small farmers are technically more effi- (1973) Punjab, Uttar Pradesh, West Bengal cient, but not different in their price effi- ciency. Post-HYV Period S. S. Sidhu (1974b) 1967-71 FMS, old and new varieties of wheat, 150 No difference in economic, technical, and farms, Punjab price efficiency by farm size. P. N. Junankar (1978a) 1969-70 FMS data, local and HYv paddy, 150 farms, Relatively equal economic, technical, and Tamil Nadu price efficiency of all farm sizes. F. S. Bagi (1979a and b, 1981) 1969-70 Farm-level data for irrigated and Small farmers are more efficient but not unirrigated farms, Haryana significantly different. " M. H. Khan and D. R. Maki 1974 Wheat and rice farms, Pakistan No differences in efficiency by farm size. (1979, 1980) K. Kalirajan (1981) 1977-78 Seventy farmers growing HYV rice in rabi Small and large farmers exhibit equal price season, Tamil Nadu and technical efficiency. - Not applicable. Note: FMS = Farm Management Surveys. Table 4-3. South Asia: Relationship between Farm Size and Productivity, the Post-HYV Period Data Inverse years Study Region Data notes relationship Remarks 1955-60/ C. H. H. Rao (1975) North and Andhra Pradesh, Punjab, Does not hold Relationship breaks down under 1966-70 South India Uttar Pradesh conditions of new technology. 1967-70 S. S. Johl (1971) North India Farm-size specific data, Lu- Does not hold Growing productivity and income dhiana disparities between large and small farmers. O\ 1968-72 S. L. Bapna West India Kota District, Rajasthan Does not hold Growing productivity and income (1973) disparities between large and small farmers. 1969-70 W. Khan and South India West Godavari, Andhra Does not hold Growing productivity and income R. N. Tripathy Pradesh disparities between large and (1972) small farmers. 1969-70 F. S. Bagi (1981) North India Farm-level data, Haryana Holds Small farmers are at least as effi- cient as large farmers. 1954-70 G. R. Saini (1980) All India and Disaggregated data, 9 states, May not hold Business incomes indicate that North India 11 districts, 25 cases along small farmers may be as pro- with business income data ductive as large farmers. for Punjab and Uttar Pradesh 1962-73 M. Chattopad- All India Disaggregated data, 10 May not hold Generally holds for all crops, does hyay and A. states, 12 districts, and 27 not hold for individual crops, Rudra (1977) cases and does not hold in all cases and at all times. 1968-71 S. S. Bhalla in All India Panel survey over three Holds Holds but is weakening. Berry and years for approximately Cline (1979) 3,000 cultivating house- holds 1970-71 A. Deolalikar All India Data for 272 districts Does not hold At low levels of technology (ferti- (1981) lizer use), inverse relationship holds; at higher levels the rela- tionship is reversed and larger farms are more productive. 1970-73 B. Dasgupta All India Village-level crop-specific Does not hold See consolidation of evidence in (1977a) studies Dasgupta (1977a), table 3.15. 1974 M. H. Khan Pakistan Irrigated Indus Basin farms May not hold Holds for all crops, positive for (1977) in Punjab (498) wheat. 1974 M. H. Khan Pakistan Irrigated Indus Basin farms May not hold Because of higher levels of non- (1979b) in Punjab (498) traditional inputs in large farms. 1975-76 P. L. Roy (1981) North India Punjab farm households Does not hold Districts with low levels of adop- (821) tion show inverse relationship; in districts with high levels of adoption relationship is re- versed. 108 Better Cultivating Practices: The Training and Visit System and choices. Both the willingness and, indeed, the efficiency with which poor households respond to opportunities within their limited means need to be kept in mind when assessing ways to improve their situation. There are two strategies for raising small-farm output, both of which are designed to draw on small farmers' main asset other than land-their labor resources. The first is intended to increase the out- put of a given crop per unit of area; the second is intended to increase the number of crops per unit of area in a given crop year. These strat- egies are discussed below, as well as the critical role of irrigation in increasing the output of food crops on small farms. The chapter con- cludes with a brief survey of small farmers' access to credit (which is essential if they are to buy the inputs needed for sustained growth in production) and markets (which provide the earnings needed to develop and expand South Asia's rural cash crop economies). Better Cultivating Practices: The Training and Visit System A noteworthy and successful approach to raising output of a given crop is the so-called training and visit (T&V) system of extension de- scribed by Benor, Harrison, and Baxter (1984). The system is de- signed to enable extension workers to convey information easily to farmers and persuade farmers to do simple things that will increase output immediately. The emphasis is on raising efficiency, especially that of costly inputs, by improving management practices and crop- ping sequences. In principle, these improvements are in land prepa- ration, seedbed and nursery maintenance, the spacing and density of plant populations, use of seed (including appropriate improved va- rieties), seed treatment, the timing of tilling, weeding, and fertilizing, and the methods of using pesticides and fertilizers. In fact, T&V SyS- tems have also introduced line transplanting of paddy, faster growing varieties of paddy, and the production of wheat on residual moisture in Assam; seed selection and treatment in Madhya Pradesh; tech- niques to improve plant populations in Gujarat; faster growing HYV paddy and early jute varieties to allow the rotation of crops (paddy- wheat and rapeseedjute) in West Bengal; new crops such as oilseeds, pulses, and HYV paddy in Orissa; lower-cost measures to protect cot- ton plants in Gujarat; and the ridge cultivation of maize in Rajasthan. There are several reasons for this emphasis on cultivating and man- agement practices. First, they are said to produce sure and immediate results, and adopting them exposes farmers to few risks. Second, al- though new practices may involve more work, they require little or no cash outlay. Third, the ability to derive the maximum benefit from new production technologies depends on first learning good crop Prerequisites for Sustained Gains in Output on Small Farms 109 management techniques. For example, if farmers apply fertilizers to a field that is not properly weeded or that has poor stands of plants, the net effect will be to fertilize weeds, which will successfully com- pete with crops for available soil, moisture, and light; the result can be a lower crop yield rather than a higher one. Under the T&V approach, new cultivation and management prac- tices are initially recommended for only a small part of a farmer's land, so that the methods do not appear unduly risky and so that the results can be compared with those of traditional practices on the farmer's own land. Since these practices are quite "well known and tested" (according to Benor, Harrison, and Baxter 1984), they can be incorporated into an extension service without elaborate and time- consuming screening and trials.2 The T&V method has been successfully used under a variety of rain- fed and irrigated conditions. The achievements and impact of T&v are difficult to quantify, but evidence suggests that they are considerable. In Orissa, the area devoted to drought-resistant crops such as oilseeds and pulses has increased significantly since the inception of the T&V system in 1977. In Assam, the area sown to HYV increased fourfold and fertilizer use increased 144 percent in one year after the intro- duction of the system. In Rajasthan, farmers have raised paddy yield from 2.1 tons to more than 3 tons per hectare in two years, a 43 per- cent improvement. In Madhya Pradesh, unirrigated and irrigated wheat yields rose from 1.3 tons to nearly 2 tons per hectare after one season, an increase of nearly 54 percent. The area under high- yielding paddy and wheat varieties in the entire state of West Bengal is reported to have grown substantially in a single year, as have crop- ping intensities and the employment of family labor. These impacts were found to be greater on small holdings than on large (see Ray 1979). These rapid gains in output were apparently not primarily because of better irrigation or weather, the discovery of a new crop variety, or greater use of fertilizer. In Rajasthan, for example, it was reported that yields rose while the use of purchased inputs fell. In irrigated areas where the T&V method was adopted, irrigation had been avail- able for years but had been poorly managed and used. A survey of beneficiaries in one of the World Bank's projects in Uttar Ptadesh found that although yields increased significantly with the installation of tubewells, they remained well below levels in irrigated areas in other states which had better extension organizations.' This suggests that better extension is critical for realizing the full potential of irriga- tion, fertilizer, and HYV investments. Some of the gains were the result of reduced applications of costly fertilizers or protective chemicals, or the result of more timely and 110 Better Cultivating Practices: The Training and Visit System better applications. Such achievements are often missed in statistics that are geared to show increased input use as a sign of progress. These gains are especially important for small farmers. Available evi- dence strongly suggests that a professional agricultural extension ser- vice developed on the same principles in each of these diverse areas was a major force behind these changes (see Benor, Harrison, and Baxter 1984). Evidence that following recommended practices has a significant impact on yields even on the smallest holdings is available from the Chambal canal project areas in Rajasthan and Madhya Pradesh (Cernea, Coulter, and Russell 1984, ch. 4). Yield increases of 50-60 percent are reported from the smallest holdings. Additional evidence on the effects of improved extension services comes from the moni- toring and evaluation of two Bank projects in Rajasthan and Haryana (see tables A6-A14). Although there are some minor differences, these and other data suggest the following generalizations: * The adoption rates for recommended practices are slightly higher for contact farmers (those regularly visited by T&V agents) than for noncontact farmers. - The adoption rates are higher in areas with T&V than in areas with traditional extension systems. • There are substantial differences in yields between those who adopt recommended practices and those who do not, and these yield differences increase with the number of recommended practices adopted. • Yields increase as the frequency of extension visits increases. * There are no significant differences either in adoption rates or in yields between farms of different sizes. • Small farmers are not being neglected by agents in favor of larger farmers. - Extension agents are the main source of knowledge leading to higher yields. Moreover, since the improved practices require more labor, rather than increased use of costly inputs, small farmers may in fact be able to adopt such recommendations more easily than larger ones (see Feder and Slade 1984 for detailed analysis of the Haryana data). These findings come from areas with previously poor records of agricultural performance and large numbers of smallholders. They clearly suggest that small farmers have an enormous potential for raising yields through simple improvements in cropping practices im- Prerequisites for Sustained Gains in Output on Small Farms 111 plemented under extension systems. It is important, however, to dis- tinguish carefully between yield increases that can be achieved merely with increased labor (associated with improved practices and of course knowledge) and those that require increases in other inputs, such as water, seeds, pesticides, nutrients, and credit. These inputs may not be available to small farmers-or they may not be profitable to use, a point that has often been ignored. It is very difficult to separate the impact of extension alone from that of other inputs that often accompany the introduction of new practices via extension. To do this requires careful comparison of data from areas where T&V has been introduced against data from areas where traditional systems prevail. Feder and Slade (1984) analyzed the impact of the T&V system in districts in Haryana by comparing them with similar districts in Uttar Pradesh where T&V had not been introduced. Their results suggest that when compared with tradi- tional extension services, the T&V system has significantly increased the diffusion of knowledge of recommended practices for HYV cere- als; there was no evidence that visits to contact farmers by extension agents were biased by farm size; and significant yield increases oc- curred after the introduction of the T&V system. Further, Feder, Slade, and Lau (1985) show that after accounting for the impact of other inputs, the impact on yield attributable to T&V alone was about 10 percent. The estimated rate of return of the T&V project, based on yield increases attributed to the new system alone, was approximately 40 percent in one district according to their calcu- lations. Their findings refute the consistent criticism that T&V has few benefits to offer, that extension agents bypass smallholders, and that contact farmers fail to pass information to other farmers. Extension services are rapidly becoming major sources of knowledge, especially with respect to more complicated and costly practices, and their im- pact on yields and production is significant (Feder, Slade, and Sundaram 1986). A growing body of research suggests that substantial increments in production and income are possible, even in semi-arid rain-fed areas, with improved cultivation practices and varieties, and with modest amounts of fertilizer properly applied. An example of the range of improvement possible is shown in table 4-4. These results suggest that technology already available can raise incomes substantially on the 70 percent of India's cropped land which is not irrigated. Much of the work done by Benor and Harrison, and others who stress improved farming practices, is based on reducing the gaps be- tween "average" yields and "best practice" yields in the same area. Al- though these gaps are smaller than those typically cited between Table 4-4. Net Income Increases as a Percentage of Gross Farm Incomes Net income increases Mean Grain Gross value as percentage of gross value yield price of mean Current' Improvedd Country and (kilograms (rupees or taka yield (rupees crop per hectare)a per kilogram) or taka)b HYV Traditional HYV Traditional Bangladesh Rice 1,043 198.0 2,065 29 7 93 63 India (irrigated) Rice 1,352 129.0 1,744 52 25 14 9 Wheat 1,572 136.2 2,141 33 9 9 28 Maize 1,749 86.0 1,504 23 _e 6 Jowar 943 120.0 1,131 1 -e - 2 - India (unirrigated) Wheat 837 136.2 1,140 -c 7 _ 5 Maize 896 86.0 771 -2 46 0 34 Jowar 486 120.0 583 6 _e 0 a. For India, 1973-74 average irrigated and unirrigated yields are used; for Bangladesh, 1972-73 yields are used. b. For India, 1976 prices are used; for Bangladesh 1977, first quarter prices are used. c. When moving from zero fertilizer use to current levels. d. When moving from current levels of fertilizer use to optimum levels. e. Less than 0.5 percent. Source: World Bank data. Prerequisites for Sustained Gains in Output on Small Farms 113 Table 4-5. India: Examples of Yield Gaps in Cultivators' Fields, 1977 Present yields Increase in (kilograms per hectare) yIelds possible State and crop Average On bestfieklsa (percent) West Bengal Rice Aman 1,100 2,500 127 Aus 1,000 2,000 100 Boro 2,500 5,000 100 Wheat 2,000 4,000 100 Pulses 600 1,200 100 Oilseeds 400 1,000 150 Madhya Pradesh Paddy 750 1,600 113 Wheat 730 2,000 174 Pulses 690 1,600 132 Other cerealsb 720 1,800 150 Rajasthan Wheat 1,300 2,000 54 Pulses 500 1,600 220 Other cerealsb 740 1,800 143 a. Average on best farmers' fields is lower than obtained under research conditions. b. Sorghum, barley, maize, millets-barley. Source: World Bank data. farmers' yields and those obtained under research conditions, they can be very large-anywhere from 50 to 250 percent depending on the crop and region (table 4-5)-and they provide an easily observ- able measure of potential gains. The few data available suggest that, without any increase in inputs, output can be raised by 15-30 percent in rain-fed areas and by 25-50 percent in irrigated areas, depending on the crop (see tables 4-5 and A15). Further increases in yields re- quire varietal changes or increased use of inputs, or both. There are thus discrete stages in the adoption of improved prac- tices, each providing an incremental yield. But not all stages are equally accessible to small farmers. Data from rain-fed areas in Assam (table A15) make this point. Stage 1 consists only of changes in crop- ping practices; stage 2a requires varietal changes, and 2b requires the use of additional nutrients; stages 3a and 3b involve not only new va- rieties and more nutrients, but also management-intensive practices such as the use of appropriate pesticides. Stage 2 requires the reorga- nization of research, extension, and marketing systems (for the deliv- ery of inputs) to make sure that small farmers have access to them. Stage 3 requires management capabilities, which means farmers must 114 Better Cultivating Practices: The Training and Visit System be educated and given a proper understanding of soils, plants, and nutrients so they can make intelligent decisions. All innovations do not come from experts or extension agencies. In Bangladesh small farmers themselves have pioneered a variety of techniques for multiple crops and intercropping. They include broad- cast upland rice sown during pre-monsoon showers and followed by a second rice crop; on land that is susceptible to flooding, a quick- maturing upland rice and a longer-maturing deep-water rice sown to- gether but harvested separately, followed by a crop of pulses, oilseeds, or wheat in the dry season using residual soil moisture (Brammer 1980). These practices demonstrate an amazing adjust- ment of the farmers' choice of crop and cropping patterns to the spe- cific micro environment in their fields. One result is the multiplicity of rice varieties that have been adopted.4 Double sowing of paddy and cultivation of wheat on soils unsuited for rice are other examples of innovations by the farmers themselves. Farmers as a community are not so dependent on research and government programs as govern- ments and aid organizations widely presume. They readily adopt practices that maximize the use of their scarce land. These intercrop- ping practices do not necessarily require mechanized cultivation, nor is irrigation essential where deep loamy soils have a high moisture content, as in parts of Bangladesh. Even fertilizers are not essential for continuous cropping on seasonally flooded soils (Brammer 1980). To summarize, small farmers can easily increase their yields by 10-50 percent without any increase in inputs if they are taught better practices. Nevertheless, the data just presented and the apparent suc- cess they reveal should be treated with caution. The samples are small, the evidence is limited, and it is too early to evaluate the full impact of the changes over time. At best, these magnitudes should be treated as only indicative of the type of yield increases that can be brought about by changing simple cropping practices. Further in- creases (of 75-150 percent) will depend on the price and availability of inputs, access to these inputs, the availability of credit with which to purchase them (see below), and better-educated small farmers. In India the reform of extension systems along the lines of the T&V approach involves the consolidation, reorganization, and strengthen- ing of existing extension agencies rather than the creation of new or- ganizations. This approach uses the experience and skills of the exist- ing staff, avoids duplication of effort, and keeps costs low. Twelve states have already adopted the T&V extension system through World Bank-assisted projects. When fully developed these programs will cover some 70 percent of India's farm holdings and some 65 percent (45 million) of India's farm families. Bangladesh and Pakistan are in the process of adapting the T&V approach to their needs. Prerequisites for Sustained Gains in Output on Small Farns 115 Scientific Multiple Cropping In addition to better cultivation practices, multiple cropping- growing more than one crop on the same piece of land in a given year-can make the little land available to small farmers more pro- ductive. Multiple cropping increases both the area cropped per year and the total yield per unit of area. Although an ancient practice on the subcontinent, multiple crop- ping has gained considerable attention lately. Until the early 1960s most multiple cropping was of vegetables grown near urban areas and irrigated with sewage. The recent concept of scientific multiple crop- ping is an outcome of the existence of reliable irrigation, awareness of the value of commercial fertilizers, and the development of quick- maturing, high-yielding varieties of cereals, grains, legumes, and other crops. The objective is to maximize production and economic returns per unit of area and per unit of time. (Rajat and Bhardwaj 1974.) Table 4-6 gives figures for multiple cropping in selected countries in the 1960s. In South Asia 15-40 percent of the planted area was multiple cropped, in East Asia 30-80 percent. The multiple cropped area did not increase much from 1950 to 1969 in the subcontinent (see table 4-7), in spite of the increase in irrigation with which it is strongly linked. The index of multiple cropping remained fairly sta- ble for India at between 112 and 115 during this period.5 For Pakistan it has been even lower, but the range is about the same. In Bangladesh Table 4-6. Index of Multiple Cropping in Selected Economies Multiple cropped area Economy Years Index (1,000 acres) Burma 1965-66 111.1 2,162 Pakistan 1966-67 111.7 4,116 India 1962-63 115.0 48,456 Indonesia 1964 126.2 5,248 Philippines 1960 136.0 4,982 Bangladesh 1968-69 139.2 8,479 China 1968 147.4 127,388 Korea, Rep. of 1969 153.4 3,074 Egypt 1961-62 173.0 4,388 Taiwan 1969 184.3 1,905 Note: The index is derived by dividing the total land area planted during the year (including that planted as a result of double cropping) by the amount of cultivated land physically available and in use. Source: Compiled from data presented in Dalrymple (1971). Table 4-7. South Asia: Multiple Cropped Area and Index Bangladesh India West Pakistan Multiple cropped Multiple cropped Multiple cropped Season area (1,000 acres) Index area (1,000 acres) Index area (1,000 acres) Index 1949-50 5,564 127.6 38,108 112.0 n.a. n.a. 1950-51 5,708 127.8 32,486 111.1 3,131 111.1 1951-52 6,080 129.4 34,181 111.6 2,333 108.4 1952-53 6,574 131.4 35,169 111.5 2,066 107.4 1953-54 6,909 133.1 38,730 112.4 2,825 109.4 1954-55 6,572 131.4 40,119 112.7 3,503 112.0 1955-56 5,509 126.9 44,861 114.1 3,875 112.7 1956-57 5,468 126.7 45,916 114.2 3,774 112.1 1957-58 5,620 127.7 41,340 113.0 2,405 111.0 1958-59 5,366 127.0 48,928 115.0 4,520 114.3 1959-60 5,902 128.7 49,136 115.0 3,105 109.6 1960-61 6,732 132.3 48,330 114.7 2,421 107.5 1961-62 6,453 130.8 51,266 115.3 3,447 110.4 1962-63 6,796 132.5 50,465 115.0 3,187 109.4 1963-64 7,095 133.7 50,468 115.0 4,191 112.9 1964-65 7,434 135.2 51,931 115.2 5,140 114.7 1965-66 7,940 136.8 47,119 114.0 4,473 112.9 1966-67 7,926 137.6 48,456 114.4 4,116 111.7 1967-68 9,694 144.6 n.a. n.a. 3,046 108.5 1968-69 8,479 139.2 n.a. n.a. n.a. n.a. n.a. Not available. Note: The index is derived by dividing the total land area planted during the year (including that planted as a result of double cropping) by the amount of cultivated land physically available and in use. Source: Dalrymple (1971). Prerequisites for Sustained Gains in Output on Small Farms 117 a larger percentage of the area is multiple cropped (30-40 percent), and the trend has been positive. Although aggregate data for South Asia suggest that multiple crop- ping could be increased substantially, they do not reflect the full pic- ture for small farms. As one would expect, multiple cropping is in- versely related to farm size. Small farmers farm their land more intensively, having the available family labor to do so; hence their yields per unit of area and per unit of time are much higher than those on large farms. Although only 9 percent of the cultivated area is multiple cropped in India and 11 percent in Pakistan, figure 4-1 shows that the multiple cropped area on farms smaller than five acres ranges from 16 to 27 percent for India and from 30 to 52 percent for Pakistan. The figures for Bangladesh are even more dramatic. Whereas 53 percent of the cultivated area in Bangladesh is multiple cropped, 58-82 percent of the area of farms smaller than five acres is multiple cropped. Indeed, for farms of less than one acre the cropping intensity is almost as high as that in Taiwan, which is often cited as the prime example of effec- tive multiple cropping systems. Raising the number of small farms in India and Pakistan that produce two crops a year ultimately depends Figure 4-1. South Asia: Multiple Cropped Area as a Percentage of Cultivated Area, by Farm Size Percent 90 B Bangladesh, 1968 0 Pakistan, 1972 80 - India, 1971 70- 60- 50- 40 30 20 10 0- < 1.0 1.0-2.5 2.5-5.0 5.0-7.5 7.5-12.5 12.5-25.0 25.0+ Farm size (acres) Sources: Bangladesh Bureau of Statistics (1972); India, Ministry of Agriculture and Irri- gation (1975); and Pakistan, Ministry of Food and Agriculture (1975). 118 Scientific Multiple Cropping on the availability of water and its control through irrigation. In Ban- gladesh, where many farmers are already getting nearly two crops a year, the objective is to get three crops a year. Tables 4-8, 4-9, 4-10, and A16 and A17 provide additional data on cropping intensities by farm size for Bangladesh, Pakistan, and India. They clearly highlight the inverse relationship between farm size and cropping intensities (see figure 4-2) as well as the importance of irri- gation to higher intensities (see below). They show how small farms overcome the handicaps of their size by farming what they have far more intensively. This is a central reason why land area is not a good proxy for the income potential of small holdings. The input requirements of multiple cropping can be substantial. Necessary (but not sufficient) conditions for its wider adoption in- clude the use of improved varieties of seed, better water control, higher inputs of fertilizer, better insect and disease controls, and more efficient postharvest technologies. New varieties of crops are of central importance in promoting multiple cropping. Early maturing varieties are generally associated with a rise in multiple cropping; the shorter the growing season required by a given crop, the greater the possibility of raising an additional one within the available growing time (Dalrymple 1971, p. 23). Other desirable characteristics, espe- cially for grains, are yield rather than vegetative responsiveness to fertilizers; short, strong stems that limit "logging"; as much insect and disease resistance as possible; and acceptability to consumers. In addi- tion, varieties that are to be grown in rain-fed areas often need to be drought-resistant. The high-yielding varieties of wheat and rice devel- oped in the late 1960s have many of these characteristics and have been extensively adopted in multiple cropping systems. Since the rainfall in large areas of South Asia is both inadequate and unsuitably distributed for multiple cropping, increased irrigation is also essential.'Irrigation is discussed separately in the next section; its role in raising cropping intensities is summarized here. A policy conflict can arise between increasing the supply of water during the single existing growing season or extending the growing season to cover other parts of the year. Although irrigation during appropriate periods helps make multiple cropping possible, the relationship works both ways: multiple cropping may be necessary to justify the expense of putting irrigation systems in place. Indeed, this is a neces- sary condition for many agricultural projects in South Asia. Many of the new irrigation projects pay particular attention to projected crop- ping intensities. The extensive existing irrigation systems in South Asia were not de- signed to maximize output per acre, however, but rather to spread Table 4-8. India: Cultivated Area, Cropping Intensity, and Irrigation by Size of Operational Holding, 1970-71 Cultivated area Size Total area of Cropping Percentage of cultivated area of holding holdings (millions Millions of As percentage Gross cropped area intensity (acres) of acres) acres of total (millions of acres) (percent) Irrigated Unirrigated < 1.0 13.46 12.52 93 15.93 127 34 66 1.0-2.5 22.48 21.14 94 25.89 122 31 69 < 2.5-5.0 47.62 44.21 93 51.40 116 28 72 C 5.0-7.5 40.38 37.52 93 42.66 114 27 74 7.5-25.0 152.85 138.76 91 150.15 108 23 77 25.0-50.0 70.44 62.47 89 63.60 102 16 84 50.0+ 53.20 41.72 78 40.80 98 12 88 Total 400.43 358.34 89 390.43 109 23a 78b Notes: Column may not add to total because of rounding. a. 88.2 million acres. b. 302.2 million acres. Source: India, Ministry of Agriculture and Irrigation (1975). Table 4-9. Pakistan: Cultivated Area, Cropping Intensity, and Irrigation by Size of Farm, 1960-61 and 1972 Cultivated area Cross cropped area Percentage ofcultivated area Percentage of total Percentage Percentage of total Percentage intensity' Irrigated Unirrigated Farm size __ increase increase (acres) 1960-61 1972 in group 1960-61 1972 in group 1960-61 1972 1960-61 1972 1960-61 1972 < 1.0 1 b - 73 1 b -67 138 152 54 65 46 35 1.0-2.5 3 1 -54 3.5 2 -46 130 137 63 72 37 28 2.5-5.0 7 4 -31 8 5 -22 128 131 66 75 34 25 5.0-7.5 8 8 2 9 9 12 127 125 70 78 30 22 4 7.5-12.5 17 20 28 18 22 37 124 119 73 81 27 19 12.5-25.0 29 29 10 30 29 15 119 111 74 78 26 22 25.0-50.0 20 19 5 19 18 10 115 103 68 70 32 30 50.0-150.0 10 13 35 9 11 53 109 97 47 60 53 40 150.0+ 5 5 22 2.5 4 87 108 88 58 60 42 40 Total 100 100 9 100.0 100 17 120 111 68 74 32 26 Millions of acres 37.25 40.69 38.31 44.98 25.2 29.98 a. Censuses provide cropping intensity based on net sown area (86 percent and 92 percent of total cultivated areas in 1960-61 and 1972 respectively). b. Less than 0.5. Sources: Pakistan, Ministry of Agriculture and Works (1960b), tables 13 and 17; Pakistan, Ministry of Food and Agriculture (1975), tables 24, 25, 26, and 27. Table 4-10. Bangladesh: Cultivated Area, Cropping Intensity, and Irrigation by Size of Farm, 1960-61 and 1967-68 Cultivated areaa Gross cropped area Cropping intensity Percentage of cultivated Farm size Percentage of total . Percentage of total . (percent) area,b 1960-61 Far1n szze ~~~~Percentage increase Percentage increase (ecn)ae, 906 (acres) 1960-61 1967-68 in group 1960-61 1967-68 in group 1960-61 1967-68 Irrigated Unirrigated < 0.5 1 1 11 1 1 25 166 186 6 79 0.5-1.0 2 3 32 2 3 40 171 181 6 76 1.0-2.5 13 17 31 16 19 36 166 172 6 75 2.5-5.0 27 30 13 28 31 15 155 158 6 75 X 5.0-7.5 20 18 -8 20 17 -7 147 149 6 75 7.5-12.5 19 16 - 19 18 15 - 16 140 145 7 75 12.5-25.0 14 11 -20 12 9 -21 132 131 8 73 25.0-40.0 3 3 29 2 3 20 127 117 9 72 40.0+ 1 1 -3 1 1 -7 117 113 13 73 Total 100 100 1 100 100 4 148 153 7 75 Millions of acres 18.85 19.03 27.88 29.08 1.32 14.3 Notes: Because of rounding, figures may not add to total. Increase in group is the absolute percentage increase. a. Cultivated area is taken as net sown area, consistent with cropping intensity. b. 1960-61 data show 18 percent of total cultivated area as flooded; this is not shown in this column. Sources: Pakistan, Ministry of Agriculture and Works (1960a), tables 12, 13, and 18; Bangladesh Bureau of Statistics (1972), table 7. 122 Scientific Multiple Cropping Figure 4-2. South Asia: Cropping Intensities by Farm Size Percent 200 - Bangladesh, 1967-68 [ Pakistan, 1972 180 - . E India, 1970-71 160 140 120 100 80 <1 0 1.0-235 2.5-50 5.0-75 7.5-25.0 25.0-50.0a 50.0+a Farm size (acres) a. For Bangladesh, the largest size groups are 25.0-40.0 and 40.0+ acres respectively. Sources: Tables 4-8, 4-9, and 4-10. the available water as widely as possible in order to reduce the proba- bility of crop failure (see Mellor and others 1968). Especially in the case of canal irrigation, which is widely used in the subcontinent, the quantity of water available depends largely on seasonal flows. The main drawback to canal irrigation in Pakistan is that the supply is un- reliable (Mohammad 1965; Kaneda and Ghaffar 1970). When the water is most needed-in the dry season or for the second crop-it is likely to be most scarce (Dalrymple 1971, p. 25). This has led to the rationing of water in the winter in Pakistan, which has in turn lim- ited opportunities for multiple cropping (Hussain 1970). Unless the seasonal distribution of water is improved, it is not possi- ble to increase cropping intensities. In this connection, both well water and tubewell irrigation are preferable to canal-based systems, because they are less subject to seasonal variations in supply. There is clear evidence that the multiple cropping index is generally higher in regions where tubewell irrigation is practiced than in canal- irrigated regions. In the Indian Punjab the multiple cropping index in regions irrigated by wells in 1968-69 was 150 percent as opposed to 125 percent in canal-irrigated areas; it was expected to rise to be- Prerequisites for Sustained Gains in Output on Small Farms 123 tween 180 and 200 percent by the 1980s (Billings and Singh 1970; I. Singh and Day 1975a). Similarly, in West Pakistan in 1963-64 the cropping index for tubewell farmers in three districts averaged about 138 percent, as opposed to 107 percent for those without tubewells. Irrigation is only one aspect of the water control problem; the other is drainage. Adequate drainage is needed in the short run to ensure proper irrigation; in the longer run, it is essential in order to prevent chronic waterlogging or salinization of the soil (which is extensive in Pakistan). If drainage schemes are in the hands of local elites, small farmers may again be the last to receive adequate assistance. One of the drawbacks of multiple cropping is that the soil is impov- erished (unless replenished by annual flooding, as in Bangladesh). Thus, in addition to quick-growing crop varieties and reliable irriga- tion and drainage systems, the further expansion of multiple crop- ping on small farms requires the use of organic or chemical fertilizers to reinvigorate the soil.6 Although extensive work has been done on the new high-yielding varieties, relatively little is known about the re- lationship between different crop rotations and fertilization needs. Ultimately, small farmers' use of fertilizer depends on whether they have access to supplies of it and whether they find its use profitable. Without the increased use of nutrients, however, extensive multiple cropping systems will be difficult to sustain (Swaminathan and Bains 1970). Finally, multiple cropping requires a longer life span for plants, which in turn requires better controls for insects and diseases. Unfor- tunately, some of the new high-yielding crop varieties are much more susceptible to local diseases and insects than are the native varieties. Four possible solutions to this problem are: stepping up chemical con- trol measures, breeding for natural plant resistance, attempting bio- logical control, and arranging rotations so that no two crops sharing the same pests and diseases are grown in sequence (Dalrymple 1971, p. 28). The first three have major difficulties and limitations. Chemi- cal control can be expensive and cause ecological problems; more- over, chemical pesticides are cash-intensive and may be out of small farmers' reach. Breeding for natural resistance and developing bio- logical controls are time-consuming processes, and it is not certain whether all the necessary research has been done, especially in Ban- gladesh and Pakistan. The last method, the use of appropriate rota- tions, is the most logical starting point. What can be accomplished in practice depends, however, on adaptive research into the develop- ment of farming systems especially suited to the conditions and needs of small farmers on the subcontinent. Thus, much needs to be done before the potential for multiple cropping in South Asia can be realized. Moreover, multiple cropping 124 Scientific Multiple Cropping presents fewer opportunities for small farmers than for larger ones. Small farmers may have relatively limited access to the necessary new inputs, and raising the area under multiple cropping, especially with existing cropping intensities of over 150-160 percent, requires a huge advance in farmers' management capabilities. Small farmers are shrewd decisionmakers, but they generally lack the scientific informa- tion needed to handle complex cropping systems. The most critical management problem has to do with timing. In South Asia, which has year-round sunshine and no climatic barriers to three or even four crops a year, the key obstacle to extensive multi- ple cropping is the length or timing of the growing season.7 As more crops are planted and the growing period between them is shortened, less time is left between the harvesting of one crop and the planting of the next. Minimizing the number of days during which the land is left idle is a difficult job for the farm manager. A large number of operations need to be carried out in a relatively short period. The small farmer's advantage of having a relatively abundant supply of labor may well be offset by his disadvantages in obtaining other in- puts. If seeds, nutrients, and irrigation water are not available on time, they might as well not be available at all; if small farmers have to wait for these inputs, their timing is likely to be upset and the prob- ability of succeeding with multiple cropping practices is likely to be correspondingly reduced. Even though adaptive research has provided appropriate cropping rotations for parts of India and Bangladesh, the problem of timing has prevented their general use. Most of these rotations leave little margin for delays caused by weather or the unavailability of inputs. Delays in crop maturity may reduce yields or create problems with harvesting one crop and planting the next.' Mechanization, which has been widely adopted in some parts of the subcontinent, can speed up the work, but it creates its own difficulties; in any event, this is not a viable solution for small farmers. Timing can therefore be expected to continue to be an obstacle to the development of extensive and sta- ble multiple cropping systems. If the various impediments to sustained increases in cropping in- tensities can be effectively reduced or removed, however, the benefits to small farmers and the agricultural sector as a whole could be con- siderable. Multiple cropping can virtually double the demand for labor and raise the net incomes of the poorest farmers by 30-60 per- cent.9 An example of the possible net impact of multiple cropping sys- tems on income is shown in table 4-11. Increases in incomes are low unless one moves beyond double cropping to triple and quadruple cropping, when increments of 50-100 percent are possible. I do not know how representative these data are, so it is difficult to predict the Prerequisites for Sustained Gains in Output on Small Farms 125 net increases in the incomes of small farmers. If a farmer has a very small holding and few off-farm employment opportunities, it may be necessary for him to move to higher cropping intensities simply to maintain a minimal income. Such a tendency is clearly evident from the data on cropping intensities on small farms in Bangladesh.' It is reasonable to suppose that the total amount of labor used rises substantially with multiple cropping, but little empirical evidence is available on this point. Data from a sample of farmers in the Indian Punjab during the 1960s indicate that labor requirements rose faster than cropping intensities (Johl 1971). Evidence from northwestern India shows that when quadruple cropping increases production about threefold, the demand for labor rises about fivefold (Rajat and Bhardwaj 1974). Again, it is difficult to assess how representative these data are. But labor requirements also vary considerably with the type of water supply and with the kinds of crops, varieties, and crop- ping intensities. Canal irrigation requires twice the labor of tubewells, while water supply by persian wheels requires ten times more again (Singh, Day, and Johl 1968). When sugarcane is included in the rota- tion, labor requirements are far higher than with other crops. Im- proved crop varieties require more labor per unit of land (although not always per unit of output) than traditional varieties. Finally, the variation in incremental labor requirements widens as one moves from single to double and from double to triple cropping. It is thus hard to assess how much additional employment could be generated if multiple cropping were adopted more widely. One such estimate for India suggests that about 17 million additional peo- ple could be employed directly, with an additional secondary and ter- tiary impact on employment in input services and transportation (Rajat and Bhardwaj 1974, p. 659). It is not certain, however, how much the additional labor for multi- ple cropping would add to the number of jobs, and how much it would simply reduce the underemployment of existing farm labor. Experience in Pakistan shows that an increase in multiple cropping led to a less than comparable increase in the farm labor force, which suggests a substantial degree of previous underemployment (Cownie, Johnston, and Duff 1970). The same must certainly be true for India and Bangladesh. If labor requirements increase substantially, large operators (who have to hire labor) might find further increases in cropping intensity unattractive; on the other hand, small operators (who rely on their families' labor) might find them very profitable, especially where there are few off-farm jobs and consequently little or no loss of wage income if workers are needed on the family farm. The extent of underemployment during much of the year, espe- cially on small family farms, would be reduced. Farm management Table 4-11. Northwestern India: Economics of Multiple Cropping Sequence ON Rupees per hectare Net profit index Gross with base of Benefit-cost Cropping system Crop income Expenditure Net profit single cropping ratio Single cropping Wheat 3,572 1,775 1,797 100 2.01 Double cropping Wheat 3,420 1,775 1,645 Maize 2,200 1,813 387 Subtotal 5,620 3,588 2,032 113 1.57 Triple cropping Green gram 1,000 636 364 Maize 2,200 1,813 387 Wheat 3,750 1,775 1,975 Subtotal 6,950 4,224 2,726 152 1.65 Maize 2,037 1,813 224 Potato 4,954 3,055 1,899 Onion 5,234 2,225 3,009 Subtotal 12,225 7,093 5,132 286 1.72 Green gram 1,095 636 459 Red gram 1,078 1,000 678 Wheat 3,040 1,775 1,265 Subtotal 5,813 3,411 2,402 134 1.70 Quadruple cropping Green gram 1,000 636 364 Maize 2,200 1,813 387 Potato 4,500 3,055 1,445 Wheat 2,128 1,575 553 Subtotal 9,828 7,079 2,749 153 1.39 Note: The table refers to the Northwest Plains zone comprising the states of Delhi, Haryana, Punjab, northeastern Rajasthan, and western Uttar Pradesh. Source: Rajat and Bhardwaj (1974), p. 654. 128 Irrigation studies in India have shown that as much as 25-30 percent of all fam- ily labor engaged in farming is not gainfully employed, because farm operations are not evenly spread over the year." Multiple cropping clearly helps to shorten slack periods, but the problem of seasonal peaks in the demand for labor during planting and harvesting is likely to be exacerbated and may lead to mechanization (Billings and Singh 1970; Johl 1971; Day and Singh 1977). How much this will affect small farmers will depend on a host of factors. As noted earlier, mech- anization is unlikely to be an option for this group in the short term. Eventually, however, some kind of labor-saving machinery for har- vesting and planting operations will need to be developed for small farmers so that they can handle production bottlenecks.'2 At any rate, multiple cropping promises to be more effective in cre- ating employment and income for rural populations than either rural industries or rural work programs (discussed in chapters 6 and 7). Small farmers and the landless will be the most likely beneficiaries, if the delivery systems for fertilizers and water can meet their needs, and if appropriate solutions can be found for the other technological and management problems associated with multiple cropping. Irrigation Irrigation and controlled water management hold the key to increas- ing agricultural productivity in South Asia. Irrigation is essential for higher yields because it permits a shift to higher production func- tions; because the proper management of complex cropping systems calls for controlled water; because high-yielding varieties perform best under irrigated conditions; because irrigation reduces the risks associated with the variability of weather and monsoons; and because crops are more responsive to organic and inorganic fertilizers under irrigated conditions. In India there is a clear inverse relationship between the size of holding and the area under irrigation. All farms in India taken to- gether irrigated about 22 percent of their gross cropped area in 1970-71, but farms smaller than five acres irrigated significantly more-28 to 34 percent of their area (table 4-8). This relationship is much weaker in Pakistan. A far higher percentage of the total area was irrigated in 1972 (roughly 74 percent on all farms); for the small- est farms the irrigated percentage (65-75 percent) was not above the average (table 4-9). In Bangladesh a very small percentage of the area was irrigated (6-7 percent) and there were no substantial differ- ences among farms of various sizes (table 4-10). By 1985, despite major investments in water development, irrigation covered no more than 22 percent of the cultivated area according to World Bank data. Prerequisites for Sustained Gains in Output on Small Farms 129 Small farms may irrigate a higher proportion of their gross cropped area, but they account for only small proportions of the total irrigated areas of India and Pakistan. In India, farms of fewer than five acres accounted for 22 percent of the total irrigated area (and 19 percent of the cultivated area); in Pakistan for only 6 percent (5 percent of the cultivated area). In Bangladesh the figures are much higher, nearly 50 percent of the irrigated area (51 percent of the culti- vated area). The aggregate data show only that small farmers irrigate propor- tionately as much, if not more, than larger ones. They give no indica- tion of the intensity of water use, the timing of the availability of water in farmers' fields, or the efficiency of water use once it gets to the fields. These issues are central to the problem of small farmers' access to irrigation. We need to establish the extent of the unutilized potential for irri- gation in the subcontinent and how much of this potential is likely to be made available to small farmers. There are two kinds of irriga- tion to be considered: surface irrigation, mainly through canals, and groundwater irrigation through wells and tubewells. Government studies indicate that there is considerable potential for increased irri- gation from both these sources in South Asia (tables A18 and A19) subject to investment expenditures that may or may not be made. But despite this substantial potential for increased irrigation from both surface and groundwater, a number of factors may prevent small farmers from getting access to it. With surface irrigation, the problems are: * Lack of investment in farm-level tributary irrigation and proper field-level water courses and drainage systems. Old-style irriga- tion projects that left the development of secondary and tertiary systems to local authorities may have unwittingly biased the devel- opment of surface irrigation in favor of larger and better-off farmers. * Lack of access to and control over water flows. Small farmers may be located fairly far downstream in the water distribution net- work and may not be able to get canal water released when they need it. Since timing is critical, especially in periods of drought or low rainfall, small farmers may suffer from being last in line to receive scarce water resources. Rationing during periods of scarcity is likely to benefit farmers who have good relations with the local authorities. * Lack of credit for financing long-term on-farm investments such as water courses, drainage systems, and wells. This can be a spe- cial problem for sharecroppers and tenants who lack clear title 130 Irrigation to the land they farm. Reliance on land and physical assets as col- lateral for long-term loans discriminates specifically against these groups because they do not own such assets. * Costs of water. Although not discriminatory, water rates can be prohibitive to the small farmer with little cash. Subsidizing water rates for small farmers is one solution but is not easy to adminis- ter. With groundwater irrigation the problems are: * Lack of investment funds or medium- and long-term credit to sink deep wells or put in tubewells. * Economies of scale that militate against the use of private tubewells on small and fragmented holdings. Irrigation coopera- tives can be established to overcome this problem, but organiza- tions of this kind suffer from a problem endemic to cooperatives in general: the larger and better-off farmers end up controlling the system and getting most of the benefits. * Lack of public wells and tubewells. A definite bias against smallholders and tenants who belong to lower castes restricts their use of existing public deep wells, and public tubewells that were installed to overcome the problems of lumpy investments have often turned out to benefit larger holders. * Fragmentation of holdings prevents groundwater irrigation from being used to its best advantage.'3 Where holdings are fragmented, cooperation between neighboring farms is essential if wells are to be sunk and used productively. The minimum economic area for well-water irrigation depends on how the water is shared among farmers (see Dhawan 1977). Alternatively, the operators of large fragments may install wells and sell water to their neighbors, a solution that permits larger farmers to become mo- nopoly suppliers of a scarce resource. As I show in chapter 8, small farms are generally made up of fewer parcels than larger farms. Nevertheless, fragmentation is a critical problem, for a small farmer may be able to borrow only a little money to install a well, and the investment cannot be spread effectively over as many as six small parcels of land. Although other diseconomies and inefficiencies are associated with the fragmentation of holdings, the barriers to the effective use of groundwater are especially crip- pling for the small farmer. Smallholders are thus likely to get less than their share of increased irrigation in the 1990s. Moreover, the further development of both surface and groundwater irrigation systems faces real difficulties. In Prerequisites for Sustained Gains in Output on Small Farms 131 India, for example, the actual use of irrigation systems lags far behind their completion. The government has introduced command area de- velopment (CAD) programs (see below) to improve the situation. The funding of such programs has not been easy, but the crucial con- straint has been the time required to build the institutional frame- work to implement and administer the complex components of CAD schemes. An estimated 6.25 million acres was potentially irrigable from existing facilities by the end of the fifth plan (1974-79). In rela- tion to the extent of the task, coverage by CAD projects has been modest-not more than 250,000 acres by 1985. Despite its ambitious plans, CAD cannot be expected to have a major impact on the gap be- tween the potential created and its actual utilization for a long time. In addition, the design and operating standards of existing and planned large and medium-size surface irrigation systems in India need to be upgraded if India's scarce water resources are to be used efficiently. Upgrading is of special concern since the introduction of high-yielding varieties, which require more water, and within more narrowly restricted periods, than do traditional varieties. Most of the irrigation systems built over the last 200 years were designed to de- liver relatively small supplements to rainfall over a wide area to pro- tect the crops. The systems generally lacked controlled structures to deliver water selectively when crops needed it. Furthermore, utiliza- tion efficiencies-that is, the proportion of the water entering the sys- tem that is beneficially used-are generally very low. These difficulties have caused India to change its strategies for the development of surface irrigation. Initially, even when projects incor- porated components to strengthen agricultural extension services or on-farm development, these features were largely ignored in practice. Experience with these earlier projects made it clear, however, that construction of irrigation infrastructure alone had little impact on ag- ricultural output, and that the integration of irrigation and agricul- ture was essential to make better use of the water and to realize the anticipated benefits from projects more rapidly. This conclusion led in the early 1970s to command area develop- ment, which constructs field channels and drains, and levels and shapes the land to ensure an effective and reliable water supply at the farm level. Some of the larger projects now involve 100,000 or more farmers, each of whom has to be dealt with individually. Before the actual work of land development can begin, the land records of every farmer must be updated, farm boundaries established, the land surveyed, and development plans prepared for loans from land de- velopment or commercial banks. As universal participation is essen- tial, special arrangements must be made for providing finance from outside normal credit channels if a farmer's creditworthiness is in 132 Irrigation doubt. These arrangements are especially necessary for the small farmers included in the schemes. The World Bank's experience with CAD has been mixed. The Raja- sthan Canal project, which also included the reorganization of exten- sion services, was very successful. This is an atypical example, how- ever, since it was also a settlement project and hence involved no problems of land records and credit. In the Chambal project in Raja- sthan and Madhya Pradesh, progress with CAD has been painfully slow, despite the special emphasis placed on organizational features associated with land tenure. It is now generally agreed that if irrigation development is ap- proached piecemeal, project by project, a significant program of on- farm works cannot be implemented, particularly in view of the esti- mated backlog of 25 million acres needing such works. To speed development, the problem is being approached on a statewide basis. It is believed that the solutions to organizational, administrative, fi- nancial, and legal problems associated with land development can be replicated across states. Other breakthroughs have been to improve water use through bet- ter engineering design and to finance all works except land leveling from budgetary sources, thus bypassing credit institutions and ending the need for clear land titles. The main problems that remain have to do with cost recovery, either by raising water charges or by refi- nancing the budgetary portion of expenditures through credit insti- tutions. To what extent have small farmers benefited from surface irriga- tion, CAD or otherwise, in India? Areas have hitherto been earmarked for surface irrigation mainly on technical grounds, with preference being given to dry and drought-prone areas. These areas tend to have larger but very poor farms, which have considerable potential once they are irrigated. Traditionally, larger farmers (and particularly those at the head of a water course) have benefited most. Bringing the water supply to the farm level and financing irrigation schemes from budgetary sources rather than by credit should, however, greatly increase the participation of small farmers in the benefits of irrigation. 14 To the extent that surface irrigation requires field channels, and to the extent that such investment requires valid and generally accepted land titles, both small farmers and tenants stand to gain little. The former lack the credit to invest in field channels, and the latter lack both credit and clear title to the land they cultivate. The inability of current organizational structures to deliver water from developed ir- rigation sources to small and tenant farmers is well known, but the reforms needed have not been defined satisfactorily. Prerequisites for Sustained Gains in Output on Small Farms 133 Similarly, enormous problems exist in the development of ground- water potential. The major force in irrigation development in India has been private investment in open wells and tubewells. The number of open wells rose from about 3.6 million in 1956 to about 6.1 million in 1971 (India, Ministry of Agriculture and Irrigation 1976, pt. 5, p. 18); tubewells proliferated further in the wake of the new technolo- gies. A private tubewell enables the cultivator to water his crops at the right time and with the required quantity-freedom of action of immense value. In India at the end of the fourth plan (1969-73) there were some 6.8 million wells in use and about 4.1 million diesel and electric pumps. From the mid-1960s to the early 1970s this type of invest- ment in irrigation added more to the irrigated area than all public programs combined. The government's role in this development has been mainly to provide rural electrification and medium-term credit."5 Private tubewell development has so far occurred mostly in Haryana, Punjab, Tamil Nadu, and western Uttar Pradesh-regions in which conditions are highly favorable for the adoption of seed- fertilizer-water technology in general and groundwater development in particular. If private tubewell development is to maintain its his- torical rate of growth, the impetus must now come from other areas. One important geographical area that still has an untapped poten- tial for large-scale groundwater development is India's eastern re- gion, comprising Assam, Bihar, Orissa, eastern Uttar Pradesh, and West Bengal. But economic, climatic, and institutional conditions in the region provide relatively poor incentives for investment in tubewells. Agriculture is not highly productive and is based on tradi- tional varieties of grain, mainly rice. Because existing rice varieties are neither as widely adaptable nor as broadly pest-resistant as the new wheat varieties, private investment in tubewells may be limited. Fur- thermore, because the area has fairly heavy rainfall, the need for irri- gation is less obvious and the incremental benefits are few. The aver- age size of holdings is small, so even the smallest private tubewell is a viable investment for relatively few farmers. In addition, the public infrastructure is not well developed in the eastern region. The area is not strong in cooperative or commercial credit institutions; conse- quently, credit for tubewell financing may be available, but there are few local institutions to channel funds to small farmers who might use them. Moreover, rural electrification is much less developed in the eastern region than elsewhere.'6 A recent innovation in Bihar is a low-cost bamboo tubewell that irri- gates five to ten acres. A large number of these were sunk in the dis- tricts of Purnea and Saharsa and were reported to perform well for 134 Irrigation more than four years. These low-cost tubewells hold great promise for rapid exploitation of groundwater where shallow aquifers exist, but how extensively they can be used is not known (India, Ministry of Agriculture and Irrigation 1976, pt. 5). Electric pumps for them are generally trouble-free, cheap, and easy to operate, but the low voltage in many areas of India means that they have not been used as effectively as they might have been. Moreover, a disproportionately small percentage of small farmers have access to electricity. Diesel pumps can be conveniently moved from one tubewell to another and can be rented; farmers thus save on capital costs, but the hourly oper- ating costs are nearly double those of electric ones. Pakistan has similar problems. Although irrigation systems have ex- isted for centuries, recent developments have consisted mainly of large infrastructure projects designed to transfer water from western to eastern rivers. Some on-farm work has been undertaken to make water courses more efficient. As a result, farmers at the tail end of water courses (generally the smaller farmers) may now receive more reliable water supplies. In Bangladesh, before independence in 1971, the emphasis was on large-scale flood control and gravity irrigation projects. These were costly and represented a poor allocation of scarce resources. Since in- dependence the trend has been away from major infrastructure pro- grams and toward simple, low-cost, quick-yielding minor irrigation projects. These newer projects have provided mainly low-lift pumps and tubewells. Bangladesh has pioneered its own institutional arrangements for ir- rigation, and much of its groundwater has been reaching smaller farmers. Given the small size of most holdings in Bangladesh, it was essential to organize farmer groups to make the best use of minor irri- gation facilities, which typically cover ten to fifty acres. Although there are few large farmers in Bangladesh, the relatively better-off farmers tend to-dominate these groups. Evidence from the Northwest Tubewells project sponsored by the International Development Asso- ciation (IDA), an affiliate of the World Bank, confirms this, although it also shows that small farmers have benefited significantly. Recent World Bank projects have placed more emphasis on reach- ing small farmers. Under the Shallow Tubewells project, for example, if a farmer group is to be eligible for credit, "at least one-half of the proposed irrigated area must belong to farmers whose total holdings are less than three acres each." Even if programs with stipulations of this kind are fully successful, however, they cannot significantly re- duce the inequalities between various segments of rural society or solve at a stroke the problems of improving small farmers' access to irrigation. The relationship between access to groundwater and credit is well Prerequisites for Sustained Gains in Output on Small Farms 135 established in Bangladesh. Because of scale economies, smallholders must purchase tubewells as a group. The Thana Irrigation Program (TIP) is the only current example of this sort of arrangement in Ban- gladesh. The TIP begins with the formation of a pump group, which is encouraged to act as a cooperative and can expect to obtain credit to purchase tubewells. But according to Blair (1974), although a large proportion of the groups do become cooperatives, only a small pro- portion of the members obtain any loans. Possibly a few individuals take advantage of the program by padding their membership lists with nonexistent members. At least some of the groups are composed of farmers in a quasi-tenurial debt relationship to the group manager. Furthermore, because credit payments are made on the basis of irri- gated area, farmers have no incentive to use water carefully. Though farmers are expected to assume all operating costs of the pumps for a five-year period, their operations were initially subsidized; more- over, the distribution of pumps has become "a distribution of favors." The utilization of deep tubewells in Bangladesh was found to be positively related to the availability of cooperative credit and other in- puts. The deep tubewells were rented to groups by the Bangladesh Agricultural Development Corporation. Although the groups in- cluded members with farms of all sizes, leadership rested mainly with medium and large cultivators. Because of the fragmentation of own- ership holdings and lack of cohesiveness in the pump group, small farmers were not able to benefit. Myers (1973), quoting a study by A. H. Khan (1971) of a cooperative trying to take advantage of tubewell irrigation, reported that 24 percent of the cooperative's members received 62 percent of all loans. The larger landholders had greater indebtedness, and the larger the loan, the greater was the per- centage of loan default. Khan found that the share of loans taken by management committee members (who were large farmers) was dis- proportionately large-managers made up 36 percent of all borrow- ers and received 65 percent of all loans. To conclude, irrigation water is essential for raising small-farmer productivity, and the potential for expanding irrigation is substantial. Nevertheless, severe problems remain in ensuring that small farmers in the subcontinent will get a larger share of projected increases than in the past. Organizational arrangements designed to improve small farmers' access to both surface and groundwater irrigation are still in their infancy, and their success is far from assured. Access to Credit, Markets, Research, and Extension The sustained and substantial growth in small farmers' production of food grains depends critically on the increased use of purchased inputs-whether for multiple cropping or for the adoption of new 136 Access to Credit, Markets, Research, and Extenision high-yielding varieties (see the following section and chapter 5). Small farmers, however, typically have few if any financial resources for purchasing inputs. Hence they are in special need of agricultural credit. To what extent is credit in fact available to small farmers in South Asia, and does a lack of credit on reasonable terms seriously constrain their prospects for prosperity? The available data and some key policy issues relevant to the availability of credit are discussed below. It might be possible to acquire cash to buy inputs by selling market- able surpluses. I have already shown that small farmers do in fact par- ticipate in the market economy. But is this participation currently lim- ited by market distortions or other impediments, and can steps be taken to improve market access for small farmers? A brief overview of the evidence on small farmers' access to markets and related ques- tions for policy is presented below. Sources and Availability of Credit for Small Farmers Credit can be obtained from both institutional sources (public agen- cies or banks) and noninstitutional sources (moneylenders, commis- sion agents, friends, relations, and other private sources)."7 Institu- tional loans make up only a very small proportion of total agricultural credit (table 4-12), and they go to only 5-20 percent of all farmers. Noninstitutional sources provide 70-86 percent of the total volume of agricultural credit lending. In Bangladesh and Pakistan, all institu- tional loans come from public sources, while most noninstitutional credit is "noncommercial," that is, privately arranged through family or friends with interest rates and repayment terms that are hard to establish. In India, however, commercial credit predominates in the noninstitutional sector, and public institutions do not monopolize the institutional sector. Figure 4-3 shows the distribution of credit by country and by main source. Small farmers rely particularly heavily on noninstitutional and non- commercial sources of credit, especially rural moneylenders and rela- tives (Reserve Bank of India 1976). Moreover, although small farm- ers account for a large proportion of the number of loans from all sources-51 percent in India, 73 percent in Bangladesh, and 40 per- cent in Pakistan-they account for a smaller proportion of the total value of lending-41 percent in India, 50 percent in Bangladesh, and only 23 percent in Pakistan (see figure 4-4).18 And they pay much higher interest rates because they depend heavily on moneylenders, whose rates are two and a half to three times as high as those of insti- tutional sources (table 4-13). There is little evidence on the interest rates charged by friends and relatives, the other major source of Prerequisites for Sustained Gains in Output on Small Farms 137 Table 4-12. South Asia: Estimated Distribution of Agricultural Loans by Type of Lender Item Bangladesh India Pakistan Institutional loans Percentage of farmers getting 15 20 5 Percentage of all agricultural credit 14 30 14 Percentage of all institutional loans Public 100 87 100 Private 0 13 0 Noninstitutional loans Percentage of all agricultural credit 86 70 86 Percentage of all noninstitutional loans Commercial 39 71 27 Noncommercial 61 29 73 Sources: Donald (1976); World Bank data. credit for small farmers, but there is no reason to suppose that these rates are merely nominal. Opinions differ on whether access to credit is a problem for small farmers. The traditional view is that it is indeed a considerable prob- lem that hinders their adoption of new technologies. Abraham (1973) argues that special small-farmer credit agencies were set up in India largely to counter the bias of the institutional credit system against small farmers."9 Commercial banks have given small farmers better service since na- tionalization, but other institutional agencies remain strongly biased toward medium and large landholders. B. Sen (1974) points out that although credit from cooperative societies is supposedly available to large and small farmers alike, loans are granted in proportion to the share capital purchased. Consequently, even cooperatives set up to help small farmers end up discriminating against them. Small farm- ers' participation in the cooperatives' decisionmaking is limited; large farmers, with larger share purchases per acre, get more credit per acre from the cooperative movement. This problem is particularly serious because of the special role of cooperatives in fostering the spread of new technology for food-grain production. As Schluter (1973) pointed out, cooperative credit became important in the adop- tion of new technologies because traditional credit sources were un- able or unwilling to meet the needs of small farmers and because farmers were unable or unwilling to use these sources owing to the high risks involved in taking up the new technologies. By contrast, co- operatives' more flexible arrangements for repayment made farmers more willing to take risks. Figure 4-3. South Asia: Sources of Agricultural Credit as a Percentage of Rural Borrowings Bangladesh, 1969 Other (8 percent) n Government and Agricultural Development Bank (6 percent) ,- Cooperatives (8 percent) Friends, relatives (26 percent).. ~~~~~~ ~ Rua oneylenders (52 percent) India, 1971 Other (1.3 percent) F-- Government (3.6 percent) Friends, relatives - Cooperatives (22.7 percent) (18.8 percent) (- Banks (4.0 percent) Rural moeledr (49.6 percent) Pakistan, 1960 l Government (2.9 percent) -Cooperatives (13.2 percent) Rural moneylenders (20.7 percent) Friends, relatives (63.2 percent) Sources: Data for India are from the National Council of Applied Economic Research. Other data are from Tyers (1977); Surjit Bhalla in Berry and Cline (1979); the World Bank and the ILO; and Lowdermilk (1972). 138 Figure 4-4. South Asia: Percentage Distribution of Loans by Farm Size, 1971-72 Number of loans Farm size (acres) Value of loans , Jb No~~/ c0 v o o -1/ J1 / Firm size (number of employees) Source: Table 7-3. 246 Rural Industrialization and Employment 247 Table 7-4. India: Employment in Selected Industries in Punjab and Haryana Combined Annual percentage Nsumber of workers sr~ , m~~~~~icrease Industry 1969-70 1976-77 (1970-77) Khadi and village industriesa 25,642 43,290 7.8 Cycle and cycle parts 18,642 40,337 11.7 Agricultural implements and machine and hand tools 32,857 35,376 1.1 Rice shelling, cotton ginning and pressing 11,673 18,777 7.0 Sewing machines and parts 6,984 8,418 2.7 Hosiery 24,521 46,509 9.6 a. Includes processed food and vegetable oil, soap, gur-khandsari, pottery, handmade paper and textiles, and leather and leather goods. Sources: StatisticalAbstract of thePunjabl 1978, pp. 334-40; Statistical Abstract of Haryana 1977-78, pp. 170-73, 182-83. also striking. In urban areas in West Bengal the percentage of em- ployment in large units with more than 500 workers is very much higher than in the other states covered by the table. By contrast, the Punjab has a much smaller percentage of employment in very large units and a moderate amount in medium-size ones, while small units (fewer than 50 workers) provide 30 percent of total employment. Punjab alone among the states for which data are given has developed a strong segment of medium-size and small industrial firms produc- ing a variety of goods. West Bengal's industry, however, is concen- trated around Calcutta and in a few subsectors such as jute textiles and steel. These differences are repeated in rural areas, except that West Bengars bias toward large units is not repeated in the rural econ- omy. Bihar, Haryana, and Punjab also have heavy concentrations of employment in very small units. Industries have grown at very different rates. The nonmetallic mineral products subsector, which is dominated by relatively large units, has generally shown the most rapid rate of growth in rural ar- eas, and its growth has been at the expense of food processing indus- tries that generally operate with smaller units (Mazumdar 1980, p. 38). This is not true everywhere. Table 7-4 shows the growth rates of selected industries in Punjab and Haryana. In these states it appears that traditional village industries, which are dominated by smaller units, have not stagnated as in the rest of India. Meanwhile, some newer industries specializing in consumer goods and agricultural pro- cessing (often dominated by small and medium-size units) have also 248 Rural and Small-Scale Industrialization shown vigorous growth. These industries tend to have higher ratios of value added to employment and lower ratios of fixed capital to em- ployment than most of the heavy industries-chemicals, steel, and power-that dominate the industrial sector in other states.' The industrial experience of Punjab and Haryana is particularly relevant because these areas were considered industrially far behind West Bengal at the time of independence.6 Partition dealt a further blow to the region's industry, and a totally new start had to be made. Yet twenty-five years later the region was recording high rates of growth and its industrial output had increased more than fiftyfold. Between 1947 and 1974 the number of registered factories rose from 547 to 7,400; by 1973 registered small-scale industrial units had grown from some 18,000 to 49,000, and there was an equal number of unregistered units in these states. The region has emerged as an important producer of many industrial goods in the country, despite the fact that it lacks both major raw materials and the large markets that generally form the basis for industrial expansion. Table 7-5 shows the share of the Indian Punjab and Haryana in national output and employment in thirteen industries, for both large- and small-scale units. It demonstrates the dominant position of the region in the country's modernized small-scale sector of many industries-the two states' share of total employment and output in India exceeds 40 percent in more than half the industries listed. Even their share in the industrial output of the large-scale sector is high, given that Punjab and Haryana together account for only about 4.5 percent of India's population. Few other states that were industrially backward before indepen- dence have achieved anything approaching this record of industrial growth. Moreover, most of the region's industrial success is based on newly emerging industries in which it has no comparative advantage in the supply of raw materials. Markets, too, are mainly in other states or abroad.7 Pandit (1978) argues that the key features of this remark- able industrialization process have been the availability of a cadre of industrial entrepreneurs (mainly persons displaced by partition) with previous experience in trade, business, or industry and the rapid emergence of a large and disciplined skilled labor force. The state government played its part by allotting materials and factory sites, providing loans for construction and machinery, training and orga- nizing industrial institutes, and giving special allotment quotas to help displaced persons. Displaced agricultural classes also spearheaded the Green Revolution, which suggests the importance of migrants-or re- migrants, in the case of the Punjab, because their ancestors had gone to settle the new canal colonies in West Pakistan only two generations before partition-in pioneering entrepreneurial activity. Table 7-5. India: Share of Punjab and Haryana in All Output and Employment for Selected Industries and Sectors Percentage of all India Small-scale sector (1972)b Large-scale sector (1969)C Industrya Output Employment Output Employment Woolen hosiery and other knitted wear 88.9 92.7d 28.8d 13 Id Rubber footwear 68.3 56.1 13.9 8.8 Small hand and cutting tools 50.2 43.0 15.5 18.5 Plumbing fixtures and metal fittings 18.9 18.5 63.2d 62.4 Nuts, bolts, and other minor hardware 36.2 36.8 3.8 5.4 Agricultural implements and hand tools 36.6 29.6 18.2 15.6 Utensils and light metal products 13.2 9.2 4.3e 4.3C Machine tools 42.5 44.4 9.6 11.0 Sewing and knitting machines 68.5 73.0 17.5 15.3d Automobile parts and machinery 17.0 18.8 14.4 16.8 Bicycles, motorcycles, and parts 59.3 55.8 30.6 28.7 Laboratory and scientific instruments 26.9 27.5 13.7e 21.3e Sports goods 48.3 47.6 78.7 57.2e All industries 13.2 10.4 4.4 3.9 a. Industries are grouped according to definitions standardized in the Annual Survey of Industries. b. The small-scale sector is defined by "Census of Small-Scale Industrial Units" of the Indian Ministry of Industry. c. The large-scale sector is composed of units employing 50 or more workers and using power or 100 or more workers without power. d. Refers to Punjab only. e. Refers to Haryana only. Source: Pandit (1978), p. 1936. 250 Rural and Small-Scale Industrialization Skilled workers make up more than two-thirds of the total number employed in most of the industries which the region dominates; this is true even in the small-scale sector. This labor force is drawn mainly from urban centers in which skilled refugees from Pakistan had set- tled. Because they and other migrants from rural areas have stressed the importance of education for their children and have acquired skills through learning-by-doing, the efficiency of this labor force has increased in the postindependence decades. Skilled labor with an entrepreneurial spirit and experience of ad- versity together with enlightened state policies have transformed this region through small-scale industrialization in a way unparalleled in the subcontinent. The region has been able to absorb the growth of its own labor force and in-migration from other regions attracted by rising real wages in agricultural and other occupations. The early de- velopment of industry, initially rural, small-scale, and modest, has promoted the formation of both skills and finance capital that should ensure rapid future growth. The region's lesson for other areas seems to be that a rapid expan- sion of small-scale industries, both as direct producers of consumer goods and as ancillaries to larger firms, is a highly effective way to expand employment. In general, however, evidence on the industrial structure in India and on the linkages discussed earlier suggests that rurally based strategies for growth have been exploited on only a very limited scale in South Asia. This has been a major failure of rural de- velopment efforts generally in the subcontinent. Lessons from Taiwan That a strategy based on rural small-scale industries can be made to work is evident not only from experience in the Indian Punjab but also from developments in several Southeast Asian economies, includ- ing China, the Republic of Korea, Malaysia, and Taiwan. The experi- ence in China and Taiwan is especially instructive. Ho (1979) has shown how in Taiwan nonfarm sources of income increased from 25 percent of total farm household income in 1962 to 43 percent in 1975. Although total rural employment grew at a slower rate than urban employment, employment in many rural in- dustries grew more rapidly (manufacturing employment in rural Tai- wan grew at 7.2 percent between 1956 and 1966) and led to a large shift toward nonagricultural ruraljobs. In 1956 agriculture employed 73 percent of the rural labor force. By 1966 this share had fallen to 54 percent, with the rest employed in manufacturing, commerce, and services. Further industrialization has not been accompanied by a Rural Industrialization and Employment 251 concentration of nonagricultural employment in major metropolitan areas. In rural areas, small-scale units with low capital intensity predomi- nate. Establishments with fewer than ten workers employed 41 per- cent of the manufacturing workers in rural Taiwan. Fixed assets per person employed in rural areas were half those in urban areas,8 the output-capital ratios were higher in some industries, and the ratio of net value added to fixed assets did not differ much from those in urban Taiwan. These facts suggest that rural-based industries used capital as efficiently as, if not better than, urban-based industries. The impact on rural employment was particularly significant. With the labor force growing by over 2.5 percent and the size of holdings averaging about one hectare (Fei, Ranis, and Kuo 1979), increasing nonfarm employment enabled farm households to combine farming with part-time or full-time (for some members of the household) em- ployment, which eased the pressure on land and on rural-urban mi- gration. In 1960, 45 percent of all farm households relied exclusively on farms while 23 percent earned more income from nonfarm activi- ties. By 1970 these ratios stood at 30 percent and 29 percent respec- tively. The share of small farmers was even higher, and this contrib- uted to the greater equality of total incomes in rural areas. Nonfarm activities grew by 12 percent a year between 1960 and 1970. The real income of farm households doubled between 1952 and 1972, mainly as a result of this growth, although rising agricultural productivity also contributed. This was accompanied by an upgrading of skills of the rural population via almost universal education and the develop- ment of rural infrastructure. The geographical size of Taiwan has certainly contributed to the decentralization of its industries in rural areas. But it is clear that rapid and equitable agricultural development (based initially on a very successful land reform) has also contributed by providing an im- portant market for nonfarm goods and services. Practically all of the services and many nonfarm consumer goods demanded by farm households are produced in rural areas. Many of these have relatively high income elasticities, and their production in small-scale, capital- efficient units with low technology and skill requirements has created rural nonfarm employment. Rural Industry in China The experience in China is of special relevance because of character- istics it shares with the countries of South Asia-notably size of land area and population. Rural linkages and nonfarm activities are 252 Rural and Small-Scale Industrialization strongly emphasized in China. Although noncrop agricultural activi- ties have always been important-especially pig production-they have been accelerated and integrated since the 1970s at both the pri- vate and the communal level.9 Nonagricultural rural employment, for example, rose from about 15 percent of the labor force in 1957 to over 24 percent by 1975. More than 100 million people worked in China's nonfarm rural sec- tor in 1975, reflecting a growth rate of over 7 percent a year between 1957 and 1975. The expansion was most marked in collective- communal industries in rural areas. Trade, food processing, personal and financial services, health, education, transport, and communica- tions all showed above average rates of growth (Rawski 1979). Social services-health, education, and population control-have been ac- tively encouraged and have grown rapidly. One unique feature of rural development in China has been the promotion of rural small-scale industries designed to meet local pro- duction (intermediate) and consumption (final) demand locally. A second unique feature has been the integration of these activities with agricultural seasons; in the slack period much of the labor force is employed in noncrop and nonagricultural rural activities. Communes and "brigades" ran small-scale enterprises organized on this basis. In the 1980s these have been largely decollectivized through the use of the personal responsibility system (PRS). These enterprises almost doubled in number and in the value of their output between 1975 and 1978. These activities were expected to contribute about half the total value of communes' output by 1985 (Perkins and Yusuf 1984). Small-scale rural industries in China include iron and steel, cement, and chemical fertilizers (over half the total output of these industries was produced in small plants in 1973),1° together with local enter- prises that manufacture and repair farm machinery, process agri- cultural output, and produce consumer goods. Most of the durable consumer goods available in rural areas, including bicycles, radios, furniture, and textiles, are also produced in small-scale units linked to communes-the result of a conscious policy "to use local materials, produce locally and distribute locally" (Sigurdson 1972, p. 324). The Chinese have also gone to unusual lengths to promote indige- nous intermediate technologies in support of the official strategy. Lo- cally developed methods, equipment, and adaptive innovations are encouraged (Riskin 1978). As a direct consequence, in addition to the considerable intensification of agriculture, the vast rural economy not only absorbed nearly 100 million new entrants into the rural work force between 1957 and 1975, but also raised the average number of days of employment for the entire farm labor force of more than 300 million men and women (Rawski 1979, p. 126 and ch. 4). Rural Industrialization and Employment 253 The development of China's rural industry before the late 1970s has since been dwarfed by a burst of growth." With the breakdown of the communes and the onset of rural reforms in 1978, the farm sector was transformed by the personal responsibility system, which was introduced in 1978 but fully implemented in 1980-83. Under the PRS land can be leased for private farming by rural households for up to fifteen years (up to fifty years in mountainous regions), and leases can be transferred to heirs. The result has been a tremendous surge in agricultural and rural economic activity. Further changes in government policies have progressively liberalized the growth of pri- vate enterprises and permitted the establishment of a variety of new enterprises in rural areas. The labor surplus, which was extensive but hidden under the old commune system, suddenly became open and visible with the growth of private family farming. This in turn has generated pressure in rural areas to use the large labor surplus effec- tively. Between 1978 and 1987 the value of gross agricultural output grew at 6.8 percent a year in real terms, the gross value of industrial output in rural areas grew at more than 23 percent a year in real terms, while rural employment in industry, commerce, transport, and construc- tion increased by 124 percent. This tremendous growth of rural in- dustrial and other activities has been accompanied by a large increase in personal incomes and in the demand for rural housing and con- sumer goods. As a result, millions of new township and village enter- prises (TVEs) have emerged to meet this demand. Large increases in rural credit have been provided to the TVES by the expanding banking system, and internal accumulation of funds has been generated in part by concessional tax treatment. Rapid growth in incomes in the rest of the economy (at rates of more than 10 percent a year) together with high investment demand (gross investment to GDP ratios in China in the 1980s have exceeded 30 percent a year) have helped sup- port this expansion. The growth in the TVE sector has also brought about massive shifts in the rural labor force. In 1978 about 11 percent of the total rural labor force of around 306 million was in nonfarm activities. By 1986 this share had grown to about 20 percent of a total rural labor force of 380 million. The nature of rural industry has also changed. Previ- ously many of the TVES were geared to processing and sideline activi- ties that supplemented incomes in the communes in the off-seasons. With few links to the larger economy, they were inward oriented and static, offering no competition to the state-owned urban-industrial sector. Now TVES are characterized by dynamism, rapid growth, and an ability to adjust quickly to changing market conditions; they compete effectively in many product areas with state-owned enter- 254 Rural and Small-Scale Industrialization prises and reach out to local and even provincial product markets. A continuing characteristic of Chinese rural industry is its strong ties to the local community. The institutional structure of China's TVE sector is linked very closely to a three-tier system of rural community governance-township, village, and production team-that has been in place since the 1950s. Community governments at the township and village levels resemble mini-states in that they have fixed mem- bership, often share incomes within the community, and exhibit a large degree of financial self-reliance. Many of them continue to be communally owned, but at the same time seem to operate as profit- oriented business entities. Agroindustries in other countries, especially in South Asia, often suffer from an inability to develop strong farm linkages with private peasant suppliers of raw materials, but this problem does not seem to be evident in China. Although forms of individual ownership and proprietorship are emerging, it is still the strong ties to local govern- ment and communities that link the Chinese TVES with the farm sector and underlie their success. These strong community links are a cul- tural trait that is shared with Japan, Korea, and Taiwan, where rural industrial activities have also been flourishing. Whether these cultural norms can be transferred to the South Asian context is debatable. rvEs in China are not without their problems. They suffer from limited factor mobility in a socialist setting where labor has never been mobile, face increasingly severe competition and market saturation in many industries, are hindered by lack of scale economies and by ex- cessive spatial dispersion, lack support services (credit and marketing networks, transport, and communication), and are denied many es- sential materials distributed by the state-controlled system. But above all they are still in a sort of legal limbo: their rapid growth has caught the government off guard, and it is not exactly sure how to deal with the expansion of this "private" sector in a socialist setting. The recent mushrooming of TVES in China is further evidence, if more were needed, of the impetus that the growth of agricultural out- put provides for growth of the rural nonfarm sector. South Asian countries generally lack this impetus and have not been able to benefit to the same extent from rapid growth in rural industry. This is partic- ularly disturbing because, given the population densities in South Asia and China, growth in industrial employment cannot be expected to keep up with the growth in the labor force, even with high rates of growth in the industrial sector. In this regard too, China has outpaced India. Whereas industry in China grew at around 10 per- cent a year all through the 1950s and 1960s, India's industrial growth stagnated around 5 percent a year. Between 1973 and 1984 the indus- trial sector grew 8.7 percent a year in China but only 4.4 percent in Rural Industrialization and Employment 255 India. Even with such high rates of industrial growth, most of the em- ployment growth in China is expected to be in the TVE sector. The same will have to be the case in South Asia. But rapid employment growth has been possible only where agricultural growth has been ro- bust (Northwest India and Punjab). For further growth, it will be nec- essary to have rapid industrial growth in the urban sector as well. Bet- ter links between rural and urban industrialization will have to be developed. China and Taiwan both point the way in this context. Another important element of Chinese strategy has been their con- scious reliance on the growth of the TVE sector to prevent rapid urban growth and to meet most of the growing employment needs in rural areas. Earlier this was done mainly by restricting any labor movement whatsoever. Now the reliance is on rapid growth of rural industrial employment. In South Asia the absence of any rational strategy to provide employment in rural industries has led to rapid urbanization with all the associated costs of heavy social overhead and urban blight. Disguised rural unemployment and poverty have too often been con- verted into open urban unemployment and despair through lack of growing opportunities in rural areas. To its credit, China has so far avoided this. It is a lesson well worth learning in South Asia. Comparison of India and China The different approaches to rural industrialization taken in India and China have been examined by Sigurdson (1978). He points out that both countries recognize the need to develop nonfarming activities through rural industrialization and both have some 20 million people engaged in rural industrial activities. Nevertheless, as he shows, the organization and structure of rural industries in the two countries ex- hibit major differences-differences that offer some helpful insights into what can be learned from the Chinese experience. Industrial activity in India's rural sector is dominated by traditional, household-based artisanal handicrafts, together with a modest but growing nonhousehold sector; China's rural industries are run mostly by collectively owned units. This denotes not only a difference in ownership-private or collective-but also differences in size, tech- nology, skill levels, productivity, and the consequent ability to gener- ate investable capital. Some of the results of these contrasting ap- proaches are briefly outlined in the next few paragraphs. First, India has tried to preserve village and artisanal activities based on Gandhian traditions. To this end, it has imposed quotas, differen- tial pricing, preferential treatment for credit and raw materials, and a "reserved" listing of goods that can be produced only by this sector. As a result, primitive (and not necessarily appropriate) technologies 256 Rural and Small-Scale Industrialization involving limited use of machinery and power have been preserved at the cost of more efficient alternatives. It has also meant that new skills have not been learned, new occupations have been discouraged, and low labor productivity has been institutionalized. The integration of these household units with other industrial processes is minimal; they have consequently not been exposed to any modernizing influ- ence. Attempts to promote more modern, technology-based rural in- dustries in semirural towns via the Rural Industries Projects (RIP) have had few results. In China, by contrast, under the communes, the collectivization of household and village industrial and artisanal activities meant that the 20 million people in the rural industrial sector in 1973 were employed in modern or semimodern industrial activities as part of an overall program for rural modernization. The total number of nonagricul- tural TVE employees, including part-time and seasonal workers, rose from 22 million in 1978 to 77 million in 1986, an increase of 55 mil- lion. Levels of technical development and skills were consciously up- graded, and access to modern tools, machinery, and power were en- couraged. The long-term benefits of skill formation have outweighed the costs of production equipment. This systematic organization of rural industrial activities continues under the personal responsibility system. The basic approaches underlying rural industrial programs are thus quite different in the two countries. China emphasizes upgrad- ing village crafts by transforming their technologies, while India pre- serves outmoded and stagnant production methods for their own sake, regardless of their high opportunity costs when compared with more appropriate systems. As a result, China operates with larger units and higher capital intensity, but also with higher labor produc- tivity and greater diversification of output. Second, Indian programs stress the labor-absorbing aspects of rural industrialization-in effect, treating it as a residual source of jobs that agriculture has failed to provide. China, however, has paid only mar- ginal attention to using industrial employment in rural areas as a means of righting the overall imbalance between labor supply and de- mand in these areas. It has instead concentrated on dovetailing indus- trial activities with seasonal labor needs in farm production and has used massive rural works to mop up rural labor. The Chinese have thus not been the victims of the obsession to use labor at any cost that has prevented Indian rural industries from choosing more pro- ductive and efficient techniques. Rural industrialization has rightly been seen not as providing employment opportunities for certain rural groups-often precisely those who are least able to take advan- tage of them-but rather as an integrated function of a moderniza- Rural Industrialization and Employment 257 tion program designed to upgrade the rural economy as a whole. Third, rural industries in India are considered in isolation both from urban industries and from programs for regional rural develop- ment. Rural industrialization is organized in an economic vacuum. Vertical integration and linkages with urban industries through sub- contracting is missing; there is no overall plan for meeting local needs through local production. Farm machinery and fertilizers come from distant urban centers or from abroad. Only primary food processing is located anywhere near rural areas. Neither intermediate nor final local demand is systematically met by local production using local ma- terials. In China rural industries are part of both an overall industrial strat- egy and an integrated rural strategy based on the TVE system. Not only is considerable attention paid to producing for local needs by local units using local raw materials and skills, but also the interde- pendence of units is planned for and promoted through links within counties, between counties, and with higher-level enterprises. Large numbers of people are mobilized to participate in overall rural devel- opment through regional planning that is dovetailed with local partic- ipation down to the brigade level. Rural industrialization is thus seen as only one element of China's strategy for rural development. A number of integrated components form a "package," and it may be difficult or impossible to hive off individual parts (see Aziz 1978). In both China and India, population pressures on land are intense, the rural labor force is growing rapidly, and the economic welfare of farm and landless households depends critically on whether they have access to rapidly expanding sources of nonfarm employment. The Chinese model suggests that this access can be provided by creat- ing and developing a process of spatially decentralized industrializa- tion, based on small-scale rural production units that are linked to the intermediate and final demand generated by a rapidly growing agri- cultural sector. All these activities need not be planned by the state, as in China, but a system of incentives must be devised to achieve simi- lar results. Rural industrialization can thus serve as a base for econo- mywide industrialization. Conclusions The experience of other countries suggests that the broad elements of a rural development strategy for South Asia include: * An accelerated expansion of food-grain production, based as much as possible on labor-intensive methods and small and mar- ginal farms 258 Rural and Small-Scale Industrialization * Massive public investment in irrigation, transportation, commu- nications, electrification, feeder roads, and other infrastructure in rural areas * A change in the structure of industrial production toward con- sumer goods whose quality, price, and product mix are specifi- cally designed to reflect the demand structure of the rural popu- lation * A rapid growth in ancillary industries-both those ancillary to agriculture (dairying, poultry, fishing, forestry, and food process- ing) and those ancillary to urban medium- and large-scale indus- tries (through subcontracting for and vertical integration with those industries)-together with the expansion of institutions to provide the inputs, credit, and marketing and extension services needed to support these industries' growth * A conscious policy of encouraging, in rural areas and provincial towns, productive small-scale enterprises that use local materials and develop local skills whenever this is cost-effective * A shift of emphasis away from household-based, low-productivity village handicraft and artisan units, and toward the rapid devel- opment of units that, although still small in scale, use machinery, power, and new skills to raise labor productivity * As in the case of ancillary industries, the development of support- ing institutions to deliver inputs, credit, and marketing and exten- sion services to these small-scale "modern" enterprises * The development of appropriate price, tax, and credit incentives and assured access to needed raw materials * The delivery of basic social services such as health and education, not only to improve the quality of rural life but also more specifi- cally to upgrade skills and labor productivity in rural areas * The promotion of a new attitude toward rural industries that de- fines them as a dynamic part of an integrated strategy to trans- form rural areas, rather than as a "make-work" means of absorb- ing labor and providing residual employment for the rural poor who cannot find agricultural work. Such a strategy would raise the incomes of peasant cultivators and small-scale rural entrepreneurs and would thus initially worsen in- come distribution. At the same time, however, the living standards of the Tural poor would improve as a result of lower food-grain prices and greater employment opportunities. The organizational arrangements (communal or private) and the means (central planning or a system of market incentives) of imple- menting the strategy are probably less important than the establish- Rural Industrialization and Employment 259 ment of an environment in which the poor have increasing access to resources and opportunities. Experience from other countries in Asia suggests that redistribution of land may be a necessary (but by no means sufficient) condition for such an environment. Both equity and growth can be served if rising incomes and their induced effects can be widely and equitably distributed. The demand structures associated with rising rural incomes can then support a de- centralized and labor-using pattern of industrialization. Growth and equity conflict when the distribution of rural incomes is highly skewed and incremental incomes are inequitably distributed. Such a situation produces a narrowly based, urbanized process of industrialization that caters to the needs of the few who are relatively rich and neglects the needs of the many who are poor. It needs to be understood that the long-run answer to the problems of rural unemployment and underemployment must be sought not in the rapid growth of agriculture alone, but in rapid and appropriate industrialization, in both rural and urban areas, that proceeds hand in hand with rising agricultural output. Whatever the long-term pros- pects for the expansion of rural employment and incomes in South Asia, however, given past and reasonably predictable future rates and types of growth in the rural economy, the problem of rural unem- ployment will not be solved quickly. Indeed, rural unemployment can be expected to increase further in the near and medium term, under even the most optimistic scenarios for accelerated agricultural growth, at least in Bangladesh and most of India. Therefore a case for medium-term relief to the unemployed becomes unassailable. Policymakers have been forced to rely on large-scale public works programs, both to create additional employment in rural areas and to construct some of the physical assets needed for the dynamic ex- pansion of the rural economy in the long term. Experience with pro- grams of this kind in the subcontinent is discussed in the next section. Rural Public Works Programs Examples of public works programs in South Asia include three pro- grams in Bangladesh (the Works Program, the Thana Irrigation Pro- gram, and the Integrated Rural Development Program), two in Paki- stan (the Rural Works Program and the People's Works Program), and three in India (the Scarcity Relief Program, the Crash Scheme for Rural Employment, and the Drought Prone Areas Program). A great deal of controversy exists about the usefulness of such programs in alleviating rural poverty and unemployment. This section summa- rizes the findings of a detailed World Bank study of experience with these programs in the 1960s and 1970s (Burki and others 1976) and 260 Rural Public Works Programs also examines the Maharashtra Employment Guarantee Scheme (EGS), which has adopted a novel, comprehensive, and apparently suc- cessful approach to public works (Thomas and Hook 1977). Main Characteristics Table 7-6 summarizes some of the characteristics of selected public works programs in South Asia. The programs have generally been geared to provide employment to specific target groups such as the rural poor, the "most needy," or "poor surplus agricultural labor," but in practice their targeting has been neither very specific nor very effective. Although all the programs have promoted long-term employment and the development of infrastructure, they have been designed pri- marily to provide relatively short-term disaster relief or income sup- plements. Moreover, they have tended to be planned and imple- mented in isolation rather than as part of any broad strategy to create new rural assets in an efficient way or to attack the fundamentals of the unemployment problem. The physical, budgetary, and political constraints inherent in these programs have made it impossible for them to make an appreciable difference. They cannot employ all those without jobs or provide supplemental sources of income to all those who need them. In Bangladesh a number of important and innovative programs provide employment opportunities for the poor in construction, while creating economically justified infrastructure in rural areas. By far the largest of these is the Food for Work Program (FWP), which started in 1974-75. The allocation of food to the FWP increased al- most twelvefold in the following decade to reach 500,000 tons in 1986, equivalent to about US$114 million (IFPRI 1985). The FwP now generates annually nearly 100 million days of employment for work- ers who would otherwise be unemployed and at least 17 days of addi- tional employment for every landless worker in Bangladesh in con- struction alone. Recent evaluation studies suggest that if FWP projects are well selected, designed, implemented, and maintained, the em- ployment effects will be enormous. A recent review by the World Food Program, however, has criticized the program's lack of integra- tion with local development schemes such as those financed by the Upazila Development Grant. It also identifies technical and manage- rial problems, the underpayment of beneficiaries, and the overestima- tion of work done. Nevertheless, those who found jobs with public works projects considerably improved their short-term income and consumption. The percentage of GDP represented by the programs' expenditures Table 7-6. South Asia: Characteristics of Selected Public Works Programs to Augment Income India Pakistan, Bangladesh, Crash Scheme Drought Prone Rural Works Item Works Program for Rural Employment Areas Program Program Duration 1962 to present 1971-74 1970 to 1962-72 present Nature of unemployment Seasonal; estimated at 30 Seasonal; estimated at Underemployment percent of available 40 percent of rural estimated at workdays labor force 27 percent Target group Rural poor Most needy None Poor surplus agricul- tural labor ' Population per square kilometer, 456 153 51 1965 Workdays of employment generated (millions) 28.8 178.0 37.9 4.4 Average annual program expenditure of GDP 0.57 0.15 0.05 0.19 Wages as a percentage of total program expenditure 63 (1962-67); 16 (1971-78) 76 50 30 Note: The main objective of income-augmenting programs is to supplement the normal earning activities of those they employ, usually in rural areas. They generally take account of seasonal variation in employment. This objective distinguishes them from disaster relief works designed to replace earning opportunities destroyed or reduced by natural calamity and from long-term employment programs designed to absorb structural unemployment. Source: Burki and others (1976). 262 Rural Public Works Programs has varied, but has generally been low (Burki and others 1976). Since their overall inflationary impact has consequently been negligible, lit- tle can be concluded about how the price of wage goods might be af- fected by an expanded program designed to reduce rural unemploy- ment rates significantly. Inflationary pressures on food prices have occurred, however, especially where crisis relief programs have con- centrated on a small geographical area. This was so in Maharashtra in 1973-74, for example, when the Scarcity Relief Program gener- ated an estimated 290 million days of labor at a time when successive crop failures had already produced strong upward pressures on ce- real prices. The need for income-augmenting rural works programs varies in- versely with the demand for agricultural labor. If they are to reach those who need additional employment, the programs must be based on a detailed knowledge of local on-farm and off-farm seasonal em- ployment patterns. In this regard, most of the programs effectively coordinated public works activities with seasonal needs and con- cluded, deferred, or interrupted projects to avoid conflicts. Programs which offer year-round jobs can present serious problems, however, especially if they pay high wages; they may outbid the agricultural sector for peak-season labor and create a shortage of workers in agriculture. The detailed evidence on alternative employment patterns and wage rates that would make it possible to calculate the real opportu- nity cost of the programs was not available. But the wage rates obtain- able under, for example, the Pakistan Rural Works Program seem high enough to run the risk of diverting rural labor from other activi- ties. The extent to which this could occur in practice depends partly on how much emphasis a particular program places on employment generation to the exclusion of other outcomes. One measure of the strength of this emphasis is the percentage of total expenditures going to wages (see the last column in table 7-6). Both programs in Pakistan were classified in the low employment-creating category, while portions of the Bangladesh Work Program (1962-67) were judged to have generated large amounts of employment. Both the Crash Scheme for Rural Employment and the Drought Prone Areas Program in India are considered medium-to-high employment- creating programs (see Thomas and Hook 1977). Most of the programs in South Asia have concentrated on directly productive projects and the provision of physical (as opposed to so- cial) infrastructure. This bias effectively counters fears that public works programs do not create useful rural assets. According to Burki and others (1976), however, only India's Drought Prone Areas Pro- Rural Industrialization and Emnployment 263 gram spent substantial sums on creating assets that could be expected to have a direct impact on agricultural growth. In many cases, ex- penditure is thought to have emphasized road construction in rural areas, an activity that has traditionally been well organized under ex- isting administrative structures in South Asia. Efforts of this kind often concentrate on the so-called construction phase of a project to the exclusion of its long-term impact on unemployment or poverty. Wages paid purely for this kind of work are unlikely to have any last- ing effect on poverty or to cause any long-term changes in the posi- tion of those employed or in the distribution of income in rural areas. In the construction phase, the programs transfer resources to those believed to be in deepest poverty through the wages paid, but their secondary and tertiary benefits depend on the type of works being constructed. Projects for the development of irrigation, drainage, or soil conservation, for example, have the greatest impact on future production and thus the highest potential for creating additional de- mand for labor. By and large, however, they help those who already own land or assets, and this help is directly proportionate to the amount of land or assets owned. Thus, where land distribution is bi- modal, as in Pakistan, secondary benefits are regressive-that is, they accrue mainly to the better-off and accentuate inequities in the own- ership of rural assets. Where land distribution is more even, as in Bangladesh, the benefits may be more evenly shared but are still dis- tributed among landowners only; the landless remain excluded from all but the primary wage benefits. An illustrative case analyzed for Bangladesh shows that the returns to land (and hence to landowners) claimed by public works projects are five times higher than the re- turns to labor (Burki and others 1976, p. 50). It is of course possible that the programs provide work for some very small owner- cultivators. Nevertheless, rural income distribution is unlikely to im- prove much if some short-term increase in employment is the only benefit this group receives. Projects that emphasize economic infrastructure (roads, reforesta- tion, markets) tend to have the largest short-run redistributive effects because they are usually located on public land. But again longer- term benefits will accrue to landowners in proportion to the size of their holdings and, more important, the size of their marketable sur- pluses. Traders and intermediaries providing farmers with inputs or buying their output will also gain. The only nonwage benefits of in- frastructure programs (notably roads) for the landless may lie in pro- viding greater labor mobility and opening up remote areas-devel- opments which may lead to better employment opportunities. Social infrastructure projects (such as schools, clinics, and low-cost 264 Rural Public Works Programs housing) can provide broad benefits, although they may not reach the poorest groups if access to these facilities is limited. The rural power structure can often restrict the effects of such programs. Thus gains from public works programs (other than short-term employment) may percolate through to smallholders and the landless only if land is reasonably evenly distributed or if small holdings are geographically segregated from larger ones. In South Asia, where small and large holdings are intermixed, it is almost impossible to di- rect the longer-term benefits of public works programs to small farm- ers or the landless. The existing distribution of scarce resources, espe- cially of land, is the single most important determinant of the flows of nonwage benefits from works programs. Both the design and the implementation of most public works pro- grams in South Asia have been subject to political influence. The two most important factors in bringing about what Burki and others (1976) call "program mutations" are (a) political commitments and objectives that are rarely articulated explicitly, and (b) local elites that bend public works programs to serve their own interests. Pressures can be applied at key decisionmaking stages, including the choice of subprojects and of the location of the program as a whole; the choice of technology, especially its labor intensity; the choice of agents for project implementation; the establishment of wage rates; and the se- lection of employees. Depending on the choices made, benefits can be diverted to suit the interests of (usually) higher-income groups. Program mutation has been especially evident in Pakistan, with its sharply skewed distribution of land assets and political power. In both the People's Works Program and the Rural Works Program, political loyalty rather than project performance became the basis for public works allocations. "Large land owners, rural contractors and small town tradesmen quickly shaped the program to serve their own inter- ests" (Thomas, in Falcon and Papanek 1971, p. 306) by shifting decisionmaking from the lowest tier of government, the union coun- cils, to the much larger district councils, which were controlled by these interests. Work was given to contractors, which led to a rapid decline in labor intensity and employment creation. In Bangladesh a more egalitarian rural structure appears to have slowed the process of mutation, the consequent emergence of power brokers, and the dilution of broad participation in project decisions and benefits. In addition, the Works Program in Bangladesh has ben- efited from giving local bodies the authority to make decisions and to implement programs within a framework structured by the central government. To summarize, overall experience with rural works programs in the Rural Industrialization and Employment 265 1960s and early 1970s was unimpressive, and much of the skepticism about their effectiveness was justified. Maharashtra's Employment Guarantee Scheme The Employment Guarantee Scheme (EGS), begun by the Ma- harashtra state government in 1972, seems to provide an alterna- tive to the earlier rural works programs."2 The EGS offers guaranteed employment to the rural poor, including small and marginal farmers, landless agricultural workers, and rural artisans. In ways that are pro- ductive to the economy, it aims to provide gainful employment to all unskilled persons in rural areas who are willing to do manual labor but cannot find work. In a country as large as India with widespread rural unemployment and underemployment, such a program clearly has enormous appeal. The main features of the scheme are: * Beneficiaries cannot choose the work for which they will be em- ployed. * Guarantees are given that work will be available within a radius of five kilometers of the village where potential beneficiaries have registered. If employment is more than five kilometers away, ar- rangements for camping (including water, sanitation, and medi- cal and shopping facilities) have to be made. * Work has be be provided within fifteen days of a request for it. * Work has to be provided for a continuous period of not less than fifteen to thirty days. It is not clear whether workers can repeat- edly reregister after having worked thirty days, but presumably this is so. -\Works are undertaken under government auspices, not through contractors, and at least 60 percent of total expenditures must be for wages to laborers. (This provision is designed to meet the criti- cism that public works programs have often spent their money through contractors and have provided only limited employment and wage benefits to participants.) * Wages are determined by piece rates calculated to ensure that, as far as possible, a laborer working seven hours a day gets the mini- mum agricultural wage, or at least Rs3 a day. Wages are paid weekly, on the basis of output or work done. * Works are chosen according to a particular set of priorities, such as the completion of earlier works or the continuation of labor- intensive activities that directly or indirectly stimulate agricultural 266 Rural Public Works Programs production. Works may include minor irrigation infrastructure, water and soil conservation, command area development, land development, or flood control. Road construction is undertaken where no other productive activities are available. * As far as possible, on-farm and related labor needs are to be met before the EGS becomes active in any area. The scheme has some unique administrative features. First, since pri- ority is given to existing or incomplete projects, only if no such proj- ects are available can new ones be started. Second, to the extent possi- ble, each project is intended to be completed within one or two work seasons, and any requiring more than 4,000 workdays to complete need prior authorization from the state government. Third, each project must have a blueprint that shows how work is to be allocated over each two-week work period, and a manpower budget showing labor demand and supply must be prepared at the block level. Fourth, a register of work seekers is used to forward requests for work to the block level where the program is coordinated by the district collector who is in overall charge of the EGS. Fifth, half of the budget allocation for the EGS comes from the state's regular budget and the other half is raised by special taxation. Sixth, local administrative units enjoy wide discretionary powers; each district has its own quarterly budget for managing the scheme. Table 7-7 shows the expenditures of the EGS and workdays of em- ployment generated. Its employment impact has been massive, even when compared with the nationwide Crash Scheme for Rural Em- ployment (which generated only 178 million workdays over a five- year period). The World Bank estimated in 1978 that rural unem- ployment and underemployment in Maharashtra amounted to roughly 620 million workdays a year. The scheme was intended to provide about 140 million workdays per year, a figure estimated to have been surpassed by 1977-78. At its planned level the EGS was thus expected to generate work for nearly a quarter of the rural unem- ployed and underemployed; if the projections in table 7-7 were fully realized, this proportion would have risen to nearly 35 percent.13 The wage component of expenditures has typically been much higher than the levels previously experienced in public works programs in South Asia, attesting to the labor intensity and employment emphasis of the scheme. Wage rates actually paid have been at or above planned levels. The EGS was supposed to supplement existing projects already in- corporated in state plans, but all evidence indicates that it has in fact been an initiator of new projects. More than 5,000 individual projects Rural Industrialization and Employment 267 Table 7-7. India: Expenditure and Employment in Maharashtra's Employment Guarantee Scheme Annual Days of expenditure employment Wage Average (millions generated component daily wage' Year of rupees) (millions) (percent) (rupees) 1972-73 18.8 4.5 n.a. 4.18 1973-74 18.9 5.1 n.a. 3.71 1974-75 137.2 48.1 99 2.85 1975-76 344.5 109.5 86 3.14 1976-77 485.4 133.0 77 3.64 1977-78 (est.) 560.0 160.0 75 3.50 1978-79b 738.6 188.0 70 3.93 1979 - 806 867.8 220.9 70 3.93 1980 - 816 1,100.0 260.0 65 4.23 1981 - 826 1,292.5 305.5 65 4.23 1982-83b 1,518.7 359.0 65 4.23 n.a. Not available. a. Annual wage expenditure divided by days of employment. b. Projected. Source: World Bank data. have been undertaken each year, but the rate of completion was ex- pected to fall. Some tentative estimates of the preliminary benefits from the first three years are given in table 7-8. Few conclusions can be drawn from the data, other than that the impact on agricultural production in the state was only slight in these years. Despite problems of project planning and selection, a majority of EGS expenditures can be classified in the "directly productive" cate- gory. Between 80 and 90 percent of all expenditures have been for irrigation of one kind or another or for soil conservation and land development projects. This emphasis is one of the distinctive features of the EGS. The costs of financing the EGS have been growing fast (table 7-7), but half of its expenditures are funded by special taxes, nearly 70 per- cent of which are in turn generated by a tax on professional incomes, mainly in urban areas. The inflationary impact of EGS spending should therefore be minimal provided that deficit financing is avoided; to do so, other state expenditures will need to fall as EGS out- lays increase. Although the primary inflationary potential of the scheme as currently constituted thus appears to be small, its second- ary impact on food-grain prices could be high because most of its out- lays are on wages and wage recipients are likely to have high marginal 268 Rural Public Works Programs Table 7-8. Maharashtra's Employment Guarantee Scheme: Area Benefited and Estimated Cumulative Impact on Production and Employment, October 1977 Additional Additional annual Area benefited production employment Program (hectares) (tons) (thousands of workdays) Minor irrigation 150,000 150,000 6,000 Completed 14,500 14,500 580 Incomplete 135,500 135,500 5,420 Percolated tanks 45,000 45,000 1,800 Completed 10,000 10,000 400 Incomplete 35,000 35,000 1,400 Contour bunding 560,000 28,000 n.a. Irrigation channel bunding 2,000 2,000 80 Completed 900 900 36 Incomplete 1,100 1,100 44 All programs n.a. 225,000 7,880 Completed n.a. 53,000 1,016 Incomplete n.a. 172,000 6,864 n.a. Not available. Note: Total food-grain production in Maharashtra in 1974-75 was 7.8 million tons. The 225,000 tons expected additional production (assumed to be food grains) from completed and incomplete projects over nearly three years of operation amounts to 75,000 tons annually, or less than 1 percent of total annual production. Source: World Bank data. propensities to consume and high income elasticities for food-grain consumption. The EGS has had its share of problems. These include: L Low completion rates of projects. Although detailed data are unavail- able, surveys indicated to the World Bank that in 1978, 35 percent of the works on the master list had been completed, but they ac- counted for less than 7 percent of the total value of works sched- uled. Thus there appeared to be a predictably strong correlation between the size of the work and the speed of completion. * Inequitable distribution of benefits. The scheme has focused mainly on intermediate outputs-workdays of employment and asset creation-rather than final benefits. Although it was intended mainly to help landless laborers and the lowest decile of landown- ers, preliminary evidence shows that only about 46 percent of the beneficiaries have come from these groups. The remaining 54 percent own larger holdings-but this may be more an indication Rural Industrialization and Employment 269 of the poverty of landowners in rural areas of Maharashtra than a sign of the failure of the EGS to reach its target groups. Large landowners have also gained from the creation of assets on their holdings. To the extent that these investments have increased the demand for labor, however, indirect distributional benefits are still progressive."4 Planning and selection of projects. The greatest weakness in the man- agement of the EGS may be the limited ability of local authorities to undertake the physical surveys needed to prepare suitable new works or to develop integrated projects. So far the state has not been prepared to pay for preliminary design work. Moreover, there is little scope in many areas for EGS projects on publicly owned land. Projects on private land usually involve a 50 percent loan and a 50 percent state subsidy. The EGS, however, offers several important lessons for the successful use of public works programs to reduce widespread rural unemploy- ment. These are: * Rural public works programs can be designed and administered on a large scale. In addition, they can have a significant impact on long-term employment prospects in rural areas. * If properly financed, such programs need not be inflationary and can even serve as a means for urban-rural transfers. *Political commitment to employment creation is needed at the highest level. The EGS represents such a commitment. Sometimes viewed as a social contract between the state and the citizen, it in- corporates concepts such as the dignity of labor, the right to em- ployment and group participation, and the state's responsibility for the timely execution of public works. It can also be seen as an opportunity for mobilizing and organizing rural labor throughout the state or as a first step toward the unionization of rural labor. * Perhaps the most important contribution of the EGS is its adoption of the watershed as the appropriate unit for the planning and management of rural development. This approach is based on the belief that a watershed is a stronger economic community than a village. It raises the possibility of using the systems ap- proach advocated by Minhas (in Srinivasan and Bardhan 1974) to undertake jointly three components usually thought of as sepa- rate and distinct-land consolidation, development, and rural public works for employment. Although the EGS does not include a land consolidation program, its choice of the watershed as the planning basis for rural public works and its emphasis on irriga- 270 Rural Public Works Programs tion and soil conservation projects certainly contain the other two elements. The Indian National Rural Employment Program The experience with Maharashtra's EGS and the earlier Food for Work Program encouraged the government of India to institute the nation- wide National Rural Employment Program (NREP) in 1980. This pro- gram, whose purposes and modes of operation are similar to those of the EGS, uses unemployed and underemployed labor in rural areas to create productive community assets. The program focuses not only on productive assets such as irrigation infrastructure, drainage sys- tems, and soil conservation schemes, but also on support for basic so- cial services such as wells, village schools, health subcenters, and rural housing and roads, which are being developed under the Minimum Needs Program (see below). In 1980 an estimated 1.5 million person years of employment were provided under NREP. This was less than 8 percent of the total rural unemployment of some 19 million esti- mated by the World Bank in that year. These programs, however, are not meant to provide full-time year- round employment; they merely attempt to supplement the income of a broad segment of the rural population-especially during the ag- ricultural slack seasons. As such they can significantly raise employ- ment opportunities in rural areas and narrow the gap between labor supply and demand. The creation of productive assets is expected to generate permanent jobs, while increases in social assets will improve the quality and productivity of rural life. Although it is too early to assess the overall impact of the NREP, it represents a major commit- ment on the part of the government of India to address the problems of rural unemployment. No such nationwide programs exist yet in Pa- kistan or Bangladesh. Conclusions Although the extension of Maharashtra's EGS experience on a nation- wide basis through the NREP offers a glimmer of hope for new jobs in rural South Asia, rural public works programs cannot solve the problem of rural unemployment. It is a structural problem caused by unduly rapid growth in the labor force coupled with unduly slow growth in the demand for rural labor. Rural-urban migration merely displaces the problem from the countryside to the cities. Extensive and ongoing labor mobilization in rural areas may well be part of the answer. But in the long run, rural works programs, no matter how comprehensive and well executed, cannot be more than a palliative. Rural Industrialization and Employment 271 Substantial long-run reductions in rural unemployment will be achieved only by structural change that raises the demand for labor through vigorous agricultural growth and associated rural industriali- zation. Rural works programs will therefore be inadequate; growth in farm and nonfarm incomes remains the only viable solution to pov- erty in the long run. In the interim there seems to be a need for spe- cial programs to assist the rural poor and to alleviate the burden of poverty. Specialized Poverty Programs In addition to rural work programs like the NREP, an extensive effort is now being made to develop programs specially designed to assist the rural poor. The commitment to such programs, at least on a na- tionwide basis, is currently confined to India. The main nationwide poverty programs now operating in India are the Minimum Needs Programs (MNP) and the Integrated Rural Devel- opment Program (IRDP). Whereas the purpose of the NREP is to pro- vide more employment opportunities in rural areas, the MNP is geared to improving the human capital and quality of life for the rural poor and the IRDP to helping poor rural households acquire productive assets. Minimum Needs Program Over the years, the government of India has sponsored a wide variety of programs to make certain minimum social services available to all. A core group of these programs-elementary education, adult educa- tion, rural health, rural water supply, rural electrification, rural roads, housing for the rural and urban poor, and nutrition-are now being implemented through the Minimum Needs Program, which started in 1980. Some of these programs (education, health, water supply, nutrition) explicitly invest in human capital, while others (such as electrification and roads) develop a minimum level of infra- structure in less advanced rural areas. Some Rs58 billion were allocated to MNP during the sixth plan pe- riod (1980-85) (see tables 7-9 and 7-10). Although performance under such umbrella programs is always difficult to assess, data avail- able to date suggest that overall results have been fairly good. Rural electrification, roads, and water supply improve living conditions for all rural households, whereas health, nutrition, education, and hous- ing programs have a more than proportional benefit for the rural poor. 272 Specialized Poverty Programs The rural poor, especially the poorest, suffer disproportionately from inadequate nutrition and lack of access to health care and edu- cation, and they typically have relatively large families and high child- to-adult ratios (see chapter 2). Therefore, improving their access to family planning, nutrition, health care, better water supplies, and housing will not only improve their quality of life but substantially im- prove their ability to work. But in the long run providing minimum or basic needs solely as consumed services, without increasing the productive capabilities of the rural poor, would only burden already poor societies. It is pre- cisely because improvements in health, nutrition, and education actu- ally increase the productive capacity of the poor that such programs make economic sense. Better health and nutrition, along with educa- Table 7-9. India: Planned Outlays on Selected Poverty Programs (billions of rupees in 1979-80 prices) 1980-81 to Original planned Percent 1983-84a allocation, 1980-85 (1) 1(2) Item (1) (2) (3) Selected poverty programs Minimum needs 34.6 58.1 59.6 Rural developmentb 26.0 34.9 74.0 Memo items Crop production 16.1 29.6 54.4 Irrigation 62.6 121.0 51.7 Total plant 596.6 975.0 61.2 a. Actuals for 1980-81 and 1981-82, except for rural development expenditures which are revised estimates; revised estimates for 1982-83 and budget estimates for 1983-84. b. Excludes community development and cooperation; includes Integrated Rural Development Program and National Rural Employment Program. Source: World Bank data. Table 7-10. India: Government Expenditure on Antipoverty Programs (millions of rupees) 1980-85, 1985-86, 1986-87 1987-88, 1988-89 Program sixth plan revised budget revised budget Integrated Rural Development Program 15,000 2,786 4,274 2,970 3,462 National Rural Employment Program 18,000 3,372 4,427 6,820 7,300 Rural Landless Employ- ment Guarantee Program 4,000 6,063 6,337 6,529 5,294 Total 37,000 12,221 15,037 16,310 16,050 Sources: Dandekar (1986); Rath (1988). Rural Industrialization and Employment 273 tion and the promotion of family planning, will also decrease popula- tion growth rates. This itself can have enormous economic benefits, particularly for the rural poor (see Quiz6n and Binswanger 1986). The Integrated Rural Development Program The IRDP was initiated in India on a nationwide basis in the early 1980s to increase the income-generating assets of the poor directly through a mixture of subsidies and credits. The IRDP was allocated nearly Rs15 billion under the sixth plan (see table 7-10). It integrates a number of earlier programs (notably the Small Farmers Develop- ment Agency and the Marginal Farmers and Landless Labor Agency) which had offered credit subsidies to the poor. The purpose of the IRDP iS to provide a group of selected beneficiaries (identified as those below the poverty line) with a mixture of credits and subsidies (the ratio specified is Rs2 credit to Rsl subsidy) in sufficient amounts to purchase assets such as draft and milk animals, small stock, and farm equipment. The intention is to generate additional income to raise the recipients above the poverty line. The program grew very rapidly in the early 1980s, from Rs4 billion in 1980-81 to RslO billion in 1982-83, the number of recipients in- creased from 2.8 million to 3.3 million families a year, and the invest- ment per family more than doubled in current prices. The IRDP and related programs now account for 5.13 percent of the total outlay for the agricultural sector in the seventh plan (Rath 1988). Any program of this size and scope and with this rapid a growth in commitments and beneficiaries is bound to have implementation problems. Funds are inadequate for the scope of the job. In addition, they are allocated in a fixed amount per block without any reference to the number of poor, which leads to anomalies in allocations. The beneficiaries are supposed to be the "poorest of the poor," but identi- fication of the targeted group is a pervasive problem. Other concerns are with the overconcentration on livestock, the question of whether the effects will be lasting, and the strain placed on the administration of other services. The problem of identifying target groups also plagued the earlier programs. Part of the problem is administrative: first, the program has grown fast; and second, because progress is measured by the number raised above the poverty line, there may be an incentive to concentrate on those at or near that line. But a more fundamental difficulty is that, since substantial subsidies are involved, relatively better-off farmers use their influence to participate in the program. In Tamil Nadu and Maharashtra the average per capita income of the beneficiaries in each district was higher than that of the control 274 Specialized Poverty Programs group of nonbeneficiaries before the program started. Moreover, a significant proportion of the beneficiaries (15-60 percent) were better-off farmers (India, Planning Commission 1982, pp. 41-49). Only in cases where local villages were involved in selecting the bene- ficiaries was the selection concentrated toward the very poor. Com- munity participation should therefore be made mandatory in select- ing beneficiaries. Only in this way can the powerful and the rich be prevented from milking such programs for their own benefit, often in collusion with the bureaucratic apparatus set up to administer such programs. The budgetary commitments do seem inadequate to make a dent in the problem. With an average annual investment of about Rs2,000 per beneficiary in 1979-80 prices, and even with an optimistic investment-output ratio of 2.0, the most that can be generated is a gross monthly income stream of about Rs8O per household (or Rs 15 per capita, assuming an average household size of 5.6). Since the aver- age monthly per capita expenditure for those below the poverty line was about Rs5O and the poverty line was about Rs75 in 1979-80, an increase of Rs8O would offer a major breakthrough in living stan- dards, especially for those in the lowest decile with per capita expendi- tures of around Rs3O. But this is before deducting loan repayments (at 10.25 percent interest these are an estimated Rs9.5 per capita), working expenses (fodder for animals, for example), and legal and illegal administrative costs and fees associated with participating in the program. Evidence suggests that few IRDP beneficiaries have been able to raise their incomes significantly more than the control group of non- beneficiaries. Also, given the extreme poverty of recipients and the risks attendant on being poor, there are strong incentives to sell the assets acquired by these loans to meet contingencies or consump- tion needs, rather than to wait for gains from the investment. At any rate, much higher sums will have to be made available if net consump- tion is to be raised. If those at the lowest levels get these loans and use them appropriately, the income-generating benefits could be sub- stantial. The largest component of IRDP assistance is for the purchase of live- stock. Because of the potentially high returns associated with dairying and the use of draft animals, however, the problems are not with this concentration as such. Rather, the difficulties are that beneficiaries often lack experience, veterinary services are lacking, funds remain inadequate to purchase more productive and higher grade animals, and working funds (for fodder or grazing payments) are unavailable. Still, poor beneficiaries who acquire dairy or small stock can make a viable return when complementary marketing and other services are Rural Industrialization and Employment 275 provided. The National Dairy Development Program is one source of assistance, but more attention needs to be given to expanding such services. With nearly 3 million families a year joining the IRDP, government agencies in rural areas are strained, and other development services, such as extension, may suffer as a result. Some rural credit agencies are already under serious financial pressure, with high and rising amounts overdue on loans of increasing maturity (see chapter 2 and Tendulkar 1983), and there is skepticism about whether these loans will be paid back. Conclusions It is too early to assess whether these special programs will make a major contribution to alleviating rural poverty. Nevertheless, they tackle the problem via direct efforts to help the rural poor, and both their scope and their expenditures are extensive. Whatever their cur- rent shortcomings, they demonstrate the government's commitment to a substantial effort to help those with the greatest needs-the rural poor. Given the pervasive poverty in South Asia, the need for such a commitment and for the practical means to make it effective is obvi- ous. Whether the programs will succeed in doing what they are in- tended to do is not yet clear. What can be said, however, is that only in India (perhaps reflecting the pressures exerted within its demo- cratic polity) is there yet even a commitment in principle to exploring ways of direct alleviation of rural poverty. Notes 1. See Hirschman (1958); Lele and Mellor (1972a and b); Johnston and Kilby (1975); and Mellor (1976). 2. See also Liedholm and Chuta (1976) for an example based on Sierra Leone, and ILO (1972, 1974) for studies of Kenya and the Philippines. 3. For Bangladesh see N. Islam (1978) and Stepanek (1979), and forPakistan see D. A. Khan (1978). 4. Analysis of data on consumer goods industries by Lele and Mellor (1972a and b) suggested that capital-labor ratios were relatively low and that such high ratios as were found applied primarily to industries catering to the demands of urban consumers. 5. Fixed capital requirements range between RslO,OOO and Rs2O,OOO per unit of employment in most food processing industries (rice milling is low with Rs5,400), around Rs3O,OOO in chemicals, and RslO,000-55,000 in bicy- cles, sewing machines, light engineering, tanning, and the aluminum, brass, and copper industries (data for 1954-56). Generally, value added per worker increases significantly with the size of establishment. Labor productivity is 276 Notes four to six times higher in large firms than in small ones. The value added per unit of employment ranges from Rs3,000-15,000 in food processing, to RsI3,000 in bicycles and sewing machines, and to Rs2O,000-35,000 in chemical-related industries. 6. Much of this section is based on an excellent article by Pandit (1978); see also D. Gupta (1980). 7. In 1971-72 the shares of Punjab alone in India's exports of various com- modities were as follows: woolen hosiery, 100 percent; sports goods, 68 per- cent; bicycle parts, 67 percent; completed bicycles, 29 percent; domestic sew- ing machines, 14 percent; and pipe fittings, 12 percent. In 1969-70 Haryana accounted for 80 percent of the country's exports of hand tools and small tools and 73 percent of exports of plumbing fixtures (Pandit 1978). 8. This estimate is biased upward because data on very small, informal, less capital-intensive units are not available below the two-digit classification used. 9. See Burki (1969); Perkins and Yusuf (1984); Rawski (1979); and for the Chinese experiences, Keesing (1975). 10. See Sigurdson (1972, 1977, and 1978); Riskin (1978); and report of the American Rural Small-Scale Industries Delegation to China, Rural Small- Scale-Industries in the People's Republic of China (Berkeley, Calif., 1972). 11. The discussion of China's rural industry that follows is based on World Bank material and data. For a more extensive treatment, see Byrd and Lin (1990). 12. The major sources for this section are Reynolds and Sunder (1977); Dandekar and Sathe (1980); Abraham (1980); World Bank data; a newspaper article, "Miles To Go in Maharashtra," Financial Express, January 29, 1977; and India and Maharashtra (1980). On the Integrated Rural Development Program, Dandekar (1986); Kurian (1987); and Saxena (1987). 13. This record may be more impressive than that of other public works programs, but it compares unfavorably with the 496 million workdays gener- ated under scarcity relief in 1973-74 in Maharashtra. 14. Minimum wage legislation may also be a factor. Some farmers have re- ported that workers prefer EGS work to farm labor, even when higher wages are offered. But workers may prefer EGS less for its wage benefits than for the opportunity to escape irksome and even degrading relations with farm employers. 8 Land Reform: The Missed Opportunity Several times in this book I have referred to the contention that rural poverty in South Asia has its roots in the structure of landholding. It is reasonable then to maintain, as many do, that the only remedy lies in the radical alteration of this structure, mainly through funda- mental measures of tenancy reform and land redistribution. This view may be too simplistic, however, and is often inconsistent with ob- served reality. This chapter briefly summarizes the institution of ten- ancy, examines land reform in general, including measures to reform tenancy and to alleviate the concentration and fragmentation of oper- ated holdings, and assesses the feasibility and desirability of land re- form as a means of reducing rural poverty in the subcontinent. How important is tenancy? Accurate data on the number of ten- ants, area under tenancy, and the type of tenurial relations are hard to find and vary from source to source. Unrecorded tenancy relation- ships are widespread, so it is safest to assume that the available figures are underestimates. The extent, type, efficiency, and changes in the patterns of tenancy in South Asia are examined in some detail by Singh (1988c). Here I summarize the main findings from that study. The best available data are for India in the early 1970s, and they show that only a quarter of all holdings were under tenancy. The vast majority of these were partial tenancies (operators own part of their area and lease-in part of it), with pure tenancies (wholly leased-in area) an insignificant 4 percent of all holdings. There had been a slow de- cline in the number of mixed tenant holdings, but a rapid decline in the number of pure tenancies. The total area under tenancies (leased- in area) of all types was small (around 11 percent) and had declined significantly countrywide (Sanyal 1977b). There was considerable var- iation among states, but nowhere was the percentage overwhelmingly large and everywhere it had been declining. The highest rates of ten- ancy (25-30 percent of the leased-in area) remained in areas where 277 278 Land Reform: The Missed Opportunity the Green Revolution was the most successful-Punjab and Haryana. But tenancy had always been highest there, and even there tenancy had dramatically declined (Sanyal 1977b). Tenancy was not exclusively associated with small holdings-those smaller than 5 acres accounted for 72 percent of all holdings but for only 37 percent of the area leased-in. It is true that in many states, mainly in eastern India, most of the area under tenancy was in small holdings, but most of the holdings were small. Of the households leasing-in land, nearly a third were landless. The practice of leasing- out land was very common among smallholders (those with fewer than 5 acres), who accounted for nearly a third of all area leased out. So the simple association between "small" and "tenant" and "large" and "landlord" cannot be made-many smallholders are landlords, and others are both landlords and tenants at the same time (tables 8-1, 8-2, and 8-3). The dichotomy between landlords on the one hand and tenants on the other is therefore invalid. Sharecropping (mainly though not exclusively 50:50 shares) is the most predominant form of tenancy in most states. Fixed-kind tenan- cies predominate in some southern states where plantation crops are important, and fixed-cash tenancies are becoming prevalent in areas most affected by the Green Revolution, such as Punjab and Haryana. The high incidence of small holdings, of tenancy among small hold- ings, and of sharecropping poses a triple threat to areas in some east- ern Indian states where tenancy was formerly characterized by rent collectors known as zamindars. There the relationship between small share tenancies and technological change and the distribution of its benefits is of special concern. The data on tenancy in Bangladesh (table 8-4) are of somewhat dubious quality, but they suggest that in the 1970s over a third of the holdings and a quarter of the cultivated area were under mainly mixed tenancies; pure tenancy accounted for only about 8 percent of both holdings and area. Tenancy seemed to be increasing in terms of both holdings and area (Jabbar 1977). Most mixed tenancies are not small but were dominated by medium and large holdings of be- tween 2.5 and 12.5 acres. Sharecropping (almost exclusively 50:50 shares) was the absolutely dominant form of tenancy, and there was little evidence of a transition to fixed-cash or other forms of tenancies (Sanyal 1977b). Most sharecropping arrangements are of short duration-a third are for a year or less-which indicates the extreme lack of security of these tenures. In Pakistan the available data show that in 1972 tenancy was a pre- dominant fact of agrarian conditions with owner-tenant and pure- tenant farms accounting for two-thirds of all holdings and area (table 8-5). Pure tenancy was significant and accounted for two-thirds of Table 8-1. India: Tenancy in Small Farms by Size of Operated Holdings, 1971-72 (percent) Size Maharashtra Gujarat Kerala Punjab All India (acres) Number Area Number Area Number Area Number Area Number Area 0-1.0 11.4 0.2 9.3 1.0 63.0 22.4 0.6 ... 19.7 3.3 1.0-2.5 13.0 0.7 17.8 3.3 19.5 20.7 2.9 0.7 28.4 13.2 2.5-5.0 24.6 10.7 14.2 6.9 10.6 26.0 19.3 9.0 24.2 20.4 5.0+ 51.0 88.4 58.7 88.8 6.9 30.9 77.8 90.3 27.7 63.1 ts, All sizes 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 t Number (hundreds) 558 - 218 - 389 - 355 - 14,655 - Area (thousands of acres) - 2,729 - 827 - 242 - 1,880 - 32,814 All operated area (millions of acres) 44.4 8.6 2.8 6.7 310.4 All operated holdings (millions) 4.7 - 2.3 - 2.3 - 0.7 - 57.1 - - Not applicable. ... Negligible. Note: The table shows the percentage distribution of the number of holdings leasing-in and the area leased-in by the size group of operated holdings. Source: National Sample Survey (NSS), Report 215, 26th Round (1971-72), "Tables on Landholdings," State Reports, table 19. 280 Land Reform: The Missed Opportunity Table 8-2. India: Operating Households by Tenancy Status Operational holdings 1953-54 1961-62 1971-72 Number (millions) 55.5 50.8 57.1 Percent Entirely owned 60.2 72.6 74.4 Partly owned, partly tenanted 22.9 21.8 21.7 Entirely tenanteda 16.9 5.5 3.9 a. These are not necessarily operated by landless households. Sources: NSS, Reports 30, 36, 74, 8th Round (1953-54); Report 144, 17th Round (1961-62); Report 215, 26th Round (1971-72). Table 8-3. India: Changes in Leased-in Area as a Percentage of Total Area Cultivated State 1953-54 1960-61 1970-71 Andhra Pradesh 19.07a 9.15 9.01 Assam 43.54 15.36 19.69 Bihar 12.39 10.25 14.50 Gujarat n.a. 5.83 3.91 Haryana - 23.26 Punjab 76.89b 35.39c 28.01 Jammu and Kashmir 22.17 14.13 8.06 Karnataka 16.37 18.16 15.90 Kerala 23.63 15.30 8.59 Madhya Pradesh 18.61 6.40 7.46 Maharashtra n.a. 8.74 6.15 Orissa 12.58 10.75 13.46 Rajasthan 12.92 4.87 5.26 Tamil Nadu 27.53 16.55 13.07 Uttar Pradesh 11.38 8.06 13.01 West Bengal 25.43 17.65 18.73 -Not applicable. n.a. Not available. a. Figure is for Andhra region. The figure for Hyderabad State, which is included in present-day Andhra Pradesh, was 18.04. b. Punjab includes present-day Haryana plus Delhi and Himachal Pradesh. c. Punjab includes present-day Haryana and parts of Himachal Pradesh. Source: Bardhan (1976). all tenancies and 30 percent of the total operated area. There were notable regional differences with mainly mixed tenancies concen- trated in medium-size (5-25 acre) holdings in the Punjab and smaller pure tenant holdings that were sharecropped in Sind (M. H. Khan 1981). The "pure-tenants at will" (the haris) were at a special disad- Land Reform: The Missed Opportunity 281 Table 8-4. Bangladesh: Changes in Tenant Holdings and Area, 1960-78 Cultivating households Operational holdings Item 1960 1967-68 1973-74 1977 1978 Operational holdings or cultivating households Number (millions) 6.1 6.9 7.0 8.2 9.3 Percent Owner-operators 61.0 66.0 67.0 61.2 64.5 Owner-tenants 37.0 30.0 27.0 32.0 28.1 Pure tenants 2.0 4.0 6.0 6.8 7.4 Area cultivated Millions of acres 21.7 21.6 n.a. 18.8 24.3 Percent Owner-operated 54.0 57.8 n.a. 54.0 58.5 Owned and leased in 45.0 38.9 n.a. 41.7 37.7 Purely leased in 1.0 3.4 n.a. 4.3 3.8 Under all forms of leasing in 18.0 17.0 25.0 22.9 23.7 n.a. Not available. Sources: Bangladesh Agriculture Census,1960; Bangladesh Bureau of Statistics (1972, 1977); Bangladesh, Ministry of Agriculture and Forests (1978); and Jannuzi and Peach (1980). vantage in relation to their landlords who were large owners, and the dichotomy between landlords on the one hand and tenants on the other was certainly valid. There were significant differences between East and West Punjab. Tenancy in the East Punjab was less important in holdings and area, pure tenancy was insignificant, sharecropping was giving way to fixed-cash tenancies, and most tenancies were concentrated in the 5-10 acre range. In the West Punjab, the proportion of total area under tenancy was almost twice as large, pure tenancy was significant, sharecropping was the dominant form, and the average size of the tenant holdings was more than two times larger. Whether medium and large mixed tenancies are a barrier to technological innovation is the pertinent question to ask in the case of the Punjabs (Singh 1988c). Tenancy as an Institutional Constraint Much of the traditional debate on agrarian conditions has been col- ored by an altogether negative view of tenancy as an institution. In particular, it is argued that tenancy, share tenancy especially, is an economically inefficient way to organize production; impedes the adoption of new, particularly yield-increasing, technologies; allows Table 8-5. Pakistan: Regional Changes in Tenancy, 1960-72 Percentage change 1960 1972 between 1960 and 1972 Total area in Total area in Total area in Number of operational Number of operational Number of operational Item far,ms holdings farms holdings farms holdings Pakistan All farms (millions) 4.9 48.9 3.8 49.1 -23 ... Percent Owner operated 41 38 42 40 -21 4 Owner-tenant operated 17 23 24 31 8 38 Tenant operated 42 39 34 30 -36 -24 Punjab All farms (millions) 3.3 29.2 2.4 31.0 -29 6 Percent Owner operated 43 38 42 39 -29 7 Owner-tenant operated 19 25 29 36 10 54 Tenant operated 39 37 29 26 -47 -26 Sind All farms (millions) 0.7 9.7 0.7 9.5 10 -2 Percent Owner operated 20 29 24 31 30 3 Owner-tenant operated 9 15 13 19 62 25 Tenant operated 71 56 63 51 -2 -12 ... Negligible. Source: M. H. Khan (1981), pp. 108-11. Land Reform: The Missed Opportunity 283 landlords a disproportionate share of the income from cultivation and the incremental gains from new technologies and leaves few gains to the tenants; and serves few, if any, functions in the agrarian econ- omy, except the exploitation of tenants by landlords (Ladejinsky 1977). As a consequence, tenancy is seen as a major barrier to the redress of poverty. The remedy sought is to abolish tenancy and "give the land to the tiller" or, where this is not feasible, to restrict tenancy by a variety of tenurial reforms. Legislation has been proposed to limit the rental shares going to landlords, give security of tenure to tenants, and force landlords to bear a larger share of cultivation costs. The Role of Tenancy Those who want to abolish tenancy fail to understand that it is a ro- bust and dynamic institution with many important functions in the existing agrarian structure. It has long been known (since Adam Smith) and widely recognized (by Marx and Marshall among others) that share tenancy is a mechanism for adjustments between land and other resources, an incentive system that brings forth more effort at less cost for supervision than does farming with wage labor, and a credit system that allows those without assets to acquire capital and credit (Braverman and Srinivasan 1980). More recent literature has added the views that share tenancy serves to spread production risks among differently risk-averse agents (Cheung 1968; Reid 1974; Stiglitz 1974), to overcome the indi- visibility of inputs where lease markets are imperfect, to reduce trans- action costs where information is incomplete and supervision costly (Jaynes 1979a and b), to allow agents to share their managerial and entrepreneurial skills, and to allow labor contracts where labor mar- kets are risky and imperfect and nonstandard labor is involved. Share contracts are seen as only a second-best solution-an imperfect re- sponse to incomplete and imperfect markets for land, credit, labor, management, information, and insurance. Indeed, much of the re- cent rigorous work suggests that with high transaction costs, incom- plete markets, and uncertainty, first-best Pareto-efficient solutions are just not possible. But it is because these markets are imperfect and incomplete that share tenancy serves important functions. If it is abolished, these functions will have to be served by other institutions-markets or gov- ernment agencies. As markets develop, the reasons for sharecropping become more tenuous, and other forms of tenancy with a fixed rent become more common. There is considerable evidence of this transi- tion in the agriculturally more advanced regions of South Asia. 284 Tenancy as an Institutional Constraint Efficiency Many economists have agreed with Marshall (1961) that share ten- ancy is inefficient; others, along with Cheung (1969) and Newbery and Stiglitz (in Roumasset, Boussard, and Singh 1979) to mention only a few, have argued that it can be efficient. Since the theoretical arguments are logically correct, the conflicting results must follow from different assumptions about the nature of share contracts. The central differences revolve around the degree of control that land- lords can exercise. Thus if landlords can control and supervise the combination of inputs to some extent, share in the costs of cultivation, and evict tenants at will by using only short-term leases, it is recog- nized that share tenancy would be as efficient as fixed-rent or owner- cultivated systems (Vyas 1970; C. H. H. Rao 1971; Bliss and Stern 1980). The question of efficiency rests on how share tenancy operates in practice, not on any theoretical grounds. Since the issue can be re- solved only empirically, what is the evidence from South Asia? Testing for differences in efficiency is not a straightforward matter. At a minimum, there is a need to control for farm size when compar- ing, say, owner-operated with tenant farms (Chakravarty and Rudra 1973). But other confounding influences such as quality of land, ac- cess to irrigation and credit, different attitudes toward risk, and dif- ferent skill and factor endowments make it difficult to interpret the results. There are very few empirical studies that are completely satis- factory. To compare sharecropped with owner-operated farms, I first ex- amine input intensities and output per hectare after controlling for farm size and then look at cost-sharing, duration of leases, and ability on the part of landlords to control inputs. In India evidence from many geographically diverse regions sug- gests that there are no marked differences in input-output patterns of owner- and tenant-operated farms of the same size and hence no significant tenancy-related inefficiencies. There are perhaps two ex- ceptions: on very small tenant farms (fewer than 5 acres) input inten- sities are lower in Bihar, and Bell (1977) shows input intensities and outputs to be lower on leased-in land than on owned land operated by the samefarmer (this is one way to control for other sources of varia- tion). But Bardhan and Rudra (1980a and c, 1981) find no differences in West Bengal. There is only one study from Pakistan (West Punjab), and it found no significant differences between tenant and nontenant farms. A few studies done in Bangladesh also found no differences, when size is controlled for, between tenant- and owner-operated farms. Where owned and sharecropped land under the same cultivators was ob- Land Reform: The Missed Opportunity 285 served, mixed tenants were found to cultivate leased-in land less in- tensely (Hossain 1977b). Even given the difficulty of controlling for confounding effects and the great diversity of agrarian conditions, the weight of the evidence suggests no significant inefficiencies associated with sharecropping systems, at least where tenanted farms are of medium or large size. Some inefficiencies have been noted on very small share tenancies (including owned and leased land cultivated by the same operator). This finding is of some significance because in many states in India- Bihar, Kerala, Orissa, Tamil Nadu, Uttar Pradesh, and West Bengal- and in Bangladesh such tenancies account for more than 50 percent of the tenanted area. But in these cases, sharecropping per se may not be the source of all the inefficiencies. They may be caused by lack of resources or poorly developed or imperfect factor and product markets that place small farmers at a disadvantage. The true source of the inefficiency is hard to identify, and removing tenancy or chang- ing its terms may not be the best way to deal with the problem. Evidence on cost-sharing reveals that in India it is a new but not unusual part of a tenancy contract, and its importance is increasing, especially where HYVS and irrigation have increased productivity. In some areas affected significantly by the Green Revolution, where costs are equally shared, efficiency is restored. But in most areas, only par- tial cost-sharing systems are prevalent. Cost-sharing is more common in the ryotwari areas-western, southern, and northwestern states in which landholders pay land taxes or rents directly to the state-than in the former zamindari areas in eastern India where intermediaries (zamindars) collected taxes for the state (Bell 1977; Newbery 1974; Jodha 1979). In Pakistan there is some evidence that costs are equally shared in the Punjab (Naseem 1971). Elsewhere costs are not shared or are shared unequally. In Bangladesh cost-sharing does not yet seem to be widely reported (Zaman 1973; Jannuzi and Peach 1980). Lease contracts are for a short time-predominantly one year but sometimes from two to four years-and getting shorter. Some are now given on a seasonal basis. The shorter leases are correlated with greater intensity of cropping and changing technology, but they may also be the result of uncertainty created by legislation. Although de- signed to give tenants "secure rights in land," this legislation may par- adoxically have increased the insecurity of tenants. There is also evi- dence that landlords are deeply involved in their tenants' decisions and in Bangladesh closely supervise the work. Both short-duration leases and close supervision and participation are means by which landlords attempt to control inputs and hence reduce inefficiencies. The weight of the overall evidence suggests that share tenancies are 286 Tenancy as an Institutional Constraint likely to show some inefficiency where the tenancies are small (fewer than 2 hectares), cost-sharing is minimal, technical change and pro- ductivity increases are slow, tenants are not under threat of eviction or have long leases, markets of all kinds are imperfectly developed, and semifeudal patron-client systems are prevalent. Such areas are found predominantly in eastern India, Bangladesh, and to some ex- tent Northwest Frontier Province in Pakistan. Elsewhere there are not likely to be any significant inefficiencies associated with share tenancy as such. Barrier to Innovation What of the argument that tenancy impedes the spread of in- novations-in particular, that small tenants will not adopt HYVS or use fertilizers because they cannot or because usurious, "semifeudal" landlords are more interested in making consumption loans than pro- ductive investments (Bhaduri 1973)? The "semifeudal" paradigm used as a basis for these arguments is, however, counterfactual. Evidence from West Bengal, the very region from which the paradigm is often drawn, shows that although land- lords do give consumption loans to their tenants, most of them are interest free. Furthermore, landlords do give production loans for new inputs, are interested in productive investments, do not have moneylending as a principal occupation, and do not bond labor at lower than market rates via consumption loans. Even small tenants themselves hire labor. Thus empirically many of the features of "semifeudalism" do not hold (Bardhan and Rudra 1980a and c, 1981). On the adoption of new technologies, extensive evidence shows no systematic relationship between tenancy status and either adoption of HYVS or nutrient use per hectare in most of the subcontinent. Where some evidence of such a relationship existed it was confined to eastern Indian states and Bangladesh. But much of it could probably be ex- plained by sharecroppers' far more limited access to credit and other inputs in these areas. Certainly where tenancies are mainly mixed or for larger plots, and where credit and new inputs have been readily available, tenants have done just as well as owner-cultivators. Thus it is safe to conclude that tenancy as such is unlikely to hinder the adoption and widespread diffusion of technologies, especially where tenants have adequate and timely access to modern inputs. Tenancy probably inhibits longer-term investment by tenants be- cause of the insecurity of small sharecropped holdings and lack of credit. But there is little direct evidence available. Where mixed ten- ancies exist and tenant holdings are larger-as in the Punjabs- Land Reform: The Missed Opportunity 287 tenants have been investing in tubewells and farm machinery, con- trary to conventional expectations of observers. The distribution of gains from share tenancy go disproportionately to landlords where the rental share is 50 percent of the harvest. Full cost-sharing (where landlords contribute 50 percent of the cost of in- puts) is the exception rather than the rule, and tenants often bear all the costs, not only for inputs but also for transactions in credit, la- bor, or product markets. The gains particularly from new technolo- gies have been generally lower on tenant- than on owner-operated holdings. Direct evidence reveals, however, that new technologies have bene- fited all classes of farmers, including tenants. Landlords seem to take a larger share of the gains, especially from smaller tenants and in areas where small share tenancies are predominant, but the absolute size of the gains is related more to the size of the holding than to ten- ancy status. The impact of new technologies on tenurial systems has been var- ied. In general, where HYVS, irrigation, multiple cropping, and other technologies are prominent the area under tenancy may even in- crease, but fixed-tenancy systems begin to replace sharecropping and pure tenancies. Poor landless sharecroppers may then have to rely on wage employment. Mixed fixed tenants with larger holdings become predominant. Reverse tenancies-with smallholders leasing out their land to larger owners-increase (Bell 1977). In the long run techno- logical progress displaces the poorer sharecroppers, reduces share- cropping in favor of fixed-rental systems and partial cost-sharing sys- tems, and may or may not reduce the area under tenancy. Although technological progress should not be slowed, policymakers must find ways of ameliorating its impact on the poorer segments of the share- cropping population who will be rendered landless. In the end, sharecropping will be replaced, and it can be viewed as a transitional (though long-lasting) phase in the development of land-lease markets. It will continue to be prevalent as long as farms are small and farmers are poor, markets are underdeveloped, and in- frastructure is weak in rural areas. It is poverty that leads to share- cropping, and not sharecropping that causes poverty. Conclusions There is no clear or direct relationship between the unequal distribu- tion of holdings and the incidence of poverty. Nor does the incidence of tenancy or the proportion of landless rural labor households in a region have a direct bearing on the extent to which the poor have 288 Tenancy as an Institutional Constraint benefited from various programs. The evidence in hand does not support the view that existing agrarian systems, or modes of produc- tion, have prevented small farmers and tenants from expanding out- put and taking advantage of new opportunities. Nor is it possible to ascribe backwardness and poverty to a semifeudal agrarian relation- ship in which exploitative landlords and moneylenders oppose tech- nological change; as has been noted in earlier chapters, the balance of available evidence supports just the opposite tendencies. Tenancy does not in fact seem to be a major barrier to efficiency, innovation, and productivity gains since few differences can be found between tenant- and owner-operated farms once size is accounted for. It does indeed appear that the gains from innovation have been un- evenly distributed between those with access to land, water, and in- puts and those without and that the "have-nots" in this context fre- quently (but by no means always) tend to be small tenant farmers. But this has been because of the imperfect nature of the markets in these goods and because institutions serving agriculture have been biased against small landowners, tenants, and the landless alike, rather than solely because of the nature of landholding arrangements as such. Where there has been technological progress, smallholders, including tenants, have gained directly while the landless have benefited less and indirectly, irrespective of the agrarian structure. The evidence does not support the view that such growth as has taken place has further impoverished the rural poor or ineluctably "proletarianized" them even where agricultural growth rates have been relatively high. Rather, the evidence seems to suggest that the main causes of agrar- ian backwardness and poverty lie in generally low or nonexistent rates of growth in agriculture and in the low productivities inherent in the traditional ways of deriving income from both land and labor. Relent- less demographic pressures and the consequent exhaustion of land resources have reduced opportunities in the countryside, especially where agricultural output has failed to grow significantly. In areas of stagnation the result has been increasing poverty, a rural economy in which holdings have become smaller (though not more unevenly dis- tributed), and a rise in the proportion of the rural population that has had to depend on wage labor for its livelihood. Thus the need is to accelerate growth-in both the farm and nonfarm sectors-and to increase the productivity of both land and labor inputs. Where this has been done and the growth of agricultural output has outpaced that of population, poverty has been reduced, though slowly. Nevertheless, the central tenet of the "structuralist" viewpoint is that the only long-run remedy to the problems of rural poverty is to change the present agrarian structure by radical land reforms. This proposition deserves serious consideration, if only because when Land Reform: The Missed Opportunity 289 stated in these simple terms it is correct and remains an influential idea. The case for reforms can be appreciated when one considers the sheer magnitude of the problem and the numerous remedies al- ready proposed: agricultural growth via new technologies in irrigated and semiarid areas; reforms of credit, irrigation, marketing, research, and other institutions serving agriculture to make this growth more accessible to small farmers (owners and tenants); growth in ancillary activities for the rural poor; supplementary growth in employment opportunities for rural labor both on farms and in rural industries, marketing, transportation, construction, and other rural services as well as in special programs for rural public works; and poverty pro- grams to provide the poor with assets or human capital. Yet all these measures may not be enough to alleviate the problems of absolute rural poverty in South Asia, at least in the short run. Stagnation only compounds the problem. But continued and sus- tained growth in agriculture, accompanied by rapid growth in indus- trial output and employment-conditions still not met in most of South Asia-could over a longer period solve at least the problems of absolute poverty. Problems of relative poverty will always remain. Extensive evidence provided here makes the case that rapid growth can in the long run solve the problem of poverty, but not that of un- equal development. If, like the structuralists, one is unwilling to wait for the uncertain prospects of the longer run, then one accepts the argument that only a radical transformation of the rural patterns of landholding will suf- fice. The nonstructuralists, however, argue that although radical re- forms may seem desirable in the abstract, the experience to date with reforms has not been salutary. The structuralists counter by saying that the reforms undertaken so far actually amount to "nonreforms," and what is needed is not only a strict implementation of current re- forms but even more radical measures. The nonstructuralist response is that, given the present balance of political forces in South Asia (or any reasonably credible alternative), the prospects for more radical reforms are extremely poor-even those already on the statute books have not been implemented. Furthermore, in the absence of radical reforms, there are many other good measures that can help and that need to be undertaken-measures such as those that have been out- lined in other chapters. There the argument rests. All those who have worked on problems of rural poverty in South Asia-structuralists and nonstructuralists alike-agree that growth is better than stagnation, that growth has not been rapid enough in ei- ther agriculture or industry, and that the benefits of growth will con- tinue to be unevenly distributed as long as the patterns of landholding remain unequal. The case for redistributive land reforms is further 290 Tenancy as an Institutional Constraint strengthened by the evidence provided here that smaller holdings could enhance employment and farming intensities and encourage wider use of new technologies. A more equitable distribution of hold- ings would also create more equitable demand patterns, which in turn should enhance nonfarm growth and remove some of the institu- tional biases in credit, marketing, and research that arise from the un- equal distribution of assets and associated power. Many argue that the best solution to rural poverty would be a radical redistribution of as- sets accompanied by rapid growth in agriculture and industry. Why then has this "best solution" been neglected? It is beyond the scope of this work to provide a complete answer, but one simple re- sponse might be that any fundamental change in the structure of agrarian relations in a predominantly agricultural economy is by its very nature revolutionary and can be undertaken only under a revo- lutionary and coercive policy, not an evolutionary and democratic one. For a variety of historical, cultural, and political reasons the polit- ical regimes in South Asia were and remain nonreformist and noncoercive in nature. And their present polity is not likely to change dramatically; on the contrary, the evidence suggests that evolution- ary, nonreformist tendencies may have been strengthened. Radical reforms which seemed at least feasible at independence seem to have become a remote possibility. In this sense such reforms represent a missed opportunity. This is not the place to argue about the socioeconomic aspects of wholesale and fully egalitarian land redistribution in South Asia. Suf- fice it to say that such an initiative could not be carried out under ex- isting social, political, or constitutional arrangements, that it would probably be only an interim stage on the road to collectivization, and that, as a revolutionary act, it could only be carried out as part of a broad, radical transformation of political, social, and economic rela- tions. South Asia seems far from developing the preconditions for genuinely revolutionary change-and experience in other countries suggests that its consummation does not always lead to better condi- tions. In the following section I examine the relevant facts and try to show that land reforms, though well intended, have not amounted to much. In particular, quite apart from the political and administrative problems associated with legislation and implementation, the reforms to date have not necessarily been helpful to the poor. Even successful and radical changes in landholding arrangements are not, under present conditions, a sufficient condition for rural prosperity, and there are no quick fixes for what is essentially a long-term problem. Because these reforms have been delayed, the task of alleviating pov- erty has become far more difficult as populations have continued to Land Reform: The Missed Opportunity 291 increase and available alternatives have continued to shrink. What was possible three decades ago seems less possible and even less desir- able today. A great deal has been written on the subject of land reform, espe- cially in South Asia.' At least three broad sets of measures are meant when land reform is discussed: tenancy reforms, the establishment of ceilings on holdings and the redistribution of land, and the consolida- tion of fragmented holdings. To the extent that data permit, I will discuss the desirability and feasibility of each of these approaches and assess the experience with them in South Asia. Tenancy Reform The term "tenancy reform" covers a bewildering variety of proposals that reflect very different perceptions of the so-called tenancy prob- lem. At one end of the spectrum are modest proposals to record the terms of leases for each crop season so as to give tenants access to crop loans. (Because land is used as collateral, recording of leases is sup- posed to encourage loans against crop shares.) At the other extreme are measures to grant tenants security of tenure in perpetuity and the right to pass on leases to their heirs, with no right of resumption for self-cultivation by owners. Between these extremes there are propo- sals for fixing rental payments, forcing landlords to share cultivating costs, or limiting landlords' rights to resumption for self-cultivation. Bangladesh, India, and Pakistan have extensive tenancy reform legis- lation.2 A review of the vast experience in this area suggests that ef- forts have all too often been ill-conceived, poorly implemented, and ineffective or counterproductive. In fact, tenancy reforms seem gen- erally to have done more harm than good to those they were meant to help. Tenancy as a Constraint: Some Myths and Realities The previous discussion of the nature of tenancy in South Asia pre- sents at least five findings, all of which run counter to the generally accepted wisdom that has influenced existing or proposed legislation. First, tenancy is not as important as it is sometimes made out to be. It does not account for an extensive area or for a large percentage of cultivated holdings-even if official figures are underestimates- except in Pakistan, where pure tenancy is mainly confined to rela- tively large holdings. Moreover, tenancy is declining in overall impor- tance, although it remains significant in some regions of India and mixed tenancies have increased somewhat in Bangladesh in recent years. This suggests that the amount of legislation and effort devoted 292 Tenancy Reform to combating it may be excessive and that it is unlikely to be an over- whelming constraint on aggregate agricultural growth. The problems of small tenants remain critical in some states, but few of the measures designed to help them are likely to be effective, as I will show. Second, tenancy performs some very important functions in rural markets and may often represent an appropriate second-best re- sponse to a host of problems confronting the parties to contracts in land. This response is made necessary in part because land and factor endowments are unequally distributed and in part because complete and perfect markets are nonexistent. The solution, however, is not to restrict tenancy contracts or to abolish tenancy. If this is done, the functions it now performs will have to be undertaken by other institu- tions, such as inefficient markets or government agencies. There is no guarantee that these alternatives will represent an improvement on the present system. Third, tenancy in itself is neither necessarily inefficient nor a bar- rier to technological innovations. In particular, tenants have adopted new HYv technologies where they have been available, profitable, and agronomically suitable. Further improvements in productivity through better farming practices, multiple cropping, and new inputs are and will continue to be possible on tenant holdings. Tenants have by and large performed as well as owner-cultivators and reaped simi- lar benefits. They have had to share these benefits with landlords- unevenly in some areas but less so in regions where relatively equal cost-sharing or fixed-rental systems have evolved. That landlords have had the lion's share of the gains from technological change is attributable not to tenancy per se, but rather to the imperfections of local markets and the fact that land is scarce, unevenly distributed, and becoming more productive. Fourth, in South Asia there is often no neat dichotomy between landlords and tenants. The poor may often rent out their land (be- coming landlords) while the better-off may often augment their exist- ing holdings by renting in land (becoming tenants). Thus legislation simplistically designed to help tenants against landlords may militate against many of the poor and benefit others who are far better-off. It is at best a blunt instrument that may end up doing the poor more harm than good. Fifth, a tenancy contract is more often than not a part of a series of linked, interdependent, and simultaneous contracts in a number of product and factor markets for land, labor, credit, and other in- puts. These linkages are organic and not some "deviation" from "nor- mal" agrarian relations. Nor are they necessarily exploitative; many have been shown decidedly to benefit tenants. In the case of those that are exploitative, attempts to redress the linkages in one market Land Reform: The Missed Opportunity 293 may be thwarted because the monopolistic or monopsonistic power involved can often be maintained by shifting the terms of the in- terlinked contracts to other markets.' Yet tenancy legislation is all too often designed as if existing conditions in other markets are im- mutable. In view of these considerations, what can be said about the desira- bility of the various kinds of tenancy reform that have been imple- mented or proposed? I review the three most widely discussed mea- sures below. Abolition of Intermediary Interests In the often confused debate on agrarian reform in South Asia it is sometimes forgotten that one of the most critical phases of tenancy reform has in fact been carried out. The zamindari system of rent- collecting intermediaries, which dated back to the Permanent Settle- ment of Bengal in 1793, was abolished in India between 1951 and 1956, despite administrative problems and widespread opposition (Ladejinsky 1977). In Bangladesh, too, intermediate rent-receiving interests were eliminated by 1956.4 As a result of these generally suc- cessful and forgotten reforms, millions of tenants were made secure on their land and freed from a host of illegal exactions; many of them acquired ownership of their land at moderate cost. The greatest shortcoming of the abolition measures was that they left unchanged the inequalities of land holdings and the position of sharecroppers and laborers. Their overall achievements were significant, however: the system of illegal exactions was abolished; permanent, heritable, and transferable rights were conferred on occupants; and the top layer of absentee landlords (those with the largest holdings) was skimmed off. But these reforms had their costs. Many zamindars acquired owner- ship rights over land from which they had previously only collected revenue, many tenants were evicted, and the large amounts of com- pensation paid to some zamindars turned them into rich agro- industrialists..5 The zamindari system has remained essentially intact in Pakistan de- spite moves to abolish it, largely because of the ability of the landed elite to sabotage reform. Landlords have traditionally held the bal- ance of power in rural areas, and by and large their position has not been changed.6 Tenancy remains a significant factor in Pakistan, and much of the current maldistribution of land holdings there may owe much to this initial failure. There is nothing in the existing features of tenancy to prevent the abolition of intermediary interests. Both this type of reform and the 294 Tenancy Reform direct granting of occupancy rights to subtenants are highly desirable. But such reforms need to be carried out at a single stroke and with no delay; where they have not been undertaken, as in Pakistan, the chance has probably been missed. Security of Tenure Many measures have been proposed for enhancing security of tenure. They include prohibitions on the eviction of tenants, granting of per- manent and transferable occupancy rights to tenants, prohibitions on the resumption of land by landlords for self-cultivation, and restric- tions on the subletting of land. Most existing legislation in South Asia has, however, been subject to the landlord's right of resumption, and this has been its Achilles' heel. The usual argument advanced in favor of these measures is that without security of tenure tenants will have no incentive to make per- manent improvements in the land, especially if there is no way to compensate them if they are later evicted. The obvious question to ask is, what would prevent landlords from making these improve- ments? They could presumably recover their costs and earn sufficient rates of return on their investments by charging higher rents. One answer lies in the supposed paradigm of the antigrowth landlord. But the paradigm is internally inconsistent and does not fit the observed facts. It is difficult to think of a situation in which those who own the land, have overwhelming economic power over their tenants, and re- tain the lion's share of the benefits of cultivation would not have an interest in making that land as productive as possible. Yet this is what is being argued by those who maintain that landlords will neither en- courage their tenants to increase productivity nor undertake produc- tive investments themselves. As I showed in the previous section, how- ever, landlords do in fact encourage yield-increasing technologies, provide production loans, and take a great deal of interest in produc- tive investments. Investment has been considerable, especially in areas in which new technologies have promised dramatic gains in pro- ductivity. Measures to provide greater security to tenants would be desirable if good intentions could be turned into reality. In fact, however, most measures of this kind have had the reverse effect of making tenants less secure. Legislative loopholes that allow resumption of land for self-cultivation have been mercilessly exploited by landlords, while the impossibility of policing every contract has flawed the legislation from its conception. The good intentions of lawmakers have been frustrated by their failure to generate political commitment through- out the system, to ensure that law enforcers have access to local Land Reform: The Missed Opportunity 295 sources of information about contracts, or to act quickly enough to prevent the counteractions of landlords. Speedy and wholehearted enforcement are crucial for the success of legislation to grant security of tenure. Any hesitation between an- nouncement of a measure and its implementation gives landlords time to develop a counterstrategy-as they have in fact done in India when the time lag has been significant. Evictions in response to half- hearted legislation started in many areas long before mechanization began to affect the eviction rate. At least evictions done to take advan- tage of mechanization have put land into the hands of cultivators with more economic assets, but evictions and shorter leases in response to poorly implemented tenancy legislation have created insecurity with- out necessarily improving productivity. Tenancy has been driven un- derground (hence my skepticism about official statistics), and the du- ration of leases has been reduced to as little as a single crop season.7 These effects of tenancy reform are, of course, the opposite of those intended by the sponsors of the relevant legislation. There is evidence, however, that with political will and commit- ment, together with access to grass-roots rural power, state govern- ments can in fact enforce measures to increase tenants' security of tenure. In Kerala land legislation was effectively implemented be- cause it was backed by a mass movement of potential beneficiaries, which was organized when the Communist party came to power (see Oomen 197 lb, 1975). For two decades Kerala was unique, but in 1979 Operation Barga was initiated by West Bengal's Left Front govern- ment with the Land Reform Amendment Act.8 This act tried to plug at least four loopholes in the previous legislation by preventing benami (illegal or fictitious) transfers of holdings, defining personal exemp- tions from land ceilings in terms of family rather than personal own- ership, shifting the onus of proof in disputes about tenancy rights from the tenant to the landlord, and attempting to record all bona fide bargardars (sharecroppers) so as to give them previously legislated occupancy rights.' This last feature is critical, because a program to grant rights to ten- ants can work only if there are records of who the tenants are. Since contracts have traditionally been oral, landlords have avoided giving security of tenure by disavowing or changing tenants, placing land in benami holdings, or resuming tenanted lands for self-cultivation. The West Bengal government's response has been described as follows: One of the important achievements of the Left Front government has been to record the rights of bargardars through group action. Group meetings are held with bargardars and they are encouraged to speak out their de facto status. This is followed by public verifica- 296 Tenancy Reform tion of bargardar claims in the presence of both landowners and the claimant sharecroppers. Landowners are given opportunity to express their objections which are also heard and verified in the field in public. Thereafter certificates of bargardari rights are dis- tributed among sharecroppers whose claims have been vindicated by the public. This group action is far more helpful to the bar- gardars than the traditional approach where bargardars would themselves have to seek the protection of the law. (Dutt 1981, p. A5a). Three ingredients seem essential for success: group action, the public nature of the proceedings, and state power exercised on behalf of the tenants.'" Even with these ingredients, speed remains crucial; as these measures were being contemplated, a large number of bargardars were evicted and land was "reformed" for so-called self- cultivation under "personal exemption" by landlords in West Bengal. It is too early to tell how Operation Barga will turn out. In its initial stages it was very successful; three-quarters of a million bargardars were identified and recorded in the first year alone. Two years later, however, only 1 million out of an estimated 2 million bargardar claims had been recorded; the remaining 50 percent were still unreg- istered (Khasnabis 1981, p. A45). Nevertheless, Operation Barga is an important achievement in view of earlier failures. It calls into question the idea that nothing can be done about tenancy reform in South Asia or that any new efforts are now too late. Above all, it shows that if measures to bring real security to tenants are to succeed, they must be backed by the power of the state at the local level." Without such backing, no amount of well- intentioned legislation will succeed in making tenants secure, for as soon as the government's intentions are known, tenants will be evicted unless the state has the local political power to enforce the measures. Thus, providing security to tenants depends on grass-roots enforce- ment and speedy action before landlords respond with evictions. A better approach would be to give tenants their land outright. Regulation of Rents and Cost-Sharing In the absence of the outright transfer of ownership rights in leased land to tenants-which has never been successfully undertaken in South Asia-the accepted route of reform has been to regulate the relations between landlords and tenants by fixing either landlords' rental share or their share of input costs. I have noted that 50:50 ren- tal shares are common in traditional sharecropping arrangements. Land Reform: The Missed Opportunity 297 Input cost shares have varied widely, but landlords generally seem to appropriate between 45 and 55 percent of any value added (Bell 1977). Extensive legislation was enacted in India in the 1950s, and in later years in Pakistan and Bangladesh, to fix the maximum rental share for landlords and to specify their input share. Legislated rents varied from one-fifth of the gross product of tenanted land in some states to one-half in others. (Tables A42 and A43 summarize legislation on rent control and security of tenure.) Although it was seldom specified just how much tenants would benefit from such measures, it was widely felt that enforcing maximum rental shares would significantly raise tenant incomes, perhaps even double them.'2 Their effect in practice was never tested, however, because the legislation was never enforced, and landlords generally continued to take their traditional share of rents. (See M. H. Khan 1981, p. 169, for a review of the Paki- stan reforms and Parthasarathy 1977 on the Indian reforms.) It is now generally agreed that legislation to fix rental ceilings did more harm than good: it prompted evictions or voluntary surrenders and, where tenancy continued, led to informal agreements with much greater insecurity for tenants."3 These measures displaced more ten- ants in the 1950s than the much criticized displacements in the name of mechanization or the Green Revolution in the 1960s and 1970s. The pace of evictions actually slowed in the early 1960s, once it was recognized that this type of legislation could not be enforced and was in fact a paper tiger. But the shift toward informal tenancy agree- ments brought about by these measures became permanent; tenancy effectively went underground and many tenants became landless la- borers.'4 Consequently, few of the official figures of the extent and nature of tenancy in South Asia can now be deemed reliable. Parthasarathy (1979, p. 342) sums the position up as follows: Powerful and enterprising landowners succeeded in evicting the tenants and in engaging them as labourers. The dramatic effect of tenancy legislation in the fifties was the gradual demise of the ten- ant and his exit into the ranks of the landless, more insecure condi- tions of tenancy and the shift of land from the urban middle class to the resident big landlord. It is not difficult to see why these measures failed. In the first place, to legislate is not the same as to enforce. And enforcement is not merely a matter of political will: given the sheer scale of the task, even the strongest political will may not be up to it. To stipulate the land- lord's rental share and share of input costs under varying agronomic conditions and in different regions would require recording, moni- 298 Tenancy Reform toring, supervising, and policing millions of field-level contracts, many of which may have been orally made. The administrative ma- chinery needed for enforcement is almost unimaginable. Second, because rural markets for labor, credit, products, and land are interdependent and interlinked, with landlords acting as agents, any measure to help tenants in land markets at the expense of land- lords can be offset by rearranging contracts in other markets. Where these interlinked contracts are prevalent-and they are particularly so in areas where share tenancy is extensive and markets are least developed-legislation to control one element in a set of linked con- tractual arrangements is an exercise in futility. Policies other than di- rect redistribution of land will leave the welfare of tenants generally unaltered." It is now realized that under these circumstances legisla- tion will fail unless the entire set of contracts is monitored and super- vised. Monitoring all contracts of all kinds seems even more impracti- cal than monitoring all tenancy contracts. Much of course depends on the extent of interlinking and the de- gree of monopoly or monopsony power that landlords exercise in var- ious markets. Empirical research in this area is lacking. But what is surprising is that the most strenuous advocates of legislated ceilings on landlords' rental shares tend also to be those who adamantly main- tain that landlords have broad "exploitative powers" which would nul- lify the effects of the ceilings. Third, as noted earlier, most of these measures assume a neat but false dichotomy between rich, big, "exploitative" landlords and poor, small "exploited" tenants. In reality, millions of smallholders are both tenants and landlords, and in many parts of South Asia relatively rich operators may hold part of their land as tenants (often of poor and small landlords). Any legislation to provide relief to poor tenants would first have to identify those who are both tenants and poor, and then discriminate in their favor. By contrast, blanket ceilings on rents are meaningless if not counterproductive. For example, as was shown in chapter 3, all leased-in operational holdings of fewer than 5 acres in India in 1971-72 accounted for only 36 percent of the total leased-in area. If income gains accruing to tenants are roughly pro- portional to the area under tenancy, then nearly two-thirds of the benefits of such reforms, if effectively implemented, would accrue to tenants outside the target group of the poor. Similarly, since only 20 percent of the area under tenancy in Bangladesh is in this size group, 80 percent of the benefits would accrue outside the target group. In Pakistan less than 3 percent of the tenanted area is in holdings of fewer than 5 acres, so over 97 percent of benefits would have accrued to the nonpoor! Basically, then, tenancy reforms do not represent a well-targeted Land Reform: The Missed Opportunity 299 instrument for redressing rural poverty, and as often as not they are unworkable or perverse in their effects. Experience suggests that ten- ancy reforms per se have been a failure in South Asia and that no amount of tinkering with existing laws-to plug loopholes and im- prove implementation-is likely to work (Appu 1975). Unless the overall balance of economic and political power shifts in favor of the weaker sections of the rural community, further tenancy reform is useless. The greatest shifts are likely to occur where technological change breaks down old class structures and brings about a new align- ment of interests. Measures designed to "give the land to the tiller"- that is, to grant ownership rights in land to persons who work on it as tenants or hired laborers-are unlikely to succeed. The alternatives to interventionist or regulatory reforms are either a laissez-faire pol- icy with no attempt to intervene on tenants' behalf or a genuinely rad- ical decision to make the actual cultivator the owner of the land. Meanwhile, part of the tenancy problem seems to be solving itself as some areas switch from pure sharecropping to partial cost-sharing leases and fixed-rent systems. Technological innovation, especially where the Green Revolution has spread, is helping to transform con- tractual arrangements in the desirable direction of both efficiency and equity. Technological changes have also helped to break down tradi- tional rural class structures, polarizing those who have gained significantly-the well-off and the moderately rich peasantry-and those who have gained less-the small peasantry made up of tenants and landless laborers. The likely outcome of this polarization is not yet clear. In some areas in northwestern India and Pakistan, for ex- ample, a powerful class of middle-income peasants has emerged and the poor remain weak. Elsewhere, as in Tamil Nadu, the organized power and militancy of agricultural laborers have led to action to pro- mote more effective reform (see Byres 1972, 1974, 1977, and 1981 on these issues). It will be through power shifts of these kinds, not through legislation in the name of simplistic abstractions, that the in- stitution of tenancy will evolve. Ceilings on Landholdings and Redistribution of Land Tenancy reform is only one element in the structuralists' agenda for land reform. A second set of proposals addresses the inequitable dis- tribution of landownership-which is less skewed than is often thought, but inequitable nonetheless-and the plight of the landless and those with very little land. Schemes for placing a ceiling on the ownership of land and redistributing holdings above these ceilings are designed to benefit these groups (see especially Dandekar and Rath 1971 on the case for ceilings). 300 Ceilings on Landholdings and Redistribution of Land It is clear that land ceilings and redistribution would promote greater equity. A more equal size structure of holdings would also have beneficial effects on employment and income distribution, al- though the impact on output and productivity may be less desir- able.'6 It would lead to better labor use, the creation of rural assets for the poor, and more appropriate patterns of output and choices of technology; both private and social ends would be well served. Arguments for the benefits of a more equal distribution of land are not new, and measures designed to limit the size of holdings and re- distribute so-called surplus land have had a long history in the re- gion." Despite the enormous volume of legislation on the subject, however, little has in fact been achieved. The Record since Independence Table 8-6 shows the impact of various redistributive measures in the three decades since independence. In Pakistan less than 4 percent of the operated area-only 0.77 million hectares-was acquired by the government under the 1959 Martial Law Regulation 64. Over half of this total was uncultivated wasteland or forest, so only about a quarter million hectares were actually available for distribution.'8 Under the 1972 decrees, the surplus land acquired amounted to only a quarter million hectares (about 1 percent of the operated area); by mid-1974 only half of this had been redistributed.'9 Similarly in Bangladesh (then East Pakistan), only about 200,000 hectares were transferred to tenants and tenants-at-will during the 1960s, although the operated area was around 8.9 million hectares. The land ceiling legislation of 1972 applied to only about 3.3 percent of the operated area; only 0.8 percent was actually acquired, although the availability of additional land as a result of silting meant that some 134,000 hectares, about 1.5 percent of the operated area, were even- tually distributed (see Abdullah 1976 and jannuzi and Peach 1980 for details). In the case of India, estimates of surplus land available for redistri- bution vary according to the source and year.20 Moreover, figures for the different states are hard to obtain or to reconcile. In the sixteen Indian states that implemented land ceiling legislation between 1958 and 1971, however, only 0.99 million hectares had been declared sur- plus by 1971, representing only 0.7 percent of the operated area. Of this total, less than 0.49 million hectares or 0.9 percent of the oper- ated area had been redistributed to poor farm households and land- less laborers (ADB 1977, pp. 100-01). Later figures do not significantly change the picture: by 1977 only some 0.7 percent of the total oper- Land Reform: The Missed Opportunity 301 Table 8-6. South Asia: Impact of Land Redistribution before 1977 (area in thousands of hectares) Operated Estimated as Acquired through Percent Redistributed Percent area surplus legislation (3) ( (1) under reforms (5) 1(1) Total area (1) (2) (3) (4) (5) (6) Bangladesh 1977 8,890 295 69 0.8 134a 1.5 India 1957-77 125,684 2,154b n.a. - 850 0.7 1971c 139,368 988 n.a. - 493 0.4 Pakistan 1959 20,206 2,217 771 3.9 252 1.2 1972 19,854 432 239 1.2 125d 0.6 - Not applicable. n.a. Not available. a. Most of this land was settled with ryots (owner-cultivators) and was acquired as a result of siltation or by means other than legislation. See Abdullah (1976), p. 95. b. Estimates of the surplus land available for redistribution range from 1.7 million to 11.0 million hectares. See India, "'Draft Five-Year Plan 1978-83," vol. 1, p. 29. c. Estimates for 1971 by the ADB (1977), p. 100. d. Sold to landless tenants or to tenants who were small owners. An additional 74,700 hectares were sold to others in 1959. In 1972 the area was allotted without cost. Data for Pakistan have been converted from acres in the original source. Sources: M. H. Khan (1981), pp. 165-78; NSS Report 215, 26th Round, p. 87; India, "Draft Five-Year Plan 1978-83," vol. 1, pp. 29-30, vol. 3, pp. 6-8; Abdullah (1976), pp. 94-96; ADB (1977), p. 100. ated area had been redistributed under land reforms (Dandekar and Rath 197 1). These figures illustrate the abject failure of land reform to match intentions. Meanwhile, population increases and landlords' conceal- ment of holdings are eroding the available surplus. Redistribution has been bedeviled by a whole series of difficulties (many of them im- posed by the relevant legislation). These include absurdly high ceil- ings for permissible holdings, long lists of exemptions, fictitious trans- fers by landholders to avoid the ceilings, and delays as reforms became bogged down in legal maneuvers designed to make them in- effective. The most egregious flaw in the legislation was that the ceilings above which land was to be declared surplus were far too high in rela- tion to the inequities of existing patterns of landownership (see table A44)-although lower ceilings would not have solved all the prob- lems, as shown below. In addition, exemptions to even these high ceil- ings provided loopholes. As a result, the amount of surplus land avail- able for redistribution was too small to secure a meaningful livelihood per capita for the landless. 302 Ceilings on Landholdings and Redistribution of Land In Pakistan, for example, the 1959 land reform set ceilings of 500 acres of irrigated and 100 acres of unirrigated land per person. Al- though these figures were lowered to 150 acres of irrigated and 300 acres of unirrigated land in 1972, this change had little impact be- cause the so-called Bhutto Reforms remained unimplemented. Fur- thermore, exemptions designed to reflect the "productivity of land" in effect permitted an individual to retain far more than the ceiling. In Sind, for example, an owner could keep at least 1,800 acres of irri- gated land of average productivity-a holding nearly four times the theoretical limit of 500 acres. Another 150 acres could be retained for orchards, livestock, and stud farms; land was allowed for house lots, and an individual could transfer large amounts of land to heirs and female dependents. The 1972 legislation lowered the ceilings, but ac- tually gave a bonus of extra land to owners of tractors and tubewells. The new regulations did abolish the exemptions on orchards and also limited intrafamily transfers but, as noted above, their impact has been minimal (M. H. Khan 198 1).21 In Bangladesh the 1950 reform legislation set the ceiling at 100 bighas (33.3 acres) per family or 10 bighas per family member, which- ever was larger. In 1961 the family ceiling was actually raised to 375 bighas (125 acres), although it was brought back down to the 1950 level in 1972. These figures are manifestly absurd in a country where an average household's operated holding is only about 2.5 acres. As in Pakistan, exemptions were also permitted (for plantation crops and mechanized and dairy units). Indian legislation followed the same flawed pattern. Ceilings were set too high in relation to the average operated area per rural house- hold, and various exemptions and loopholes were adopted by individ- ual states. Generally, however, ceilings were lower than in Pakistan and loopholes were fewer. A comparison of ceilings in Pakistan and in the Indian Punjab and Haryana show that those on the Indian side of the border were one-tenth to one-twentieth of those on the Paki- stani side. This fact, together with better implementation in India, has contributed to the substantial differences in landholding structures in the two Punjabs noted in chapter 3. The Indian legislation, however, does seem to have been somewhat of a deterrent; in Pakistan very large units continue to operate with impunity, but in India the threat of legislated ceilings seems to have prevented the further expansion of larger holdings. The failure of land redistribution efforts is not simply a matter of faulty legislation. Even the existing legislation has not been effectively implemented. This is due partly to a lack of political will, but there are other difficulties. Accurate land records are lacking, and local patwaris (record keepers) are open to bribery; land is illegally trans- Land Reform: The Missed Opportunity 303 ferred (with the connivance of the patwaris) into the names of relatives or fictitious (benami) owners to evade the law; legal delays and lawsuits often drag on for decades; compensation for acquired land is inap- propriate (it tends to be set too high for states to afford the land but too low in relation to market values for landowners to find it attrac- tive); and those holding the levers of local power actively oppose the law.22 Further problems arise even after land has been acquired. Much of it is either very poor or even uncultivatable, and landless families have all too often been the least likely recipients of any distributed land-which is not necessarily given away free.23 Sale, rather than free distribution, was required by many of the reform acts that tried to make redistributive measures self-financing (India, Ministry of Ag- riculture and Irrigation 1976). To sum up the position, "the ceiling legislation of the late fifties and early sixties was a great hoax perpe- trated on the land hungry of India" (Parthasarathy 1979, p. 344). Political and constitutional traditions in the region also place severe limitations on radical reform. Private property is protected by the law and cannot be alienated from owners without 'just compensation." The processes of alienation and compensation are subject to endless legal delays-and even if procedures are rationalized, their imple- mentation is far from assured. Alternative Redistribution Schemes Even if redistributive measures were to be fully implemented, how effective would they be in offering the smallest cultivators and the landless the chance to earn a tolerable livelihood from agriculture? Table 8-7 shows the effects of alternative approaches to redistribu- tion, scheme A giving priority to the landless, scheme B giving it to marginal cultivators to bring them up to subsistence levels. (The data sources, assumptions, and implications are discussed in detail in ap- pendix B). The results of the calculations indicate that redistribution would be of little value, even if existing ceilings were reduced. Scheme A, "land to the landless first," appears to be workable only in areas where it is least needed (Pakistan and northwestern India); in areas of real need (Bangladesh and southern and eastern India) there is simply too little surplus land to provide the landless with holdings large enough for basic subsistence. Another disadvantage of this sce- nario is that it would set the poorest against the poor by discriminat- ing against the latter. A major drawback is that possession of land alone is not enough to ensure a livelihood from agriculture; comple- mentary inputs (water, seeds, draft power) are also needed. More- over, labor previously hired out by landless households would now Table 8-7. South Asia: Alternative Land Redistribution Scenarios Scheme A Scheme B Landless labor Below subsistence farmers Landless labor Below subsistence farmers Country and region Thousands Percent Thousands Percent Thousands Percent Thousands Percent c Bangladesh Case 1 71 2 0 0 173 7 0 0 Case 2 2,763 84 0 0 2,451 100 75)1 53 India 6,377 44 4,612 12 9,937 84 2,963 21 Jammu and Kashmir 13 100 301 100 310 100 13 100 Punjab 414 100 54 100 54 100 414 100 Haryana 174 100 72 100 72 100 174 100 Uttar Pradesh 108 8 0 0 159 2 0 0 North India 709 36 427 5 586 7 601 30 Rajasthan 111 100 857 100 851 100 111 100 Gujarat 468 55 0 0 664 31 0 0 Madhya Pradesh 781 100 1,846 100 1,846 100 781 100 Maharashtra 394 57 0 0 863 25 0 0 West India 1,954 70 2,697 33 4,224 51 892 32 Karnataka 438 49 0 0 569 19 0 0 Kerala 21 8 0 0 25 1 0 0 Tamil Nadu 788 38 0 0 1,584 100 155 7 Andhra Pradesh 1,944 100 1,488 68 2,187 100 1,315 68 South India 3,191 62 1,488 16 4,365 47 1,470 28 Orissa 280 67 0 0 441 23 0 0 West Bengal 151 7 0 0 198 5 0 0 Bihar 90 6 0 0 120 2 0 0 Assam 2 1 0 0 2 ... 0 0 East India 523 12 0 0 761 6 0 0 Pakistan 2,000 100 720 100 720 100 2,000 100 ... Negligible. Notes: This table presents the results of alternative redistribution schemes. Scheme A: The surplus land available for redistribution is allocated first to landless labor households in amounts large enough to produce subsistence requirements, and then if any land remains it is allocated to households cultivating below-subsistence holdings so as to bring them up to subsistence requirements. Scheme B: The available surplus is first allocated to cultivating households with holdings below subsistence levels so as to bring them up to subsistence requirements; any land that remains is then allocated to landless labor households. The number and percentage of households in each group likely to benefit under both schemes are given. Sources: The data, assumptions, and sources are discussed in appendix B. 306 Ceilings on Landholdings and Redistribution of Land need to be devoted to farming the household plot, so incomes would be depleted by the wage earnings forgone. Scheme B, "land to marginal cultivators first," should face fewer difficulties with the need for complementary inputs because this group is likely already to have access to at least some inputs. But there is still the drawback that household members who formerly hired out their labor would have to divert some of their effort into work on the newly acquired land. Furthermore, this scheme is open to the objec- tion that its primary target group is not the very poorest (the landless) but a group that already has some land; this priority may make eco- nomic sense but would be politically controversial. The critical objec- tion, as with the first alternative, is that the available land would be insufficient to satisfy the objective of the scheme-in this case, raising all marginal cultivating households to subsistence levels. In principle, if the ceilings on holdings were lowered, more surplus land ought to be made available for redistribution. Can ceilings in fact be lowered further? Realistically, it is hard to avoid the conclusion that, if the present ceiling levels have not been enforced, there are even slimmer chances of lower ones coming into effect.24 Lower ceil- ings are possible in Pakistan and some states in northern and western India; where they are being lowered, however, the results are not quite what might be expected by advocates of land reform as a solu- tion to rural poverty. In 1972 the Indian state of Maharashtra enacted legislation under which new surplus lands were acquired in 1975 and distributed to the landless poor in 1976. But as Pardeshi and Kazar (1981) have shown, the effect on the prospects of the poor was marginal. Although their average net sown area was around 2.5 acres (and acquired land raised the value of their assets threefold), crop incomes remained insignifi- cant. The contribution of newly allotted land to total income was mar- ginal (around 5 percent) because this land was of very poor quality; meanwhile, lacking essential inputs for cultivation, beneficiaries con- tinued to rely mainly on wage income for their livelihood. Animal husbandry showed real improvement: livestock incomes increased by 130 percent. But by and large the landless poor who were now given access to land remained what they had been before-poor and mainly dependent on wage income. Two specific lessons seem to emerge: (a) even if enough surplus land were available, redistribution (especially if the land is poor) is not helpful unless other inputs-draft animals, irrigation, credit, fertilizers-are provided; and (b) the distribution of small parcels of poor land could, however, open up prospects for livestock and other ancillary activities that might make better use of the land than crop cultivation. Land Reform: The Missed Opportunity 307 More generally, the record of how little has actually been achieved by land ceilings and redistribution, together with the discussion in ap- pendix B of the two hypothetical examples shown in table 8-7, sug- gests that radical measures for the redistribution of land are not the panacea for poverty they are often believed to be. What Dandekar and Rath (1971, p. 106) correctly concluded a decade ago is even more true today: "A major means of production in the Indian econ- omy is land. Therefore a suggested remedy to the problem of poverty is to redistribute available land among all those who depend on it. But there is not enough land to redistribute so that everyone may employ himself on his land and earn a minimum desirable living." So, how- ever simple it may appear, it is futile to attempt to resolve the problem of rural poverty in an overpopulated land by redistributing land that is in short supply. Moreover, as the hypothetical examples show, land surpluses and the potential beneficiaries of redistribution are not themselves distrib- uted evenly over the subcontinent; available data suggest that redistri- bution seems to be feasible where it may not be necessary, and less feasible where it seems more urgently needed. It can no longer be argued that the redistribution of land into smaller holdings would in- crease productivity. Evidence cited in chapter 4 suggests productivity may actually fall if holdings are made smaller unless other market and institutional biases are also corrected. Does this mean that all thoughts of land redistribution should be set aside, at least in states and regions where there is not enough land to distribute into viable holdings for everyone? Not necessarily: some redistribution, even of land parcels too small to form viable units for crop farming, could still be valuable. In particular, a relatively modest redistribution program could usefully give the landless plots for simple homesteads and space for small-scale vegetable or livestock farming. Giving the millions who do not have a roof over their heads some land on which to build shelter seems to be a worthwhile and achiev- able goal. In Bangladesh in 1978 nearly 3.5 million rural households (out of a total of 12 million) owned either no land or only enough for their houses. Nearly half did not have any land at all, not even house lots. I have no comparable figures for India and Pakistan, but if the same ratio is used, landless families without homesteads would be estimated at 7.5 million in India and 1 million in Pakistan in 1980. Serious consideration should be given to formulating and imple- menting homestead legislation that would provide enough land for the landless at least to build themselves shelter and have a small vege- table garden or space to rear small stock and poultry. Of course land alone will not be enough; money and effort will be needed to help 308 Consolidation of Fragmented Holdings those who are given land to build low-cost housing and learn to grow vegetables and rear small stock. However flimsy or poorly constructed the houses, they would be better than what most of the landless have now. Moreover, even the simplest homestead arrangement would provide an asset base whose value would rise over the long term. The amount of land needed for a homestead is very small. Even small amounts of land can be effectively used for activities such as vegetable gardening, dairying, poultry, and fisheries-all of which could greatly increase the incomes of the landless over their present levels. For dairying, as little as a quarter or a half acre can provide fodder for a cow; one-eighth of an acre is enough to set up a poultry unit of economic size; in fisheries, a quarter-acre pond represents a large unit if farmed intensively. A vegetable plot of a quarter acre can yield a large income. Realistically, redistribution of the surpluses available cannot pro- vide holdings large enough to support even subsistence food-grain farming for the landless and marginal cultivators. Once the focus shifts to ancillary activities, however, the redistribution of land parcels as small as one-eighth to one-quarter acre in size begins to make sense. In some cases-poultry and fisheries for example-even poor land can be effectively used; locational factors (distance from mar- kets, roads, electricity supplies, extension services) are probably a more important consideration. In the case of dairying and vegetable farming, although land quality remains important, other factors such as credit, marketing, and extension are probably critical. Thus a land redistribution scheme that provides a minimum amount of land-say, a quarter to a half acre per household-is both desirable and feasible everywhere. Such a scheme will not make the landless into self-sufficient farmers, as past reforms have misleadingly promised and fruitlessly attempted to do, but it will give them access to sources of income other than casual wage labor and the opportu- nity to build shelter for their families. Unlike the grandiose abstrac- tions of the radical structuralists, this seems an achievable goal that would give all poor rural families some modest but tangible gains. Consolidation of Fragmented Holdings The concentration of holdings in the hands of relatively large opera- tors is not the only troublesome feature of landholding arrangements in South Asia. The fragmentation of even the smallest holdings se- verely constrains the evolution of productive peasant agriculture. A fragmented holding consists of physically scattered parcels of land op- erated by one household; if they are widely scattered, the individual parcels define the scale of farming operations regardless of the size Land Reform: The Missed Opportunity 309 of the total holding. As has rightly been pointed out, "For a country's agricultural economy, whether owner- or tenant-operated, fragmen- tation is an unmitigated evil for which no advantages can be claimed" (Ladejinsky 1977, p. 374). Fragmentation has occurred primarily through subdivision. Under the traditional system of inheritance, landholdings have been subdi- vided among heirs. Inheritances may include pieces of land from dif- ferent parts of the original holding, so as to even out any inequities in land quality and productivity. In 1928 the Royal Commission on Agriculture had this to say: "In extreme cases the result is ludicrous . . . the size of individual plots is sometimes as small as 1/160th of an acre . . . fields have been formed a mile long and but a few yards wide, while areas have been brought to notice where fragmentation has been carried so far as to prevent all attempts at cultivation."25 The severity of the problem of fragmentation in the three countries can be seen in tables 8-8 to 8-10 and figure 8-1. In 1960-61 the av- erage operational holding in India consisted of six parcels of land. In Bangladesh more than half the farms reported six or more frag- ments in 1960, and by 1977 the average number of fragments had risen to ten, with over 70 percent of the farms reporting more than seven fragments on average. Data from Pakistan for 1960 and 1972 suggest that fragmentation may be becoming more acute as a smaller percentage of farms report unfragmented areas. Fragmentation usually increases with the size of operational hold- ings, but even the very smallest holdings are highly fragmented. In India holdings smaller than 5 acres contain an average of two to five parcels of land, each with an average area of 0.1 to 0.4 acres. In Ban- gladesh more than half of all holdings smaller than 0.5 acres reported two to three fragments or more, as did 90 percent of holdings of 0.5-1.0 acres. A somewhat similar picture emerges for Pakistan, espe- cially for small and marginal farms (see M. H. Khan 1979b). Regional data for India show that fragmentation is higher than the Indian average in the eastern states of Bihar, Orissa, Uttar Pradesh, and West Bengal, with holdings of 2.5-4.9 and 5-7.5 acres recording as many as eight to ten parcels each (see table A45). In these states the countryside is a patchwork of tiny parcels of 0.1 to 0.2 acre on which millions of small cultivators-by no means all of them owners-eke out a living. Fragmentation leads to economic inefficiency and depresses pro- ductivity. Apart from the time lost in moving human and animal labor from one parcel to another, scarce land is wasted in the construction of boundaries and irrigation channels between fields. Hired labor and land-labor contracts become hard to supervise. Above all, however, 310 Consolidation of Fragmented Holdings Table 8-8. India: Average Number and Size of Fragments in Operational Holdings, by Size of Holding, 1960-61 Fragments per holding Size of Estizmated Estimated area operated Averag operational number of Average holding holdings Percentage Acres per Average acres per (acres) (millions) of total holding number fragment <0.5 4.3 0.3 0.2 1.8 0.1 0.5-1.0 4.3 0.9 0.7 3.1 0.2 1.0-2.5 11.1 5.6 1.6 4.4 0.4 2.5-5.0 11.4 12.3 3.5 6.0 0.6 5.0-7.5 6.5 11.7 5.9 6.8 0.9 7.5-10.0 3.5 9.0 8.4 7.6 1.1 10.0-12.5 2.6 8.2 10.6 7.6 1.4 12.5-15.0 1.5 5.9 13.3 8.0 1.7 15.0-20.0 1.9 9.6 16.6 7.9 2.1 20.0-25.0 1.2 7.4 21.0 8.8 2.4 25.0-30.0 0.7 5.3 26.3 8.0 3.3 30.0-50.0 1.1 12.0 35.8 8.1 4.4 50.0+ 0.5 11.6 73.4 9.4 7.8 Total 50.7 100.Oa 6.5 5.7 1.1 Notes: Columns may not add to total because of rounding. a. Estimated total area operated is 329.6 million acres. Source: NSS data. fragmentation makes it extremely difficult to use irrigation water ef- fectively or to implement measures for water conservation. Even small tubewells or mechanized tillers are uneconomic on many small holdings, and mobile units become harder to manage. In a broader context it becomes very difficult to carry out any rational plan for effi- cient exploitation of land and water resources within a watershed. Consolidation of fragmented holdings is thus essential for improv- ing the efficiency and productivity of small-farm agriculture. The small farmer needs a compact area equal in value as well as in size to his former patchwork of scattered plots. Consolidation is not just a matter of physical regrouping of plots, however; it is a complex and difficult process, requiring villagers to make voluntary agreements (at best difficult) to undertake consolidation on an equitable basis. This in turn entails a conviction on the part of the farmer that in the process of consoli- dation (correction of land records, land measurements, land classi- fication, land valuation, and reassignment of the consolidated land) his interest will be protected. The touchiest of questions is land val- Table 8-9. Bangladesh: Agricultural Land Holdings and Fragmentation per Holding by Farm Size, 1977 Fragments per holding Size Holdings Acreage Average Average acres (acres) Number Percent Number Percent Average number per fragment 0.0-0.5 342,040 5.5 109,009 0.5 0.319 2.9 0.109 0.5-1.0 648,302 10.4 472,029 2.1 0.730 4.4 0.167 1.0-1.5 799,282 12.8 961,580 4.4 1.203 5.7 0.209 1.5-2.5 1,322,258 21.1 2,574,475 11.7 1.947 7.5 0.260 2.5-5.0 1,829,458 29.2 6,402,140 29.2 3.499 10.5 0.334 5.0-7.5 726,272 11.6 4,335,620 19.7 5.970 14.2 0.420 7.5-10.0 269,296 4.3 2,279,487 10.4 8.465 17.4 0.488 10.0-12.5 140,796 2.3 1,540,086 7.0 10.938 20.1 0.545 12.5-15.0 62,901 1.0 849,186 3.9 13.500 21.8 0.620 15.0-25.0 93,357 1.5 1,691,215 7.7 18.116 25.9 0.699 25.0+ 23,211 0.4 744,664 3.4 32.082 32.8 0.977 Total or average 6,257,173 100.0 21,959,476 100.0 3.509 9.6 0.367 Source: Bangladesh Bureau of Statistics (1977). 312 Consolidation of Fragmented Holdings Table 8-10. Pakistan: Fragmentation of Farms by Size of Operational Holding, 1960 and 1972 Size of Total Unfrag- Total Percentage Percentage offarms reporting operational area mented number of of unfrag- holding (millions of area farms mented 2-3 4-5 6+ (acres) acres) (percent) (millions) farms fragments fragments fragments 1960 <1.0 0.334 70 0.742 77 21 2 1.0-2.5 1.345 51 0.856 55 34 8 3 2.5-5.0 2.911 37 0.806 37 37 14 12 5.0-7.5 3.546 32 0.581 32 35 15 18 7.5-12.5 7.357 25 0.759 26 36 15 23 12.5-25.0 12.533 20 0.729 21 35 17 27 25.0-50.0 9.468 13 0.286 14 29 21 36 50.0-150.0 6.539 8 0.088 9 19 19 53 150.0+ 4.896 5 0.014 6 17 15 62 All 48.929 19 4.860 39 33 12 16 1972 <1.0 0.076 66 0.154 73 24 3 1 1.0-2.5 0.563 55 0.351 56 33 8 3 2.5-5.0 1.859 45 0.527 45 35 12 8 5.0-7.5 3.349 40 0.566 41 39 10 10 7.5-12.5 8.732 33 0.903 34 45 11 10 12.5-25.0 12.778 27 0.778 29 41 15 11 25.0-50.0 8.832 19 0.277 21 41 19 19 50.0-150.0 6.848 12 0.095 13 37 22 29 150.0+ 3.594 9 0.014 10 27 20 41 All 46.672 25 3.664 38 40 12 11 a. Less than 0.5 percent. Sources: 1960 Pakistan Census of Agriculture: A Summary of West Pakistani Data (Ka- rachi: Agricultural Census Organization), p. 63; 1972 Pakistan Census of Agriculture: All Pakistan Report (Labore: Agricultural Census Organization, 1975), p. 4. uation. While it is a truism to say that plots of land vary greatly in quality of soil, facilities of irrigation, productivity, and distance from the village, it is a Herculean task to evolve methods of valua- tion by which different classes of land can be reduced to a compara- ble basis. Unless valuation of land satisfies the community, consoli- dation cannot succeed. (Ladejinsky 1977, p. 374) Even the correction of land records-another prerequisite for suc- cessful consolidation-is likely to be enormously difficult, since rec- Land Reform: The Missed Opportunity 313 Figure 8-1. India and Bangladesh: Average Number of Fragments per Holding by Size of Holding Number offragments 40 [ India, 1960-61 Bangladesh, 1977 ;O 0 0 S :5< 55550.00 66.7 - - - Gujranwala 97.2 50.3 < 12.50 98.8 48.7 - - 12.50-25.00 94.9 36.5 - - 25.00-50.00 98.0 46.3 - - >50.00 95.1 63.2 - - Sahiwal 89.1 8.7 92.6 100.0 <12.50 83.3 1.3 91.6 100.0 12.50-25.00 82.0 5.8 87.6 100.0 25.00-50.00 84.1 16.0 93.5 100.0 >50.00 94.8 7.2 98.0 100.0 Lyallpur 100.0 - 100.0 97.8 <12.50 100.0 - 100.0 87.0 12.50-25.00 100.0 - 100.0 94.8 25.00-50.00 100.0 - 100.0 100.0 >50.00 100.0 - 100.0 100.0 Rahimyar Khan 79.5 1.9 83.7 - <12.50 65.8 - 74.8 - 12.50-25.00 70.7 33.3 82.1 - 25.00-50.00 67.4 - 89.8 - >50.00 96.1 - 82.6 - Punjab 91.7 31.0 93.7 98.8 Jacobabad 64.5 15.8 100.0 - < 12.50 - 2.0 - - 12.50-25.00 44.4 - - - 25.00-50.00 72.3 18.2 - - >50.00 70.4 21.0 100.0 - Larkana - 98.1 - - < 12.50 - 100.0 - - 12.50-25.00 - 100.0 - - 25.00-50.00 - 97.7 - - >50.00 - 97.6 - - Statistical Appendix 359 Table A31 (continued) District and farm size Mexi-Pak IRRI Improved Improved (acres) wheat rice cotton maize Nawabshah 46.1 - 91.8 <12.50 29.6 - 73.4 - 12.50-25.00 64.8 - 79.9 - 25.00-50.00 48.5 - 100.0 - >50.00 42.2 - 93.5 - Hyderabad 100.0 95.9 88.2 <12.50 100.0 93.1 90.7 - 12.50-25.00 100.0 98.7 87.8 - 25.00-50.00 100.0 92.8 82.9 - >50.00 100.0 94.8 89.2 - Sind 89.6 61.4 90.8 - Not applicable. Source: M. H. Khan (1975), p. 20. Reprinted with permission. Table A32. Bangladesh: HYV Paddy Adoption by Farm Size, Four Villages Index of Size of holding Number of Adopters' HYV areaa participation (acres) holdings (1) (2) (1) x (2) Very small (< 1) 127 25.2 31.3 7.8 Small (1-3) 267 53.9 22.6 11.9 Lower medium (3-5) 168 63.4 18.4 11.7 Upper medium (5-7.5) 46 73.9 14.3 10.6 Large (7.5+) 20 50.0 14.3 7.2 All 628 52.0 18.5 9.6 a. As a percentage of all paddy cultivators or total paddy area. Source: Asaduzzaman and Islam (1976). Table A33. Eastern India: Rice Yields by Farm Size in Three Districts, 1971-72 (kilograms per hectare) Burdwan Shahabad Sambalpuar Farm size Percentage Percentage Farm size Percentage (hectares) Local HYV increase Local HYV increase (hectares) Local HYV increase 0.01-1.0 2,152 3,543 64.6 1,977 3,163 60.0 0.1-2.50 1,483 3,875 161.3 1.01-2.0 2,686 4,117 53.3 1,636 - - 2.51-5.00 1,483 3,800 156.2 2.01-3.0 2,392 4,186 75.0 1,648 2,735 66.0 5.01-7.50 - 3,753 - 3.01-4.0 2,471 3,370 36.4 1,480 2,511 70.0 7.51-10.0 2,224 3,845 72.9 4.01-6.0 2,236 4,193 87.5 1,510 2,382 57.8 10.01-15.00 1,977 2,849 44.1 6.01-8.0 2,511 3,188 27.0 1,761 2,261 28.4 15.01-21.00 - 3,489 - 8.01-10.0 - - - 1,441 2,459 70.6 21.01-25.00 - 3,123 - 10.01-12.0 - - - 1,574 2,337 48.5 25+ 2,471 4,359 76.4 All 2,414 3,926 62.6 1,587 2,434 53.4 All 2,152 3,558 65.3 - Not applicable. Source: Mandal and Ghosh (1976). Statistical Appendix 361 Table A34. India: Application of Chemical Fertilizer on HYV Wheat Crops by Size Group, 1967-69 (kilograms per acre) Size group State and district I If III IV Punjab Amritsar 260 213 243 232 Ferozepur 155 135 133 131 Ludhiana 250 278 301 254 Patiala 187 172 227 183 Haryana Hissar 204 167 197 159 Rohtak 105 134 99 154 Uttar Pradesh Aligarh 93 123 147 148 Allahabad 253 277 279 284 Basti 98 154 229 346 Muzaffarnagar 123 237 240 292 Saharanpur 207 269 272 275 Sitapur 145 139 163 269 Rajasthan Sriganganagar 116 156 176 252 Bihar Gaya 122 126 140 144 Note: The chemical fertilizer consists of nitrogen, phosphorus, and potash. The size groups are as follows: 1, the smallest 30 percent (deciles 1-3); II, the next 30 percent (deciles 4-6); 111, the next 30 percent (deciles 7-9); IV, the largest 10 percent. Source: India, Planning Commission (1971). 362 Statistical Appendix Table A35. India: Application of Chemical Fertilizer on HYV Paddy Crops by Size Group, 1967-69 (kilograms per acre) Size group State and district I 11 III IV Tamil Nadu Coimbatore 286 182 163 225 North Arcot 48 156 100 90 Thanjavur 30 51 85 66 Karnataka Shimoga 189 129 185 135 Orissa Cuttack 0 123 11 116 Sambalpur 159 168 377 207 West Bengal Burdwan 100 139 84 129 Hooghly 94 162 148 191 Midnapur 105 169 182 307 Kerala Palghat 50 158 156 80 Trichur 34 111 65 297 Andhra Pradesh Krishna 224 270 208 442 Nellore 0 0 173 128 Nizamabad 344 210 266 238 West Godavari 251 149 181 159 Uttar Pradesh Basti 164 247 300 259 Varanasi 8 161 221 27 Punjab Amritsar 0 279 284 116 Bihar Gaya 92 48 76 167 Shahabad 169 232 220 173 Maharashtra Bhandara 11 52 71 107 Thana 72 109 87 103 Note: The chemical fertilizer consists of nitrogen, phosphorus, and potash. Size groups are defined in table A34. Source: India, Planning Commission (1971). Statistical Appendix 363 Table A36. India: Percentage of Fertilizer to Total Cropped Area, 1975-76 Size of farm (hectares) Zone and state <1 1-2 2-4 4-10 10+ All Northern Haryana 45.0 46.7 38.6 47.8 64.2 48.7 Himachal Pradesh 21.1 32.6 28.9 31.2 70.0 27.6 Jammu and Kashmir 38.9 27.1 24.8 19.7 - 28.5 Punjab 63.4 69.0 74.9 78.8 76.8 76.3 Western Gujarat 45.9 39.9 46.7 41.9 43.4 43.1 Maharashtra 28.0 32.8 25.1 27.8 26.5 27.3 Central Madhya Pradesh 2.9 5.1 8.9 11.6 15.6 10.8 Rajasthan 11.8 19.4 24.0 25.3 16.2 20.1 Uttar Pradesh 22.7 28.6 37.8 52.0 54.5 52.1 Eastern Assama 5.4 3.8 3.5 3.4 57.5 4.9 Bihar 24.1 34.2 40.3 37.8 36.7 35.3 Orissa 8.3 15.7 19.0 23.6 47.1 20.7 West Bengal 47.8 47.6 53.9 49.5 15.7 49.8 Southern Andhra Pradesh 47.5 46.7 49.2 38.8 32.7 41.7 Karnataka 46.0 43.0 40.4 26.6 26.5 33.4 Kerala 69.4 78.9 69.6 100.0 - 72.6 Tamil Nadu 60.6 53.9 56.5 51.4 37.8 55.4 - Not applicable. a. Includes Manipur, Meghalaya, and Tripura. Source: NCAER (1978). 364 Statistical Appendix Table A37. India: Fertilizer Use by Size of Holding in Various States, 1975-76 (kilograms per hectare) Size offarm (hectares) Zone and state <1 1-2 2-4 4-10 10+ All Northern Haryana 54.6 64.2 57.0 74.5 98.2 76.6 Himachal Pradesh 41.7 26.0 20.9 22.3 17.3 28.5 Jammu and Kashmir 45.6 50.8 47.0 34.4 - 47.0 Punjab 88.1 80.3 90.9 90.8 93.6 90.8 Western Gujarat 78.7 64.7 59.7 43.1 34.7 45.8 Maharashtra 90.5 95.8 85.4 75.3 63.3 77.3 Central Madhya Pradesh 72.1 66.9 55.4 41.3 41.9 46.5 Rajasthan 59.2 55.6 57.9 59.0 48.1 55.5 Uttar Pradesh 73.5 66.0 65.4 65.6 47.7 64.6 Eastern Assama 69.1 51.2 56.2 50.1 15.4 49.4 Bihar 64.7 50.8 45.7 45.4 49.3 49.7 Orissa 86.9 71.5 77.6 97.3 111.5 90.8 West Bengal 100.0 103.6 77.3 65.9 177.5 89.5 Southern Andhra Pradesh 109.0 118.5 115.9 117.4 85.6 111.7 Karnataka 169.1 131.7 106.3 97.4 39.3 104.6 Kerala 93.4 88.6 73.4 171.4 - 92.0 Tamil Nadu 133.5 133.0 122.5 120.5 127.3 128.1 - Not applicable. a. Includes Manipur, Meghalaya, and Tripura. Source: NCAER (1978). Table A38. India: Percentage of Area Fertilized and Fertilizer Use per Hectare for Irrigated Paddy by Size of Holding in Selected States Size of holding (hectares) < 1 1-2 10+ Crop Seasona State Useb Percentc Useb Percent' Useb Percentc Irrigated HYV paddy K Haryana 100.0 88 94.4 98.0 119.8 99.0 K Punjab 103.1 57 87.3 72.0 114.6 100.0 K Gujarat 62.6 81 n.a. n.a. 203.4 100.0 K Uttar Pradesh 47.5 27 44.8 44.0 85.1 99.0 K(A) Bihar 60.8 58 71.0 93.0 35.9 100.0 K(A) West Bengal 121.7 100 117.4 96.0 133.2 100.0 K(A) Andhra Pradesh 114.8 62 121.3 80.0 159.5 100.0 K(A) Tamil Nadu 124.3 86 130.7 96.0 297.3 100.0 K(W) Bihar 65.2 87 49.7 93.0 33.9 100.0 K(W) Andhra Pradesh 131.2 61 114.8 86.0 159.5 67.0 K(W) Tamil Nadu 163.4 96 148.6 99.0 75.8 100.0 R(S) Andhra Pradesh 136.2 65 187.8 77.0 135.6 100.0 R(S) Tamil Nadu 114.7 99 146.3 100.0 18.3 100.0 R(S) Orissa 181.3 98 174.1 100.0 243.2 100.0 Irrigated non-HYV paddy K Haryana 63.6 83 39.9 86.0 79.0 100.0 K Gujarat 60.1 71 68.6 61.0 50.1 100.0 K Maharashtra 105.3 64 71.2 62.0 49.5 22.0 K Andhra Pradesh 34.6 1 78.8 23.0 29.7 72.0 K Rajasthan 143.3 6 63.1 32.0 56.8 100.0 K Uttar Pradesh 53.7 22 47.7 32.0 30.7 5.0 (Table continues on the following page.) Table A38 (continued) Size of holding (hectares) < 1-2 10+ Crop Seasona State Useb Percent' Useb Percent' Useb Percentc Irrigated non-HYV paddy K(A) Andhra Pradesh 109.9 85 73.3 63.0 80.8 100.0 K(A) Karnataka 194.8 86 170.7 88.0 95.9 100.0 K(W) Bihar 64.0 37 44.2 54.0 67.5 70.0 K(W) Orissa 42.3 17 42.3 14.0 28.7 94.0 K(W) Andhra Pradesh 90.0 68 102.6 82.0 93.7 60.0 K(W) Tamil Nadu 162.0 67 142.5 78.0 140.8 100.0 R(S) Andhra Pradesh 117.5 85 108.5 78.0 26.7 100.0 R(S) Tamil Nadu 147.5 82 130.2 97.0 156.1 100.0 g> Unirrigated non-HYV paddy K Gujarat 98.6 33 38.4 55.0 18.7 86.0 K Maharashtra 80.3 33 48.6 36.0 63.2 68.0 K Madhya Pradesh 54.4 1 22.5 4.0 23.8 33.0 K Uttar Pradesh 21.8 12 30.2 2.0 34.3 15.0 K(A) Orissa 24.4 8 27.5 11.0 6.9 17.0 K(A) Karnataka 90.7 21 76.7 21.0 114.9 100.0 K(A) Tamil Nadu 77.0 32 47.1 15.0 25.2 100.0 K(W) Bihar 33.6 13 25.0 30.0 9.2 27.0 K(W) Orissa 29.8 2 34.1 12.0 65.5 21.0 K(W) Karnataka 79.7 98 51.9 100.0 21.1 100.0 n.a. Not available. a. K, kharif; A, autumn; W, winter; S, summer; R, rabi. b. Fertilizer use in kilograms per hectare. c. Percentage of area fertilized. Source: NCAER (1978). Table A39. Pakistan: Use of Fertilizers, 1974 (bags per acre) Mexi-Pak wheat crop Size of holding (acres) Jhelum Gujranwala Sahiwal Lyallpur Rahimyar Khan Nawabshah Hyderabad <12.5 0.7 1.7 0.9 2.5 1.7 1.3 12.5-25.0 1.0 1.6 1.5 1.9 1.6 2.3 25.0-50.0 - 1.4 1.6 2.1 2.4 1.6 1.7 50.0+ 0.1 1.0 1.7 2.0 3.6 1.6 2.6 IRRI rice crop Gujranwala Sahiwal Rahimyar Khan Jacobabad Larkana Hyderabad <12.5 0.5 1.0 n.a. 0.5 1.4 0.7 12.5-25.0 1.2 2.2 2.0 0.8 1.5 0.8 25.0-50.0 1.0 1.6 n.a. 0.8 1.4 1.4 50.0+ 1.0 2.2 n.a. 0.4 1.4 1.4 -Not applicable. n.a. Not available. Source: M. H. Khan (1975), pp. 160-77. Reprinted with permission. 368 Statistical Appendix Table 40. India: Poverty, Production, and Prices in Rural Punjab, including Haryana Per capita Percentage of population below food-grain the poverty line production CPIALb Year Below WVell below Far below (kilograms)a (1960-61 = 100) 1963-64 39.49 31.42 20.63 346 114 1964-65 41.12 30.35 18.57 422 139 1965-66 40.89 32.93 19.06 313 138 1966-67 38.52 30.53 18.78 386 174 1967-68 44.58 36.18 24.02 527 193 1968-69 34.03 26.22 15.13 512 193 1969-70 37.86 30.25 16.19 627 196 1970-71 35.44 27.42 13.80 648 194 1971-72 31.31 24.60 15.26 662 205 1972-73 25.18 18.64 9.52 616 228 1973-74 34.00 25.26 13.79 597 273 a. Per capita food-grain production has been calculated by dividing total estimated food-grain production of Punjab (including Haryana) by its total rural population. b. Consumer price index for agricultural laborers. Source: Mundle (1982). Statistical Appendix 369 Table A41. India: Annual Income of Agricultural Labor Households in Punjab, including Haryana Agricultural labor households 1956-57 1964-65 1974-75 Money income (current rupees) With land 626 987 4,773 Without land 656 917 3,388 All households 731 928 3,508 Real income (rupees at 1956-57 prices) With land 626 705 1,404 Without land 656 655 996 All households 731 663 1,032 Average size of household With land n.a. 5.9 6.6 Without land n.a. 5.4 5.6 All households 5.2 5.5 5.7 Per capita real income (rupees at 1956-57 prices) With land n.a. 119 213 Without land n.a. 121 178 All households 141 121 181 n.a. Not available. Note: The annual income of all households in 1956-57 is larger than for households with and without land because the latter pertain to casual labor households only. The real incomes have been calculated using the consumer price index for agricultural la- borers as deflator and 1956-57 as base. For 1974-75 Punjab and Haryana data have been combined using the number of households in the relevant category as Weights. Source: Mundle (1982). Table A42. South Asia: Summary of Land Reform Legislation Region Major legislation and period Summary enactment and salient features North India Land Reform Act (1977) Ceiling (100-200 acres); compensation for resumption of land Uttar Pradesh Zamindari Abolition Acts (1950-58) Zamindars compensated up io eight times net assets with rehabili- tation grant Punjab and Four acts (1952-54) Approximately 650,000 occupancy tenants acquired 2 million acres CZ° Haryana with the abolition of rent-receiving intermediaries and compen- sation of landowners Himachal Punjab and East Punjab States Union Acts Government compensated zamindars according to land revenue, Pradesh (1952-54) and another in 1971 conferring proprietary rights on tenants West India Rajasthan Land Reform and Resumption of Jagirs Act Problematic implementation in princely and religious estate hold- (1952) with subsequent amendments ings; staged resumption of all jagirs; erstwhile tenants elevated through registration Maharashtra Series of seventeen acts (1949-58) Abolition of ryots and various kinds of land grants; compensation dependent upon intermediary interests forgone Madhya Abolition of proprietary rights (1950) Zamindari and jagirdari abolished; staged compensation with Pradesh minimal or no rehabilitation grant Gujarat Numerous acts for Saurashtra and Bombay Abolition of intermediaries, upgrading of tenant cultivators to states (1949-62) occupancy status with no payments for occupancy rights South India Tamil Nadu Series of twelve acts (1948-72) Abolition of intermediaries with multiple compensations based on returns from estates Pakistan Muslim League Agrarian Committee (1949) Abolition of jagirs; ownership to tenants; security of tenure; fix- and Punjab and Sind Tenancy Acts (1950) ation of rent; land over ceilings (150-450 acres) resumed by state with compensation; redistribution to tillers and smallholders (<12.5 acres) Land Reform Regulation no. 64 (1959) Ceiling (500-1,000 acres) based on land productivity; resumption and redistribution of excess land; abolition of jagirs; tenant security; consolidation of holdings Land Reform Regulation no. 115 (1972) Ceiling (150-300 acres) based on productivity; no eviction with- out cause, no compensation for resumed land; tenants have first choice on sale Note: A ryot is a landowner who pays land taxes or rent directly to the state. A jagir is a land grant awarded by states for services rendered. Sources: M. H. Khan (1981); Abdullah (1976); India, Ministry of Agriculture and Irrigation (1976). Table A43. South Asia: Summary of Provisions on Rent Control and Security of Tenure State Fixation of rent Security of tenure Bangladesh Yes Government acquires rent-receiving interests for tenants-at-will. North India Jammu and 1/4 of produce on wetland, 1/3 of produce on Tenants have secure tenure subject to landlord's right of re- Kashmir dryland if landlord holds under 12.5 acres sumption. Tenants in possession have right of purchase. Himachal 1/4 of produce Tenants whose landlords pay more than Rs125 a year in revenue Pradesh have security. Landlord resumption privilege has expired. Non- occupancy tenants of small landowners have little security. Punjab 1/3 of gross produce Security of tenure subject to landlord right of resumption (expired) leaving tenant 5 standard acres. Tenants have purchase option in some areas. Uttar Pradesh No; no official leasing although bataia is prac- Tenancy abolished except for those with batai, who have no ticed, and payment is 1/2 of output security of tenure. \ West India 4 Gujarat Lesser of 1/6 of gross produce or 2 to 5 times Tillers deemed to be owners. Many unable or unwilling to assessment purchase at 20 to 200 times assessment. Resumption for personal cultivation by landlord restricted to half the holdings except for surrendered land. Rajasthan 1/6 of gross produce Some tenants have fixity of tenure. Security of tenants of khudkasht (personal land of landlords) and subtenants subject to resumption (expired), which left tenant 15.6 to 125 acres. Madhya 2 to 4 times assessment All tenants are occupancy tenants. Tenure is secure but subject Pradesh to landlord resumption (expired). Tenant to be left with at least 25 acres depending on length of lease and class of land. Maharashtra Lesser of 1/6 of produce or 2 to 5 times as- Compulsory transfer except in Vidharba area. Many pur- sessment chases by tenants are "ineffective." Efforts to confer own- ership may have decreased security. South India Karnataka 1/5 to 1/4 of produce depending on land class Tenure is fixed subject to landlord resumption of up to half the leased area. Kerala Paddy land 1/8 to 1/4 of produce; garden 1/10 Tenure for all tenants permanently fixed subject to landlord to 1/3; dry land 1/8 for cultivated; Rs4.00 resumption (expired). Tenants purchase ownership for 16 per acre for other larfd times fair rent. Madras 40% of normal gross produce for wetland; Interim act (1955) forbidding eviction expired in 1965. Ten- 33.3% of normal gross produce for dry- ants have right to reclaim up to 6.6 acres if evicted. land (for less than 3/4 of normal yields, rent may be reduced) Andhra In Andhra area 1/2 of gross produce on land In Andhra area eviction of tenants prohibited; in Hydera- Pradesh irrigated by baling; in Hyderabad area bad tenants in possession 6 years protected from eviction; lesser of 1/5 of produce or 3 to 5 times as- landlord resumption limited to 3 family holdings (expired). sessment; 1/2 of output on irrigated lands ffi East India Bihar 1/2 of gross produce Tenant obtains occupancy right after 12 years of possession; under-ryots have little security. Orissa 1/4 of gross produce Fixity of tenure subject to landlord's resumption (expired) of up to half area. West Bengal Noregulationexceptforbargadars(sharecrop- Tenancy abolished. Bargadars do not have tenancy status but pers); 1/2 of produce if landlord supplies have security subject to owner's right to resume 2/3 of the plow, manure, and seed; 40% if cropper area leased. If area is less than half an acre, owner may supplies resume all. Assam 1/2 of produce if landlord supplies bullocks; Rights of resumption expired 1963. Occupancy ryots have 1/5 of produce if tenant (adhiar) supplies permanent transferable interest. Nonoccupancy ryots and bullocks under-ryots have no security. Pakistan Yes; rents to be paid in kind according to es- Ejectment only for just cause; occupancy tenants to be made tablished tradition owners. a. A batai is an arrangement under which tenants exchange their labor and a share of the output for use of the land. Source: Wunderlich (1970), pp. 40-42. 374 Statistical Appendix Table A44. South Asia: Statewide Legislation on Ceiling on Holdings, 1969-74 (hectares) Range of Operated Median ceilings per area per ceiling Region and state household rural household levels' Bangladesh 1.3-13.7b 0.6 8.7c North India Jammu and Kashmir n.a. 1.0 5.1 Himachal Pradesh 4.1-12.2 n.a. 8.1 Punjab 7.0-21.8 1.6 14.4 Haryana 7.3-21.8 1.9 14.5 Uttar Pradesh 7.3-18.3 1.1 12.8 West India Rajasthan 7.3- 1.9d 4.1 12.6 Gujarat 4.1-21.9 2.3 13.0 Madhya Pradesh 4.1-7.3e 2.3 5.7 Maharashtra 7.3-19.4 2.6 13.4 South India Karnataka 4.1-8.2e 2.3 6.2 Kerala 2.4-8.1 0.3 5.3 Tamil Nadu 6.2-12.2 0.7 9.1 Andhra Pradesh 4. 1-21.9 1.5 13.0 East India Orissa 0.8_7.3e 1.1 4.1 West Bengal 5.0-7.0 0.6 6.0 Bihar 6.1-18.2 0.8 12.1 Assam 6.7-8.7 0.6 7.8 Pakistan 1959 (per person) 207-414 2.0 61.0 1972 (per person) 62 or 124 n.a. Not available. a. These median ceiling levels are used for the exercise on land redistribution schemes in appendix B, table B-1. b. Originally set at 100 bighas (33.3 acres) per household in 1950, raised to 375 bighas in 1961, and reset at 100 bighas in 1972. c. The median size of 8.7 hectares is too large, so a minimum value of 1.3 hectares was used to construct alternative scenarios for Bangladesh in appendix B, table B-1. d. Desert land is excluded. e. Based on the standard acre (approximately 1 acre of wetland). Sources: India, Ministry of Agriculture and Irrigation (1976), pp. 135-44; M. H. Khan (1981); Abdullah (1976). Statistical Appendix 375 Table A45. India: Fragmentation of Operational Holdings by State, 1960-61 Number offragments per operational holding All size State classes 2.5-4.9 acres 5.0-7.5 acres Andhra Pradesh 4.32 4.32 5.03 Assam 2.75 2.96 3.50 Bihar 7.18 8.04 10.97 Gujarat 4.40 3.49 4.17 Jammu and Kashmir 5.09 5.83 6.63 Kerala 2.01 3.40 3.73 Madhya Pradesh 5.30 4.31 4.95 Madras 4.96 5.40 6.58 Maharashtra 3.78 3.50 3.51 Mysore 3.79 3.18 3.85 Orissa 6.39 6.08 8.32 Punjab (induding Haryana) 4.76 4.32 4.65 Rajasthan 4.27 3.66 3.89 Uttar Pradesh 7.78 8.33 9.27 West Bengal 7.12 7.48 10.02 All India 5.66 6.05 6.79 Source: Minhas (in Srinivasan and Bardhan 1974), tables 3 and 4. Bibliography Abdullah, Abu. 1976. "Land Reform and Agrarian Change in Bangladesh." Bangladesh Development Studies 4 (1) January: 67-112. Abdullab, Abu, M. Hossain, and R. Nations. 1976. "Agrarian Structure and the IRDP: Preliminary Considerations." Bangladesh Development Studies 4 (2) April. Abraham, A. 1980. "Maharashtra's Employment Guarantee Scheme." Eco- nomic and Political Weekly 15 (32) August 9. Abraham, S. 1973. "Cooperative Credit for Small Farmers in India." Small Farmer Credit in South Asia. USAID Spring Review of Small Farmer Credit. Washington, D.C.: U.S. Agency for International Development. Acharya, K. T, and V. K. Huria. 1986. "Rural Poverty and Operation Flood." Economic and Political Weekly 21 (37) September 13: 165 1-56. Acharya, S. S. 1973. "Green Revolution and Farm Employment." Indian jour- nal of Agricultural Economics 28 (3). Adams, D. W. 1973. "The Case for Voluntary Savings Mobilization: Why Rural Capital Markets Flounder." Analytical Papers 19. USAID Spring Review of Small Farmer Credit, Washington, D.C.: U.S.. Agency for International Development. Adams, D. W., M. L. Ong, and 1. J. Singh. 1975. Voluntary Rural Savings Capac- ities in Taiwan, 1960 to 1970. Economics and Sociology Paper 175. Colum- bus, Ohio: Department of Agricultural Economics and Rural Sociology, Ohio State University. ADB (Asian Development Bank). 1977. Asian Agricultural Survey 1976: Rural Asia. Manila. Adnan, S. 1977. Differentiation and Class Structure in Village Shamraj. Village Studies Project Paper 8. Dhaka: Bangladesh Institute for Development Studies. Adnan, S., and others. 1978. "A Review of Landlessness in Rural Bangladesh, 1877-1977." Rural Economics Program, Department of Economics, Uni- versity of Chittagong, Bangladesh. August. 377 378 Bibliography AERC (Agro-Economic Research Center). 1969. "A Study of High-Yielding Varieties Programme in the District of Cuttack, Orissa, with Special Refer- ence to Credit, 1967." Vishwa-Bharati, Shantiniketan. - 1970a. "Evaluation of the High-Yielding Varieties Programme, Khariff, 1968: A Study of IR8 Paddy in Karnal District, Haryana." Delhi. - 1970b. "Report on High-Yielding Variety Programme in Paddy in Sibsagar District, 1968-69." Jorhat. . 1970c. "Report on the High-Yielding Varieties Programme in Vara- nasi District, 1968." University of Allahabad. - 1970d. "Report on the Study of the High-Yielding Varieties Pro- gramme (Rabi 1968-69)." Waltair: Andhra University. - 1972. "Study on the Problems of Marginal Farmers and Landless Ag- ricultural Labourers in the District of Hooghly, West Bengal." Vishwa- Bharati, Shantiniketan. - 1974. "Study of the Small Farmers Development Agency, Nalgonda District, Andhra Pradesh (Rabi 1972-73)." Waltair: Andhra University. Oc- tober. - 1975a. "Study on Marginal Farmers and Agricultural Labourers in Bankura, West Bengal." Vishwa-Bharati, Shantiniketan. July. - 1975b. "Study of the Marginal Farmers and Agricultural Labourers Development Agency, Vishakapatnam District, Andhra Pradesh (1972-73 and 1973-74)." Waltair: Andhra University. August. - 1975c. "Study of the Small Farmers Development Agency, Nalgonda District, Andhra Pradesh (1973-74)." Waltair: Andhra University. Decem- ber. - 1976a. "Marginal Farmers and Agricultural Labourers in Pondi- cherry." University of Madras. - 1976b. "Report on the Evaluation Study of Small Farmers Develop- ment Agency in South Arcot District, Tamil Nadu (1972-73)." Department of Economics, University of Madras. . 1976c. "Study of Incomes, Savings and Investments of Selected Culti- vator and Labour Households in East Godavari, Andhra Pradesh." Waltair: Andhra University. - 1976d. "A Study of Rural Poverty and Inequality in a Developed Dis- trict (East Godavari)." Waltair: Andhra University. - 1976e. "Study of Small Farmers Development Agency, Ganjami, Orissa." Vishwa-Bharati, Shantiniketan. - . 1978. Annual Report on: (1) Farm Structure Study: 1976-77 and (2) Eco- nomic Analysis of New Farming Practices: Kharif 1976. Jodhpur, Rajasthan: Central Arid Zone Research Institute. Afzal, M. 1974. The Population of Pakistan. CICRED series. Islamabad: Pakistan Institute of Development Economics. Agarwal, Bina. 1980a. "Effect of Agricultural Mechanization on Crop Out- Bibliography 379 put: A Study of Operation-wise Effects for HYV Wheat in the Punjab." In- dian Economic Review 15 (7) January-March: 29-51. . 1980b. "Tractorisation, Productivity and Employment: A Reassess- ment." Journal of Development Studies 16 (3) April: 375-86. . 1981. "Agricultural Mechanization and Labor Use: A Disaggre- gated Approach." International Labour Review 120 (1) January-February: 115-27. Agarwal, N. L., and R. K. Kumawat. 1974. "Green Revolution and Capital and Credit Requirements of the Farmers in the Semi-Arid Regions of Raja- sthan." IndianJournal of Agricultural Economics 29 (1): 67-75. Agarwal, S. 1968a. "Impact of Consolidation of Land Holdings." Agricultural Situation in India 23 (12) March. . 1968b. "Impact of Consolidation of Land Holdings: A Field Study." Artha Vijnana 10 (1) March. . 1971. Economics of Land Consolidation in India. New Delhi: S. Chand and Co. Agarwala, R. 1964a. "Size of Holdings and Productivity: A Comment." Eco- nomic and Political Weekly, April 11. . 1964b. "Size of Holdings and Productivity: Further Comments." Eco- nomic and Political Weekly, November 21. Ahluwalia, I. J. 1985. Industrial Growth in India. Oxford: Oxford University Press. Ahluwalia, Montek S. 1978. "Rural Poverty and Agricultural Performance in India." Journal of Development Studies 14 (April): 298-323. Ahmad, Bashir. 1972. "Farm Mechanization and Agricultural Development: A Case Study of the Pakistan Punjab." Ph.D. diss., Michigan State Univer- sity, East Lansing. Ahmad, Ehtisham, and Stephen Ludlow. 1989. "Poverty, Inequality and Growth in Pakistan." Paper presented at the World Bank-International Food Policy Research Institute Policy Research Conference, Airlee, Va., October. Ahmad, M. 1974. "Farm Efficiency under Owner Cultivation and Share Ten- ancy." Pakistan Economic and Social Review 12 (2) Summer: 131-43. Ahmed, Iftikhar. 1975. The Green Revolution, Mechanization, and Employment. ILO Working Paper. Geneva: International Labour Office. . 1978. "Unemployment and Underemployment in Bangladesh Agri- culture." World Development 6: 1281-96. . 1981. Technological Change and Agrarian Structure: A Study of Bangla- desh. Geneva: International Labour Office. Ahmed, R. 1979a. Foodgrain Supply, Distribution and Consumption Policies within a Dual Pricing Mechanism: A Case Study of Bangladesh. Research Report 8. Washington, D.C.: International Food Policy Research Institute. . 1979b. "Production and Consumption Linkages among Rural Enter- 380 Bibliography prises: A Case Study of Bangladesh: A Project Proposal." International Food Policy Research Institute, Washington, D.C. Draft. - 1987. A Structural Perspective of Farm and Non-Farm Households in Ban- gladesh. Washington, D.C.: International Food Policy Research Institute. Ahn, Choong Yong, and Inderjit Singh. 1981. "A Model of an Agricultural Household in a Multi-Crop Economy: The Case of Korea." Review of Eco- nomics and Statistics 63 (4) November: 520-25. Aiyaswamy, U., and H. G. Bhole. 1981. "Market Access as Constraint on Mar- ginal and Small Farms." Economic and Political Weekly 16 (13) March 28. Akbar, Ali. 1975. "1974 Famine in Bangladesh." Rajshahi University. Novem- ber. Akram, C. M., ed. 1973. Manual of Land Reforms. Lahore: Government of Pa- kistan. Alagh, Y. K., G. S. Bhalla, and A. Bhaduri. 1978. "Agricultural Growth and Manpower Absorption in Indian Agriculture." In P. K. Bardhan, A. Vaidyanathan, and Y. K. Alagh, eds., Labor Absorption in Indian Agriculture. Bangkok: International Labour Organisation and Asian Regional Pro- gramme for Employment Promotion. Alagh, Y K., and P. S. Sharma. 1980. "Growth of Crop Production: 1960-61 to 1978-79: Is It Declining?" Indian Journal of Agricultural Economics 35 (2) April-June. Alam, M. S. 1974. "The Economics of Landed Interests: A Case Study of Pa- kistan. Pakistan Economic and Social Review 12 (Spring): 12-24. Alamgir, M. 1975. "Poverty, Inequality and Social Welfare: Measurements, Evidence and Policies. Bangladesh Development Studies, April. - 1978. Bangladesh: A Case of Below Poverty Level Equilibrium Trap. Dhaka: Bangladesh Institute of Development Studies. - 1980. Famine in South Asia: Poltical Economy of Mass Starvation. Cam- bridge, Mass: Oelgeschlager, Gunn and Hain. Alauddin, T 1975. "Mass Poverty in Pakistan: A Further Study." Pakistan De- velopment Review, Winter. Alavi, Hamza. 1973. "Elite Farmer Strategy and Regional Disparities in the Agricultural Development of Pakistan." Economic and Political Weekly, Re- view of Agriculture 8 (12) March 24: A31-A39. Albrecht, Herbert. 1976. Living Conditions of Rural Families in Pakistan. Frithjof Kuhnen, ed.; V. June Hager, trans. Socioeconomic Studies on Rural Devel- opment. Saarbrucken: ssip-Schriften Breitenbach. - 1981. "New Technologies and Income Distribution in Agriculture: The Case of the Pakistan Punjab." Institute for Rural Development, Uni- versity of Gottingen. Draft. Alderman, H. 1987. Co-operative Dairy Development in Karnataka, India: An As- sessment. Research Report 64. Washington, D.C.: International Food Policy Research Institute. Alderman, H., G. Mergos, and R. Slade. 1987. Co-operatives and the Commer- Bibliography 381 cialization of Milk Production in India: A Literature Review. Working Paper on Agriculture 2. Washington, D.C.: International Food Policy Research Insti- tute. Ali, A. 1980. "Part-Time Farming in Bangladesh." Labour, Capital and Society 13 (2) November. Allen, R. 1978. "Employment of the HYv Strategy." U.S. Agency for Interna- tional Development, Dhaka. Draft. Allison, Christine. 1989. "Poverty, Inequality and Public Policy in Pakistan." Paper presented at the World Bank-International Food Policy Research Institute Poverty Research Conference, Airlee, Va., October. Allison, S. 1974. "Progress Towards Development of a More Appropriate Technology for Land and Water Development in Bangladesh." World Bank, Dhaka. March. Amjad, Rashid. 1972. "A Critique of the Green Revolution in West Paki- stan." Pakistan Economic and Social Review 10 (1) June. Anderson, Dennis. 1987. The Economics of Afforestation: A Case Study in Africa. Baltimore, Md.: Johns Hopkins University Press. Anderson, Dennis, and Mark Leiserson. 1978. Rural Enterprise and Nonfarm Employment. Washington, D.C.: World Bank. . 1980. "Rural Non-Farm Employment in Developing Countries." Eco- nomic Development and Cultural Change 28 (2) January: 227-48. Appa Rao, V., and R. R. Krishna. 1978. "Impact of Dairying on Small and Marginal Farmers and Agricultural Labourers: A Case Study of Two Vil- lages in Arakapalli." Waltair: Technical Cell, Andhra University. Draft. Appu, P. S. 1975. "Tenancy Reform in India." Economic and Political Weekly 10 (33-35) August. Asaduzzaman, M., and M. Hossain. 1974. Some Aspects of Agricultural Credit in Two Irrigated Areas in Bangladesh. Research Report 18. Dhaka: Bangla- desh Institute of Development Studies. Asaduzzaman, M., and Faridul Islam. 1976. Adoption of HYvs in Bangladesh: Some Preliminary Hypotheses and Tests. Research Report, n.s. 23. Dhaka: Ban- gladesh Institute of Development Studies. Asian and Pacific Center for Women and Development. 1979. Women's Employ- ment: Possibilities of Relevant Research. Section 5, Impact on Women: The Anand Pattern of Dairy Development. Anand: National Dairy Development Board. Athreya, V. B, G. Boklin, G. Djurfeldt, and S. Lindberg. 1986. "Economies of Scale or Advantage of Class." Economic and Political Weekly 21 (13) March 29: A2-A26. Attwood, D. W. 1979. "Why Some of the Poor Get Richer: Economic Change and Mobility in Rural Western India." Current Anthropology 20 (3) Septem- ber: 495-515. Auden-Laerina, T., and R. Barker. 1978. "The Adoption of Modern Vari- eties." In Interpretive Analysis of Selected Papers from "Changes in Rice Farming in Selected Areas of Asia." Los Bafios: International Rice Research Institute. 382 Bibliography Azad, M. P., R. N. Yadar, and H. K. Nigan. 1980. "Economics of Poultry Farming in Kanpur: Case Study, 1979-80." Indian Journal of Agricultural Economics 35 (4) October-December: 170. Aziz, S. 1978. Rural Development: Learning from China. London: Macmillan. Bagchee, S. 1987a. "Rural Development Strategies." Economic and Political Weekly 22 (30) July 18: 1189-90. - 1987b. "Poverty Alleviation Programmes in the Seventh Plan: An Ap- praisal." Economic and Political Weekly 22 (4) January 24: 139-48. Bagi, E S. 1979a. "Irrigation, Farm Size and Economic Efficiency: An Analy- sis of Farm-Level Data in Haryana (India) Agriculture." Tennessee State University, Nashville. June. - 1979b. "Relationship between Farm Size, Tenancy, Productivity and Returns to Scale in Haryana (India) Agriculture." Tennessee State Univer- sity, Nashville. October. - 1979c. "Technical and Allocative Efficiencies of Share-Cropping in Haryana (India) Agriculture." Tennessee State University, Nashville. May. - 1980. "The Stochastic Frontier Production Function and Sources of Measured Technical Efficiency." Tennessee State University, Nashville. April. - 1981. "The Stochastic Frontier Production Function and the Effect of Irrigation on the Average Technical Efficiency." Department of Rural Development, Tennessee State University, Nashville. July. Ballabh, V., and B. Sharma. 1987. "Adoption of HYV Paddy and Wheat in Flood-Prone and Flood-Free Districts of Uttar Pradesh: Implications for Research Strategy." Indian Journal of Agricultural Economics 42 (1) January- March: 76-90. Bandhopadhyaya, N. 1981. "Operation Barga and Land Reform Perspectives in West Bengal: A Discursive Review." Economic and Political Weekly 16 (25-26) June 20-27: A38-A42. Bangladesh Bank. 1978. "Problems and Issues of Agricultural Credit and Rural Finance." Deliberations of the International Workshop on Providing Financial Services to the Rural Poor, Dhaka, October 23-25. Bangladesh Bureau of Statistics. 1972. Master Survey of Agriculture in Bangla- desh, 1967-68. 7th Round, 2nd phase. Dhaka. 1977. Agriculture Census, 1977. Dhaka: Government Press. 1983. Report on Bangladesh Pilot Agricultural Census 1982. Dhaka: Gov- ernment Press. - 1985. Statistical Pocket Book of Bangladesh, 1985. Dhaka: Government Press. - 1986a. The Bangladesh Census of Agriculture and Livestock: 1983-84. Dhaka: Government Press. - 1986b. Final Report: Labor Force Survey 1981-82. Dhaka: Ministry of Planning. - 1986c. Household Expenditure Survey for Bangladesh, 1981-82. Dhaka. Bibliography 383 1986d. Report of the Bangladesh Household Expenditure Survey, 1981- 82. Dhaka: Ministry of Planning. Bangladesh, Ministry of Agriculture and Forests. 1978. Agrarian Structure and Change: Rural Development Experience and Policies in Bangladesh. Background Papers to the Bangladesh Country Report for the World Conference on Agrarian Reform and Rural Development, Dhaka, May. Bansie, P. 1977. Agricultural Problems in India. New Delhi: Sterling. Bapna, S. L. 1973. "Economic and Social Implications of the Green Revolu- tion: A Case Study of Kota District." Vallabh, Vidyanagar: Agro-Economic Research Center, Sardar Patel University. BARC (Bangladesh Agricultural Research Council). 1978. Incidence of Landless- ness and Major Landholding and Cultivation Groups in Rural Bangladesh. Agri- cultural Economics and Rural Social Science Paper 5. Dhaka. Bardhan, Pranab K. 1970. "The Green Revolution and Agricultural Labour- ers." Economic and Political Weekly 5 (29-31) January: 1239-46. . 1971. "Trends in Land Relations in India: A Note." Economic and Polit- ical Weekly, Annual Number. . 1973a. "Size, Productivity and Returns to Scale: An Analysis of Farm-Level Data in Indian Agriculture." Journal of Political Economy 81: 1370-86. . 1973b. "Variations in Agricultural Wages: A Note." Economic and Polit- ical Weekly 8 (21) May 26: 947-51. . 1976. "Variations in Extent and Forms of Agricultural Tenancy." Eco- nomic and Political Weekly 11 (37-38) September 11-18. . 1977a. "Rural Employment and Wages with Agricultural Growth in India: Some Intertemporal and Cross-Sectional Analyses." Studies in Em- ployment and Rural Development 38. World Bank, Development Econom- ics Department, Washington, D.C. March. . 1977b. "Rural Employment, Wages and Labor Markets in India: A Survey of Research." Economic and Political Weekly 12 (26-28) June 28, July 2-9. 1977c. "Variations in Forms of Tenancy in a Peasant Economy."Jour- nal of Development Studies 4 June): 105-18. . 1978. "On Measuring Rural Unemployment." University of Califor- nia, Berkeley. Unpublished. =. 1979. "Agricultural Development and Land Tenancy in a Peasant Economy: A Theoretical and Empirical Analysis." AmericanJournal of Agri- cultural Economics 61 (1) February. . 1980. "Interlocking Factor Markets and Agrarian Development: A Review of Issues." Oxford Economic Papers, March. . 1981. "Labor Tying in a Poor Agrarian Economy: A Theoretical and Empirical Analysis." University of California, Berkeley. May. Unpublished. 1984. Land, Labour and Rural Poverty. Essays in Development Econom- ics. Oxford: Oxford University Press. 384 Bibliography - 1986. The Political Economy of Development in India. Oxford: Oxford University Press. Bardhan, P. K., and A. Rudra. 1978. "Interlinkage of Land, Labor and Credit Relations." Economic and Political Weekly 13 (6-7) February. - 1980a. "Labour, Employment and Wages in Agriculture: Results of a Survey in West Bengal, 1979." Economic and Political Weekly 15 (45-46) No- vember 8-15. - 1980b. "Terms and Conditions of Sharecropping Contracts: An Anal- ysis of Village Survey Data in India." Journal of Development Studies 16 (April): 287-302. - 1980c. "Types of Labor Attachment in Agriculture: Results of a Survey in West Bengal, 1979." Economic and Political Weekly 15 (35) Au- gust 30. . 1981. "Terms and Conditions of Labour Contracts in Agriculture." Oxford Bulletin of Economics and Statistics 43: 89-111. Bardhan, P K., and T N. Srinivasan. 1971. "Crop Sharing Tenancy in Agri- culture: A Theoretical and Empirical Analysis." American Economic Review 61 (1) March: 48-64. Barker, K., and R. Herdt. 1978. "Equity Implications of Technology Changes." In Interpretive Analysis of Selected Papersfrom " Changes in Rice Farm- ing in Selected Areas of Asia." Los Bafios: International Rice Research Insti- tute. Barker, R., and V. Cordova. 1978. "Labor Utilization in Rice Production." In Economic Consequences of the New Rice Technology. Los Bafios: Interna- tional Rice Research Institute. Barnum, Howard N., and Lyn Squire. 1978. "Technology and Relative Eco- nomic Efficiency. Oxford Economic Papers 30 (2) July: 181-98. - 1979. A Model of an Agricultural Household: Theory and Evidence. Balti- more, Md.: Johns Hopkins University Press. Bartsch, W. H. 1977. Employment and Technology Choice in Asian Agriculture. New York: Praeger. Basant, R. 1987. "Agricultural Technology and Employment in India: A Sur- vey of Recent Research." Economic and Political Weekly 22 (31) August 1: 1297 and (32) August 8: 1348. Basu, K. 1981. "Food for Work Programmes: Beyond Roads That Get Washed Away." Economic and Political Weekly 16 (1-2) January 3-10: 37-40. Basu, S. 1979. Commercial Banks and Agricultural Credit: A Study of Regional Dis- parity in India. Bombay: Allied Publishers. Baviskar, B. 1988. "Development and Controversy." Economic and Political Weekly 23 (13) March 26: A35-A43. Baviskar, B., and S. George. 1987. "Operation Flood: A Different View." Eco- nomic and Political Weekly 21 (44-45) November 1-8: 1961-62. Bell, Clive. 1972. "The Acquisition of Agricultural Technology: Its Determi- nants and Effects." Journal of Development Studies 9 (1) October: 123-59. Bibliography 385 - 1974a. "Crop Sharing Tenancy in Agriculture: A Rejoinder." Ameri- can Economic Review 64 (6) December: 1007-09. . 1974b. "Ideology and Economic Interests in Indian Land Reform." In David Lehmann, ed., Peasants, Landlords and Governments: Agrarian Re- form in the Third World. New York: Holmes and Meier. -. 1976. "Production Conditions, Innovation and the Choice of Lease in Agriculture." Sankhya: The Indian Journal of Statistics 38 (ser. C, pt. 4) De- cember. . 1977. "Alternative Theories of Sharecropping: Some Tests Using Evi- dence from Northeast India." Journal of Development Studies 13 (July). 1978. "Some Reflections on Tenancy in South Asia." Development Research Center, World Bank, Washington, D.C. January. Draft. Bell, Clive, and Peter Hazell. 1980. "Measuring the Indirect Effects of an Ag- ricultural Investment Project on Its Surrounding Region." American Journal of Agricultural Economics 62 (1) February. Bell, Clive, Peter Hazell, and Roger Slade. 1982. Project Evaluation in Regional Perspective: A Study of an Irrigation Project in Northwest Malaysia. Baltimore, Md.: Johns Hopkins University Press. Bell, Clive, and P. Zusman. 1976. "A Bargaining Theoretic Approach to Cropsharing Contracts." American Economic Review 66 (September): 578- 87. Benor, Daniel, James Q. Harrison, and Michael Baxter. 1984. Agricultural Ex- tension: The Training and Visit System. Washington, D.C.: World Bank. Bergmann, T, and D. Eitel. 1976. Promotion of the Poorer Sections of the Indian Rural Population. Socioeconomic Studies on Rural Development. Saar- brucken: ssip-Schriften Breitenbach. Berry, R. Albert, and William L. Cline. 1979. Agrarian Structure and Productiv- ity in Developing Countries. Baltimore, Md.: Johns Hopkins University Press. Berry, R. Albert, and R. Sabot. 1978. "Labour Market Performance in Devel- oping Countries: A Survey." World Development 6 (11-12): 1199-1246. . 1984. "Unemployment and Economic Development." Economic Devel- opment and Cultural Change 33 (October): 99-116. Bertocci, P. 1972. "Community Structure and Social Relations in Two Vil- lages in Bangladesh." Contributions to Indian Sociology (6): 28-35. Beteille, A. 1965. Caste, Class and Power: Changing Patterns of Stratification in a Tanjore Village. Berkeley: University of California Press. . 1974. Studies in Agrarian Social Structure. Delhi: Oxford University Press. . 1975. "Agrarian Relations in Tanjore District, South India." Univer- sity of New Delhi. Bhaduri, A. 1973. "Agricultural Backwardness under Semi-Feudalism." Eco- nomic Journal 329 (83) March. . 1977. "On the Formation of Usurious Interest Rates in Backward Ag- riculture." Cambridge Journal of Economics 1 (4): 341-52. 386 Bibliography Bhagwati, J. 1988. "Poverty and Public Policy." World Development 16 (5): 539-55. Bhagwati, J., and Sukhamoy Chakravarty. 1969. "Contributions to Indian Economic Analysis: A Survey." American Economic Review 59 (4, pt. 2) Sep- tember. Bhalla, G. S., and Y K. Alagh. 1979. Performance of Indian Agriculture: A Distrietwise Study. New Delhi: Sterling. Bhalla, G. S., and G. K. Chadha. 1982. "Green Revolution and the Small Peas- ant: A Study of Income Distribution in Punjab Agriculture." Economic and Political Weekly 17 (2) May 15-22: 826-33, 870-77. - 1983. Green Revolution and the Small Peasant: A Study of Income Distribu- tion among Punjab Cultivators. New Delhi: Concept Publishers. Bhalla, Sheila. 1976. "New Relations of Production in Haryana Agriculture." Economic and Political Weekly, Review of Agriculture 9 (13) March 27: A23-A30. - 1977. "Changes in Acreage and Tenure Structure of Land Holdings in Haryana, 1962-1972." Economic and Political Weekly 12 (13) March 26: A2-A15. - 1979. "Real Wage Rates of Agricultural Labourers in Punjab, 1961- 1977: A Preliminary Analysis." Economic and Political Weekly, Review of Agri- culture 14 (26) June 30: A57-A68. Bhalla, Surjit. 1980. "Measurement of Poverty: Issues and Methods." Devel- opment Economics Department, World Bank, Washington, D.C. January. Draft. - 1983. "Does Land Quality Matter: Theory and Measurement." Devel- opment Economics Department, World Bank, Washington, D.C. Decem- ber. Processed. Bhalla, Surjit, and P. L. Roy. 1983. "Mis-Specification in Farm Productivity Analysis." Development Economics Department, World Bank, Washington, D.C. November. Processed. Bharadwaj, Krishna. 1974a. "Notes on Farm Size and Productivity." Economic and Political Weekly 9 (13) March 30. - 1974b. Production Conditions in Indian Agriculture: A Study Based on Farm Management Surveys. Department of Applied Economics, Occasional Paper 33. Cambridge: Cambridge University Press. Bharadwaj, Krishna, and P. K. Das. 1975a. "Tenurial Conditions and the Mode of Exploitation: A Study of Some Villages in Orissa." Economic and Political Weekly 10 (5-7) February: 221-39. - 1975b. "Tenurial Conditions and the Mode of Exploitation: A Study of Some Villages in Orissa-Further Notes." Economic and Political Weekly 10 (25-26) June 21-28. Bhatt, V. 1988. "Growth and Income Distribution in India: A Review." World Development 15 (5): 641-47. Bhattacharya, G., and M. Chattopadhyay. 1986. "Growth of Indian Agricul- Bibliography 387 ture: A Reappraisal." IndianJournal of Agricultural Economics 42 (1) January- March: 67-76. Bhattacharya, N., M. Chattopadhyay, and A. Rudra. 1987. "Changes in Level of Living in Rural West Bengal: Social Consumption." Economic and Political Weekly 22 (28) July 11 and (33) August 15: 1410-13. Bhattacharya, N., and G. R. Saini. 1972. "Farm Size and Productivity: A Fresh Look." Economic and Political Weekly, June 24. Bhattacharya, P., and A. Majid. 1976. "Impact of the Green Revolution on Output, Cost and Income of Small and Big Farmers." Economic and Political Weekly 11 (52) December 25: A147-A150. Bhende, M. 1986. "Credit Markets in Rural South Asia." Economic and Political Weekly 21 (38-39) September 20-27: Al19-A125. Billings, Martin H., and Arjan Singh. 1970. "Farm Mechanization and the Green Revolution, 1964-1984: The Punjab Case." U.S. Agency for Inter- national Development, New Delhi. April. Processed. Binswanger, Hans P. 1979. The Economics of Tractors in South Asia: An Analytical Review. New York: Agricultural Development Council. Binswanger, Hans P., and J. McIntire. 1987. "Behavioral and Material Deter- minants of Production Relations in Land-Abundant Tropical Agriculture." Economic Development and Cultural Change 36 (1) October: 73-99. Binswanger, Hans P., and M. R. Rosenzweig. 1981. Contractual Arrangements, Employment and Wages in Rural Labour Markets: A Critical Review. New York: Agricultural Development Council; and Hyderabad: International Crops Research Institute for the Semi-Arid Tropics. Blair, H. 1974. The Elusiveness of Equity: Institutional Approaches to Rural Devel- opment in Bangladesh. Dhaka and Ithaca, N.Y.: Cornell Center for Interna- tional Studies. . 1986. "Social Forestry: Time to Modify Goals?" Economic and Political Weekly 21 (30) July 26: 1317-22. Bliss, C. J., and N. H. Stern. 1980. Palampur: Studies in the Economy of a North Indian Village. New Delhi: Oxford University Press. Blyn, George. 1979. "The Green Revolution and Rural Social Structure in India." Rural Development: Contributions to Asian Studies 13, ed. Mohinder Chaudhry. Leiden: E. J. Brill. . 1983. "The Green Revolution Revisited." Economic Development and Cultural Change 31 (4) July: 705-25. Bose, Ajit N. 1978. Calcutta and Rural Bengal: Small Sector Symbiosis. Calcutta: Minerva Associates. Bose, Swadesh R. 1968. "Trend of Real Income of the Rural Poor in East Pa- kistan, 1949-66." Pakistan Development Review 5 (3) Autumn. . 1974a. "The Comilla Co-operative Approach and the Prospects for Broad-Based Green Revolution in Bangladesh." World Development 2 (8) August: 21-28. 388 Bibliography - 1974b. "Movement of Agricultural Wage Rates in Bangladesh." In A. Mitra, ed., Economic Theory and Planning: Essays in Honour of A. K. Dasgupta. Oxford: Oxford University Press. Bose, S. R., and E. M. Clark. 1969. "Some Basic Considerations on Agricul- tural Mechanization in West Pakistan. Pakistan Development Review 9 (Au- tumn): 273-308. Bottomley, Anthony. 1963. "The Premium for Risk as a Determinant of In- terest Rates in Underdeveloped Areas." Quarterly Journal of Economics 76: 637-47. . 1964a. "The Determination of Pure Rates of Interest in Underdevel- oped Rural Areas." Review of Economics and Statistics 46: 301-04. - 1964b. "Monopoly Profit as a Determinant of Interest in Underdevel- oped Rural Areas." Oxford Economic Papers 16: 431-37. - 1964c. "The Structure of Interest Rates in Underdeveloped Rural Areas." Journal of Farm Economics 46 (May): 313-22. - 1973. "A Tide in the Affairs of Men." Small Farmer Credit Summary Papers 20. Washington, D.C.: U.S. Agency for International Development. June. Bottrall, A. 1974. Serving the Small Farmer: Policy Choices in Indian Agriculture. London: Overseas Development Institute. Bowonder, B., and others. 1987. "Further Evidence on the Impact of Dairy Development Programmes." Economic and Political Weekly 22 (13) March 28: A6-A15. Brammer, Hugh. 1980. "Some Innovations Don't Wait for Experts: A Report on Applied Research by Bangladeshi Peasants." Ceres: FAO Review on Agricul- ture and Development, March-April. Brannon, R., and D. Jessee. 1977. Unemployment and Underemployment in the Rural Sectors of the Less Developed Countries. Occasional Paper 4. Washington, D.C.: Economics and Sector Planning Division, Technical Assistance Bu- reau, U.S. Agency for International Development. Braverman, Avishay, and J. L. Guasch. 1986. "Rural Credit Markets and In- stitutions in Developing Countries: Lessons for Policy Analysis from Prac- tice and Modern Theory." World Development 14 (October-November): 1253-69. Braverman, Avishay, and T. N. Srinivasan. 1980. Agrarian Reforms in Develop- ing Rural Economies Characterized by Interlinked Credit and Tenancy Markets. World Bank Staff Working Paper 433. Washington, D.C. Braverman, Avishay, and J. E. Stiglitz. 1980. "Sharecropping and Interlink- ing of Agrarian Markets." Development Research Center, World Bank, Washington, D.C. October. Draft. - 1981a. "Cost Sharing Arrangements under Sharecropping: A Multi- dimensional Incentive Program." Development Research Center, World Bank, Washington, D.C. April. Draft. - 1981b. "Landlords, Tenants and Technological Innovations." Devel- opment Research Center, World Bank, Washington, D.C. April. Draft. Bibliography 389 Brown, D. 1971. Agricultural Development in India's Districts. Cambridge, Mass.: Harvard University Press. Brown, L. 1970. Seeds of Change: The Green Revolution and Development in the 1970s. New York: Praeger. Bruce, R., Jr. 1982. "Agrarian Organization and Resource Distribution in South India: Bellany District, 1800-1979." Ph.D. diss., University of Wis- consin, Madison. Brumby, P. 1978. "Complementarity between Milk and Foodgrain Produc- tion: A Linear Programming Exercise." South Asia Projects Department, World Bank, Washington, D.C. March. Draft. . 1980. "The Dairy Situation in India." In Proceedings of the Agricultural Sector Symposia. Washington, D.C.: World Bank. Burki, S. J. 1969. A Study of Chinese Communes. Cambridge, Mass.: East Asian Research Center, Harvard University. Burki, S. J., and others. 1976. Public Works Programs in Developing Countries: A Comparative Analysis. World Bank Staff Working Paper 224. Washington, D.C. Bussink, W., and K. Subbarao. 1986. "Living Standards Improvement in India: An Agenda for Studies." World Bank, New Delhi. July. Processed. Byerlee, D. 1973. Indirect Employment and Incom.e Distribution Effects of Agricul- tural Development Strategies: A Simulation Approach Applied to Nigeria. African Rural Employment Paper 9. East Lansing: Department of Agricultural Economics, Michigan State University. Byrd, William A., and Lin Qingsong, eds. 1990. China's Rural Industry: Struc- ture, Development, and Reform. New York: Oxford University Press. Byres, T 1972. "The Dialectic of India's Green Revolution." South Asian Re- view 5 (2) November-January. . 1974. "Land Reform, Industrialization and the Market Surplus in India: An Essay on Rural Bias." In David Lehman, ed., Agrarian Reform and Agrarian Reformism. London: Faber and Faber. . 1977. "Agr-arian Transition and the Agrarian Question." Journal of Peasant Studies 4 (3) April. . 1981. "The New Technology, Class Formation and Class Action in the Indian Countryside." Journal of Political Studies 8 (4) July. Cain, M. 1981. Landlessness in India and Bangladesh: A Critical Review of Data Sources. Working Paper 71. New York: Population Council. Cancian, Frank. 1977. "The Innovators' Situation: Upper-Middle Class Conservatism in Agricultural Communities." Social Science Working Paper 132. School of Social Sciences, University of California, Irvine. November. Cartillier, M. 1975. "Role of Small-Scale Industries in Economic Develop- ment: Irrigation Pumpsets Industry in Coimbatore." Economic and Political Weekly 10 (44-45) November 1. Castillo, G. 1973. "All in a Grain of Rice: A Review of Philippine Studies on 390 Bibliography the Social and Economic Implications of the New Rice Technology." Uni- versity of the Philippines, Los Bafios. December. Center for Management of Indian Economics. 1986. Basic Statistics Relating to Indian Economy. Bombay. Cernea, Michael M., ed. 1985. Putting People First: Sociological Variables in Rural Development. New York: Oxford University Press. Cernea, Michael M., John K. Coulter, and John E A. Russell, eds. 1984. Agri- cultural Extension by Training and Visit: The Asian Experience. Washington, D.C.: World Bank. Chaddha, G. K. 1978. "Farm Size and Productivity Revisited: Some Notes from Recent Experience of Punjab." Economic and Political Weekly 13 (39) September 30: A87-A96. Chadha, V. 1984. "The Landless and the Poor in Green Revolution Regions in India." Agricultural Situation in India 39 (5) August: 295-303. Chakravarty, Aparijita. 1981. "Tenancy and Mode of Production." Economic and Political Weekly 16 (13) March 28. Chakravarty, Aparijita, and A. Rudra. 1973. "Economic Effects of Tenancy: Some Negative Results." Economic and Political Weekly 13 (28) July 14: 1239-46. Chakravarty, Sukhamoy. 1987. Development Planning: The Indian Experience. Oxford: Clarendon Press. Chambers, Robert. 1979. Health, Agriculture and Rural Poverty: Why Seasons Matter. Discussion Paper 148. Brighton: Institute of Development Studies, University of Sussex. - . 1981. "A Lesson for Rural Developers: The Small Farmer Is a Profes- sional." Development Digest 19 (3) July: 3-12. Chandrashekhar, D., K. Murti, and S. Ramaswamy. 1987. "Social Forestry in Karnataka." Economic and Political Weekly 22 (24) June 13: 935-41. Chattopadhyay, M. 1979. "Relative Efficiency of Owner and Tenant Cultiva- tion: A Case Study." Economic and Political Weekly 14 (39) September 29: A93-A96. Chattopadhyay, M., and A. Rudra. 1976. "Size-Productivity Revisited." Eco- nomic and Political Weekly 11 (39) September 25: A 100-16. - 1977. "Size-Productivity Revisited-Addendum." Economic and Politi- cal Weekly 12 (11) March 12: 476-89. Chaudhri, D., and A. Dasgupta. 1985. Agriculture and the Development Process. Kent: Croom Helm. Chaudhry, M. G. 1982. "The Green Revolution and the Redistribution of Rural Incomes: Pakistan's Experience." Pakistan Development Review 21 (3) Autumn: 173-205. - . 1983. "The Green Revolution and the Redistribution of Rural In- comes: Pakistan's Experience-A Reply." Pakistan Development Review 22 (2) Summer: 117-23. Bibliography 391 1986. "Mechanization and Agricultural Development in Pakistan." Pakistan Development Review 25 (4) Winter: 431-46. Chaudhry, M. G., and M. Anwar Chaudhry. 1974. "Cost of Living In- dexes for Rural Labourers in Pakistan." Pakistan Development Review, Spring. Chaudhry, P. K. 1960. "Ceilings on Land Holdings: August Measures to Hoodwink People." Economic and Political Weekly 12 (12) March 19. Chaudhry, Rafique Huda. 1978. "Some Aspects of Seasonal Dimensions to Rural Poverty in Bangladesh." Paper presented at the Conference on Sea- sonal Dimensions to Rural Poverty, Institute of Development Studies, Uni- versity of Sussex, Brighton, July 3-6. Chaudry, N. K. 1976. "Farm Efficiency under Semi-Feudalism: A Critique of Marginalist Theories and Some Marxist Formulations." Economic and Politi- cal Weekly 11 (39) September. Chauhan, K., S. Mundle, N. Mohanan, and D. Jadhav. 1973. Small Farmers: Problems and Possibilities of Development: A Study in Sangli District of Maharashtra. Ahmedabad: Center for Management in Agriculture, Indian Institute of Management. Chawdhari, T. P. S., and others. 1969. Resource Use and Productivity on Farms: A Comparative Study of Intensive and Non-Intensive Agriculture Areas. Hydera- bad: National Institute of Community Development. Chayanov, A. N. 1976 [1929]. The Theory of the Peasant Economy. B. Kerblay, D. Thorner, and R. Smith, eds. Homewood, Ill.: Irwin. Chen, Lincoln. 1975. "An Analysis of Per Capita Foodgrain Availability, Con- sumption and Requirements in Bangladesh: A Systematic Approach to Food Planning." Bangladesh Development Studies 3 (2) April: 93-126. - , ed. 1973. Disaster in Bangladesh: Health Crisis in aDevelopingNation. New York: Oxford University Press. Cheung, S. N. S. 1968. "Private Property Rights and Sharecropping."Journal of Political Economy (76): 1107-22. . 1969. The Theory of Share Tenancy. Chicago, Ill.: University of Chicago Press. Child, F C., and H. Kaneda. 1975. "Links to the Green Revolution: A Study of Small-Scale Agriculturally Related Industry in Pakistan Punjab." Eco- nomic Development and Cultural Change 23 Uanuary): 249-75. Chothani, A. n.d. "Identifying the Critical Components in a Successful Coop- erative." National Dairy Development Board, Anand. Processed. Choudhry, K. M. 1965. "Factors Affecting Acceptance of Improved Agricul- tural Practices: A Study of an I.A.D.P. District in Rajasthan." Agro-Eco- nomic Research Center, Vallav, Vidyanagar. Chowdhury, A. 1978. A Bangladesh Village: A Study of Social Stratification. Dhaka: Center for Social Studies. Chowdhury, A., T. Alauddin, and L. C. Chen. 1977. "The Interaction of Nu- 392 Bibliography trition, Infection and Mortality during Recent Food Crises in Bangladesh." Food Research Institute Studies 16 (2): 47-61. Chowdhury, A., and M. Haque. 1976. "Agrarian Hierarchy in a Kushtia Vil- lage." Indian Rural Development Project Paper. Dhaka: University of Dhaka. Chuta, Enyinna, and Carl Liedholm. 1979. Rural Non-Farm Employment: A Re- view of the State of the Art. MSU Rural Development Paper 4. East Lansing: Department of Agricultural Economics, Michigan State University. Clay, Edward J. 1975. "Equity and Productivity Effects of a Package of Tech- nical Innovations and Changes in Social Institutions: Tubewells, Tractors and High-Yielding Varieties." Indian Journal of Agricultural Economics 30 (4) October-December. . 1976. "Institutional Change and Agricultural Wages in Bangladesh." Bangladesh Development Studies 4 (4) October. . 1978. "Environment, Technology and the Seasonal Patterns of Agri- cultural Employment in Bangladesh." Paper presented at the Conference on Seasonal Dimensions to Rural Poverty, Institute of Development Stud- ies, University of Sussex, Brighton, July 3-6. - 1979. "The Rice Harvester Revisited: A Micro Study of Forms of Pay- ment to Agricultural Labor in Bangladesh." Agricultural Development Council and International Crop Research Institute for Semi-Arid Tropics, Hyderabad. August. Clay, Edward J., and M. S. Khan. 1977. Agricultural Employment and Underem- ployment in Bangladesh: The Next Decade. Agricultural Economics and Rural Social Science Paper 4. Dhaka: Bangladesh Agricultural Research Council. Colmenares, J. 1975. Adoption of Hybrid Seeds and Fertilizers among Colombian Corn Growers. Mexico City: Centro Internacional de Mejoramiento de Maiz y Trigo (cIMMYT). Connell, J., and others. 1976. Migration from Rural Areas: The Evidence from Village Studies. Delhi: Oxford University Press. Cownie, J., B. F Johnston, and B. Duff. 1970. "The Quantitative Impact of the Seed-Fertilizer Revolution in West Pakistan: An Exploratory Study." Food Research Institute Studies 1 (January). Critchfield, R. 1978. "India Ready to Reap a Farming Revolution." Agricul- tural World, May. . 1979. The Changing Peasant: Part III, The Modern Farmer. American University Field Staff Report 46. Hanover, N.H. - 1980. "The Greening of Ghungrali." Asia, May-June. Crotty, R. 1980. Cattle, Economics and Development. Commonwealth Agricul- tural Bureau. Old Woking, Surrey: Gresham Press. Dalrymple, Dana G. 1971. Survey of Multiple Cropping in Less Developed Nations. FEDR-12. Washington, D.C.: Foreign Economic Development Service, U.S. Department of Agriculture. - . 1975. Measuring the Green Revolution: The Impact of Research on Wheat Bibliography 393 and Rice Production. Foreign Agricultural Economic Report 106. Washing- ton, D.C.: U.S. Department of Agriculture. . 1976; 6th ed. 1978. Development and Spread of High-Yielding Varieties of Wheat and Rice in Less Developed Nations. Foreign Agricultural Economic Re- port 95. Washington, D.C.: Economic Research Service, U.S. Department of Agriculture and U.S. Agency for International Development. Damodaran, A. 1987. "Structural Dimensions of the Fodder Crisis." Economic and Political Weekly 22 (13) March 28: A 16. Dandekar, V. M. 1962. "Economic Theory and Agrarian Reform." Oxford Economic Papers 12 (February): 69-80. . 1964. "Prices, Production and Marketed Supply of Foodgrain." Indian Journal of Agricultural Economics 19 (July-December): 186-95. . 1976. "Crop Insurance in India." Economic and Political Weekly 11 (26) June: A61-A80. . 1986. "Agriculture, Employment and Poverty." Economic and Political Weekly 21 (38-39) September 20-27: A90-A1O1. Dandekar, V. M., and N. Rath. 1971. "Poverty in India." Economic and Political Weekly 6 (1-2) January 2-9. Dandekar, V. M., and M. Sathe. 1980. "Employment Guarantee Scheme and Food for Work Programme." Economic and Political Weekly 15 (5) April 12: 707-13. Dantwala, M. L. 1958. "Prospects and Problems of Land Reforms in India." Economic Development and Cultural Change 6 (1) October. 1962. "Failure in Land Reform." Janta 17 (1) January 26. 1972. "Financial Implications of Land Reforms: Zamindari Aboli- tion." Indian Journal of Agricultural Economics 17 (4) October-December. 1973. "Poverty and Unemployment in Rural India." Report of the Study Conducted with the Assistance of the International Development Re- search Center, Ottawa, Canada, September. . 1985. "Garibi Hatao: Strategy Options." Economic and Political Weekly 20 (11) March 16. . 1986. "Agrarian Structure and Agrarian Relations in India." In In- dian Society of Agricultural Economics, Indian Agricultural Development since Independence. Bombay: Oxford University Press and IBH Publishing Co. . 1987a. "Growth and Equity in Agriculture." IndianJournal of Agricul- tural Economics 42 (2) April-June: 149-60. . 1987b. "Rural Assets: Distribution and Composition of the Labour Force." Indian Journal of Agricultural Economics 42 (3) July-September: 294-301. Dantwala, M., and C. Shah. 1971. "Evaluation of Land Reforms." University of Bombay. Dantwala Committee. 1970. Report of the Committee of Experts on Unemployment Estimates. New Delhi: Government of India, Planning Commission. 394 Bibliography Darling, M. 1977 [1925]. The Punjab Peasant in Prosperity and Debt. Reprinted with a new introduction. New Delhi: Oxford University Press. Dasgupta, Biplab. 1977a. Agrarian Change and the New Technology in India. Ge- neva: United Nations Research Institute for Social Development. - 1977b. Village Society and Labor Use. Delhi: Oxford University Press. Dasgupta, B., and R. Laishley. 1975. "Migration from Villages." Economic and Political Weekly October 18. Datta, S. 1980a. "Choice of Agricultural Tenancy in the Presence of Uncer- tainty and Transaction Costs." Ph.D. diss., University of Rochester, Roches- ter, New York. - 1980b. Sharecropping as a "Second-Best" Form of Tenancy in Traditional Agriculture. Modelling Research Group Working Paper 8011. Los Angeles: Department of Economics, University of Southern California. Datta, S., and D. O'Hara. 1980. Choice of Agricultural Tenancy in the Presence of Transaction Costs. Modelling Research Group Working Paper 8010. Los Angeles: Department of Economics, University of Southern California. Day, R. H. 1967. "The Economics of Technological Change and the Demise of the Share-Cropper." American Economic Review 57 (June): 427-49. Day, R. H., and Inderjit Singh. 1977. Economic Development as an Adaptive Pro- cess: The Green Revolution in the Indian Punjab. New York: Cambridge Univer- sity Press. de Haan, H. 1980. Rural Industrialization in India. Discussion Paper 54. Rotter- dam: Center for Development Planning, Erasmus University. de Janvry, A., and K. Subbarao. 1986. Agricultural Price Policy and Income Distribution in India. New Delhi: Oxford University Press. de Kruijk, H. 1986a. When Poverty Declines and Inequality Increases: The Case of Pakistan during the 1970s. Discussion Paper 37. Rotterdam: Center for De- velopment Planning, Erasmus University. - 1986b. "Income Inequality in the Four Provinces of Pakistan." Paki- stan Development Review 25 (4) Winter: 685-705. Deolalikar, A. 1981. "The Inverse Relationship between Productivity and Farm Size: A Test Using Regional Data from India." American Journal of Ag- ricultural Economics 63 (2) May. Desai, B. M., ed. 1979. Intervention for Rural Development: Experiences of the Small Farmer Development Agency. Ahmedabad: Institute of Management. Desai, Gunvant M. 1971. "Some Observations on the Economics of Cultivat- ing High-Yielding Varieties of Rice in India." In Viewpoints on Rice Policy. Los Bafios: International Rice Research Institute. de Silva, M. W A. 1977. "Structural Change in a Coastal Fishing Community in Southern Sri Lanka." Marga 4 (2): 67-88. Dhanagare, D. 1987. "Green Revolution and Social Inequalities in Rural India." Economic and Political Weekly 22 (19) May 21: 137-45. Dhar, P., and H. Lydall. 1961. The Role of Small Enterprises in Indian Economic Development. Bombay: Allied Publishers. Bibliography 395 Dhawan, B. D. 1977. "Tubewell Irrigation in the Gangetic Plains." Economic and Political Weekly 12 (39) September 24. . 1987. "How Stable Is Indian Irrigated Agriculture?" Economic and Po- litical Weekly 22 (39) September 26: A93-A97. Divatia, V. V. 1976. "Inequalities in Asset Distribution of Rural Households." Indian Reserve Bank Staff Occasional Papers 1 (1) June: 1-45. Dobbs, T, and P Foster. 1972. "Incentives to Invest in New Agricultural In- puts in Northern India." Economic Development and Cultural Change 21 (1) October. Dogra, B. 1986. "Can Seventh Plan Targets Be Achieved?" Economic and Politi- cal Weekly 21 (16) April 19: 689. Donald, Gordon. 1976. Credit for Small Farmers in Developing Countries. Boul- der, Colo.: Westview. Dorner, P., and D. Kanel. 1971. "The Economic Case for Land Reform: Em- ployment, Income Distribution and Productivity." In Land Reform, Land Set- tlements, and Cooperatives. Rome: Food and Agriculture Organization. Dutt, Kalyan. 1970. "A Critique of Land Redistribution." Mainstream 9 (5) Oc- tober 3. . 1981. "Operation Barga: Gains and Constraints." Economic and Politi- cal Weekly 16 (25-26) June 20-27: A58-A60. Dwivedi, H., and A. Rudra. 1973. "Economic Effects of Tenancy: Some Fur- ther Negative Results." Economic and Political Weekly July 21: 1291-94. Eckert, J. B. 1970. "The Beginnings of Change in West Pakistan's Agricul- ture." Ford Foundation, Lahore. June. . 1972. "Rural Labor in Punjab: A Survey Report." Punjab Planning and Development Department, Lahore. July. Eckert,J. B., and D. A. Khan. 1977. "Rural-Urban Labor Migration: The Evi- dence from Pakistan." Economic Research Institute, Lahore. Processed. Eckert, J. B., and others. 1973. "An Employment Strategy for Rural Areas of Pakistan." Paper by Ad Hoc Working Group, Economic Research Insti- tute, Lahore. Eckholm, E. 1975. The Other Energy Crises: Firewood. Worldwatch Paper 1. Washington, D.C.: Worldwatch Institute. . 1979. The Dispossessed of the Earth: Land Reform and Sustainable Develop- ment. Worldwatch Paper 30. Washington, D.C.: Worldwatch Institute. Elkinton, C. 1970. "Land Reform in Pakistan." In Land Reform in Turkey, Paki- stan and Indonesia. USAID Spring Review Country Paper. Washington, D.C.: U.S. Agency for International Development. Emmerson, Donald K. 1980. Rethinking Artisanal Fisheries Development: Western Concepts, Asian Experiences. World Bank Staff Working Paper 423. Washing- ton, D.C. Epstein, T. 1973. South India: Yesterday, Today and Tomorrow. London: Macmil- lan. Ercelawn, A. 1984. "Income Inequality in Rural Pakistan: A Study of Sam- 396 Bibliography ple Villages." Pakistan Journal of Applied Economics 3 (1) Summer: 1-29. Esman, M. J. 1978. "Landlessness and Near-Landlessness in Developing Countries." Rural Development Committee, Cornell University, Ithaca, N.Y September. Faiz, Mohammed. 1986. "Wealth Effects of the Green Revolution." Pakistan Development Review 25 (4) Winter: 489-511. Falcon, Walter P. 1970. "The Green Revolution: Generation of Problems." American Journal of Agricultural Economics 52 (December). Falcon, Walter P, and Gustav E Papanek, eds. 1971. Development Policy II: The Pakistan Experience. Cambridge, Mass.: Harvard University Press. FAO (Food and Agriculture Organization). 1973. Monthly Bulletin of Agricul- tural Economics and Statistics, January. - 1977. Report of the Seminar on Agricultural Perspective Planning, Islama- bad. Rome. FAQ and UNDP (United Nations Development Programme). 1975. Pakistan: Re- port of FAO Sector Programming Mission. FAO Sector Programming Mission Se- ries 6, PAK/75/010. Rome: Food and Agriculture Organization. Farmer, B., ed. 1977. Green Revolution? Boulder, Colo.: Westview Press. Farouk, A., and others. 1976. The Vagrants of Dacca City. Dhaka: Bureau of Economic Research, University of Dhaka. Feder, Gershon. 1979. "Adoption of Interrelated Agricultural Innovations: Credit, Risk and Scale Effects." Development Research Center, World Bank, Washington, D.C. December. Processed. - 1980. "Farm Size, Risk Aversion, and the Adoption of New Technol- ogy under Uncertainty." Oxford Economic Papers 32 (2) July: 263-82. Feder, Gershon, Richard just, and David Zilberman. 1981. Adoption of Agricul- tural Innovations in Developing Countries: A Survey. World Bank Staff Working Paper 444. Washington, D.C. Rev. ed. 1982, World Bank Staff Working Paper 542. Feder, Gershon, and Gerald O'Mara. 1981. "On Information and Innovation Diffusion: A Bayesian Approach." Development Research Center, World Bank, Washington, D.C. January. Processed. Feder, Gershon, and Roger Slade. 1984. Aspects of the Training and Visit System of Agricultural Extension in India: A Comparative Analysis. World Bank Staff Working Paper 656. Washington, D.C. Feder, Gershon, Roger Slade, and Lawrence Lau. 1985. The Impact of Agricul- tural Extension: A Case Study of the Training and Visit Systems in Haryana, India. World Bank Staff Working Paper 756. Washington, D.C. Feder, Gershon, Roger Slade, and A. Sundaram. 1986. "The T and V Exten- sion System: An Analysis of Operations and Effects." Agricultural Adminis- tration 21: 33-59. Fei, John C. H., Gustav Ranis, and Shirley W. Y. Kuo. 1979. Growth with Eq- uity: The Taiwan Case. New York: Oxford University Press. Franda, M. 1979. Small Is Politics: Organizational Alternatives in India's Rural Bibliography 397 Development. American University Field Staff. Ncw Delhi: Wiley Eastern Limited. Frankel, E 1971. India's Green Revolution: Economic Gains and Political Costs. Princeton, N.J.: Princeton University Press. Gadre, N., and B. Sapte. 1980. "Impact of Dairy Projects on Rural Economy: A Case Study." Indian Journal of Agricultural Economics 35 (4) October-De- cember: 173. Gafsi, S. 1976. Green Revolution: The Tunisian Experience. Mexico City: Centro Internacional de Mejoramiento de Maiz y Trigo (ciMMYr). Galgalikar, V., N. Gadre, and C. Joshi. 1976. "Changes in the Structural Dis- tribution of Land Holdings: A Case Study." Indian Journal of Agricultural Economics 3 (July-September): 32-40. Gangadharan, T P. 1980. "Feed Economy in Milk Production: A Probe under New Dairy Farm Technology in Kerala." IndianJournal of Agricultural Economics, October-December. Garg,J., and M. Azad. 1975. "Economics of Cross-Bred Cows: A Case Study." Indian Journal of Agricultural Economics 30 (3): 149-50. Garg, J., and V. Prasad. 1975. "An Economic Investigation into the Problems of the Cooperative Milk Board, Kanpur." Indian Journal of Agricultural Eco- nomics 30 (3) July-September: 143. George, M. V., and P. T Joseph. 1976. "Changes in the Structure of Land Ownership and Use in Kerala and Their Economic Implications." Indian Journal of Agricultural Economics 31 (3) July-September: 27-31. George, P. S. 1985. Some Aspects of Procurement and Distribution of Foodgrains in India. Working Paper on Food Subsidies. Washington, D.C.: Interna- tional Food Policy Research Institute. . 1986. "Emerging Trends in Size Distribution of Operational Holdings in Kerala." Economic and Political Weekly 21 (5) February 1. George, P. S., and V. V. Choukidar. 1972. Production and Marketing Pattern of Paddy. CMA Monograph 27. Ahmedabad: Center for Management in Ag- riculture, Indian Institute of Management. . 1973. Dynamics of the Paddy-Rice System in India. CMA Monograph 42. Ahmedabad: Center for Management in Agriculture, Indian Institute of Management. George, P. S., and V. K. Srivastava. 1975. "Institutional Finance for Dairy De- velopment." Indian Journal of Agricultural Economics 30 (3) July-September: 90-96. George, Shanti. 1985. Operation Flood: An Appraisal of Current Indian Dairy Pol- icy. New Delhi: Oxford University Press. . 1987. "Stemming Operation Flood: Towards an Alternative Dairy Policy for India." Economic and Political Weekly 22 (39) September 26: 1654- 64. Ghafar, M., M. Amir, and E. Clark. 1968. Statistical Series on Private Tubewell Development in West Pakistan: Size of Holding of Private Tubewell Owners. Re- search Report 69. Karachi: Pakistan Institute of Development Economics. 398 Bibliography Ghose, A. K. 1979a. "Farm Size and Land Productivity in Indian Agriculture: A Reappraisal." Journal of Development Studies 16 (1) October. -. 1979b. "Institutional Structure, Technological Change and Growth in a Poor Agrarian Economy: An Analysis with Reference to Bengal and Pun- jab." World Development 7 (April-May): 385-96. - 1980. "Wages and Employment in Indian Agriculture." World Develop- ment 8 (5-6) May-June. - 1987. "Agricultural Research and Rural Development." Economic and Political Weekly 22 (14-15) April 4. - 1988. "The 1988-89 Budget: Does It Tackle Any of India's Prob- lems?" Economic and Political Weekly 23 (14-15) April 2-9: 711-19. Ghose, A. K., and A. Saith. 1976. "Indebtedness, Tenancy and the Adoption of New Technology in Semi-Feudal Agriculture." World Development 4 (4) April. Ghosh, Arabinda. 1973. "Size-Structure, Productivity and Growth: A Case Study of West Bengal Agriculture." Bangladesh Economic Review 1 (1) Janu- ary: 59-70. Ghosh, M. G. 1970. "A Study of High-Yielding Varieties Programme in the District Birbhum, West Bengal, with Reference to Kharif Paddy, 1968-69." Agro-Economic Research Center, Vishwa-Bharati, Shantiniketan. - 1986. "Farm-Size Productivity Nexus under Alternative Technology." Indian Journal of Agricultural Economics 41 (1) January-March: 17-29. Ghosh, Ratan. 1976. "Effect of Agricultural Legislation on Land Distribution in West Bengal." Indian Journal of Agricultural Economics 31 (3) July-Septem- ber: 40-46. Gill, M. S. 1976. "The Green Revolution: Success in the Indian Punjab." Pro- gressive Farming December: 7-15. Gotsch, Carl H. 1971. "The Distributive Impact of Agricultural Growth: Low Income Farmers and the System." Paper presented to the Seminar on Small-Farmer Development, Agricultural Development Council and Ohio State University, Columbus, September. - 1972. "Technical Change and the Distribution of Incomes in Rural Areas." American Journal of Agricultural Economics 54 (2) May. - 1973. "Tractor Mechanization and Rural Development in Pakistan." International Labour Review 107 (2) February. - 1977. "Agricultural Mechanization in the Punjab: Some Comparative Observations from India and Pakistan." Food Research Institute, Stanford, Calif. Draft. Gotsch, Carl H., and others. 1975. "Linear Programming and Agricultural Policy: Micro Studies of the Pakistan Punjab." Food Research Institute Studies 14 (1). Grawe, R., J. Krishnamurti, and J. Baah-Dwomoh. 1979. "India: Employ- ment and Employment Policy: A Background Paper." South Asia Projects Department, World Bank, Washington, D.C. October. Grewal, S. S., and P. S. Rangi. 1980. "Economics and Employment of Dairying Bibliography 399 in the Punjab." Indian Journal of Agricultural Economics, October-December. Griffin, Keith. 1974. The Political Economy of Agrarian Change: An Essay on the Green Revolution. Cambridge, Mass.: Harvard University Press. Griffin, Keith, and A. K. Ghose. 1979. "Growth and Impoverishment in the Rural Areas of Asia." World Development 7 (4-5) April-May. Guha, H. 1980. "Operation Flood II: Some Constraints and Implications: A Comment." Economic and Political Weekly 15 (17) April 26. Guisinger, S., and N. Hicks. 1978. "Long-Term Trends in Income Distribu- tion in Pakistan." World Development 6: 1271-80. Gupta, D. 1980. "A Critical Evaluation of Government Policies and Programs of Rural Industrialization with Special Reference to the Punjab Region in Northern India." Institute of Economic Growth, Delhi. Draft. Gupta, T., and D. Mohan. 1979. Economics of Trees vs. Annual Crops on Mar- ginal Agricultural Lands. CMA Monograph 81. Ahmedabad: Center for Man- agement in Agriculture, Indian Institute of Management. Gupta, V. K., and D. P. Mathur. 1976. Wheat Production, Marketing and Pro- curement. CMA Monograph 58. Ahmedabad: Center for Management in Ag- riculture, Indian Institute of Management. Gupta, V. K., and R. K. Pandey. 1975. "Consumption and Availability of Milk in India." Indian Journal of Agricultural Economics 30 (3) July-September: 142. Guttman, J. 1980. "Villages as Interest Groups: The Demand for Agricul- tural Extension Services in India." Kyklos 33 (1): 121-41. Haggblade, S., P. Hazell, and J. Brown. 1987. "Farm/Nonfarm Linkages in Rural Sub-Saharan Africa: Empirical Evidence and Policy Implications." Development Economics Department, World Bank, Washington, D.C. De- cember. Haider, A. S. 1977a. An Economic Survey and Analysis of Factors Affecting Farm Production and Income in Selected Areas of Punjab. Faisalabad: Department of Agricultural Economics, University of Agriculture. . 1977b. Emerging Occupations in a Rural Setting: Selective Evidence from Punjab, Pakistan. Faisalabad: University of Agriculture. Halse, M. 1980. "Increasing the Incomes of Landless Labourers and Small- Holders: Part I, The Experience of India's Dairy Cooperatives." Agricul- tural Administration 7: 259-72. Hameed, N. D. Abdul, and others. 1977. Rice Revolution in Sri Lanka. Geneva: United Nations Research Institute for Social Development. Harriss, B. 1972. "Innovation Adoption in Indian Agriculture: The High- Yielding Variety Program." Modern Asian Studies 6 (1) November. . 1987. "Regional Growth Linkages from Agriculture and Resource Flows in Nonfarm Economy." Economic and Political Weekly 22 (1-2) January 3-10. Havens, A. E., and W. Flinn. 1975. "Green Revolution Technology and Com- munity Development: The Limits of Action Programs." Economic Develop- ment and Cultural Change 23 (3) April. 400 Bibliography Hayami, Y., and R. Herdt. 1977. "Market Price Effects of Technological Change in Income Distribution in Semi-Subsistence Agriculture." American Journal of Agricultural Economics 59 (2): 245-56. Hazell, P 1984. "Sources of Increased Instability in Indian and U.S. Cereal Production."AmericanJournalofAgriculturalEconomics 60 (3) August: 302-11. Herdt, R., and C. Capule. 1983. Adoption, Spread, and Impact of Modern Rice Varieties. Manila: International Rice Research Institute. Herdt, R. W. 1980. "Changing Asian Rice Technology: Impact, Benefits and Constraints." Paper prepared for the World Bank Seminar on the Intro- duction and Sustenance of High-Yielding Varieties, Washington, D.C., Jan- uary 8. Herdt, R. W., and E. A. Baker. 1972. "Agricultural Wages, Production and the High-Yielding Varieties." Economic and Political Weekly 7 (13) March 25: A23-A30. Herdt, R. W., and A. Mandac. 1981. "Modern Technology and Economic Ef- ficiency of Philippine Rice Farmers." Economic Development and Cultural Change 29 (2) January. Herring, R., and M. G. Chaudhry. 1974. "The 1972 Land Reforms in Paki- stan and Their Economic Implications: A Preliminary Analysis." Pakistan Development Review 13 (3) Autumn: 245-79. Herring, Ronald. 1983. Land to the Tiller. New Haven, Conn.: Yale Univer- sity Press. Hinton, William. 1966. Fanshen: A Documentary of Revolution in a Chinese Vil- lage. New York: Monthly Review Press. IHirashima, S. 1978. The Structure of Disparity in Developing Agriculture. Tokyo: Institute of Development Economics. Hirschman, Albert 0. 1958. The Strategy of Economic Development. New Haven, Conn.: Yale University Press. Hirway, I. 1985. "Garibi Hatao: Can IRDP Do It?" Economic and Political Weekly 20 (13) March 30. Ho, Samuel P S. 1979. "Decentralized Industrialization and Rural Develop- ment: Evidence from Taiwan." Economic Development and Cultural Change 28 (1) October: 79-96. Hodgdon, L. 1966. Adoption of Agricultural Practices in Madhya Pradesh: Factors Associated with the Adoption of Recommended Agricultural Practices in Two Vil- lages. Hyderabad: Hyderabad National Institute of Community Develop- ment. Honavar, R., and others. 1984. An Economic Assessment of Poverty Eradication and Rural Unemployment Alleviation Programs and Their Prospects. Madras: In- stitute for Financial Management and Research. Hopper, W. D. 1965. "Allocative.Efficiency in a Traditional Indian Agricul- ture." Journal of Farm Economics 47 (August): 611-29. - 1977. "Distortions to Agricultural Development Resulting from Gov- ernment Prohibitions." Paper presented to workshop on Constraints in Bibliography 401 World Agricultural Production with Special Reference to Distortion of In- centives, University of Chicago, Illinois, September. Hossain, Mahabub. 1977a. "Agrarian Structure and Land Productivity in Bangladesh." Ph.D. diss., Cambridge University. . 1977b. "Farm Size, Tenancy and Land Productivity: An Analysis of Farm Level Data in Bangladesh Agriculture." Bangladesh Development Studies 5 (3) July: 285-348. . 1986. "A Note on the Trend of Landlessness in Bangladesh." Bangla- desh Development Studies 14 (2) June: 95-100. . 1988. CreditforAlleviation of Rural Poverty: The GrameenBank inBangla- desh. Research Report 65. Washington, D.C: International Food Policy Re- search Institute. Huang, C. J., and F. S. Bagi. 1980. "Stochastic Frontier Production Functions and Average Efficiency in Indian Agriculture." Department of Rural De- velopment, Tennessee State University, Nashville. October. Huffman, W. E. 1977. "Allocative Efficiency: The Role of Human Capital." Quarterly Journal of Economics 91 (February): 59-79. . 1981. "Black-White Human Capital Differences: Impact on Agricul- tural Productivity in the U.S. South." American Economic Review 71 (1) March: 94-117. Huq, A., ed. 1976. Exploitation and the Rural Poor. Comilla: Bangladesh Acad- emy for Rural Development. Huria, V. K., and K. T. Acharya. 1980. "Dairy Development in India: Some Critical Issues." Economic and Political Weekly 15 (45-46) November 8-15. Hussain, A., and M. Muazzami. 1973. "Co-operative Farming in Bangladesh." Report of the Workshop. Bangladesh Agricultural University, Mymen- singh. April. Hussain, S. M. 1970. "Price Incentives for the Production of High-Yielding Mexican Varieties of Wheat." Pakistan Development Review, Winter. IFDC (International Fertilizer Development Center). 1980. "Bangladesh: Eq- uity Effects of Fertilizer Use." Joint Report of the Bangladesh Agricultural Research Council and the IFDC, Muscle Shoals, Alabama. Draft. . 1981. Agricultural Production, Fertilizer Use, and Equity Considerations: Results and Analysis of Farm Survey Data, 1979-80. Muscle Shoals, Alabama. . 1983. Bangladesh: Policy Options for the Development of the Fertilizer Sector. Muscle Shoals, Alabama. IFPRI (International Food Policy Research Institute). 1985. Development Impact of the Food for Work Program in Bangladesh. Technical Papers. Washington D.C. ILO (International Labour Organisation). 1971. Problems of Employment Promo- tion in Pakistan. Geneva. . 1972. Employment, Income and Equality: A Strategy forDiscoveringEmploy- ment in Kenya. Geneva. 402 Bibliography - 1974. Sharing in Development: A Programme of Employment, Equity and Growth in the Philippines. Geneva. - 1977a. Poverty and Landlessness in Rural Asia. Keith Griffin and A. R. Khan, eds. Geneva. - 1977b. Report on the National Seminar on Landless Labour andJob Oppor- tunities in Rural Areas. Dhaka. . 1978. Labor Absorption in Indian Agriculture. Bangkok. . 1980. Problems of Rural Workers in Asia and the Pacific. Asian Regional Conference in Manila, Report 3. Geneva. ILO and ARTEP (Asian Regional Team for Employment Promotion). 1980. Em- ployment Expansion in Asian Agriculture: A Comparative Analysis of South Asian Countries. Bangkok: International Labour Office. India, Government of, and Government of Maharashtra. 1980. Joint Evalua- tion Report on Employment Guarantee Schemefor Maharashtra. New Delhi. India, Ministry of Agriculture and Irrigation. 1972. Interim Report on Ceiling on Land Holdings. Working Group, National Commission on Agriculture. New Delhi. 1975. All India Report on Agricultural Census, 1970-71. New Delhi. 1976. Report of the National Commission on Agriculture. New Delhi: Gov- ernment Press. India, Ministry of Agriculture and Rural Development. 1985a. Indian Agricul- ture in Brief. 20th ed. New Delhi: Directorate of Economics and Statistics, Department of Agriculture and Cooperation. - 1985b. Instructions for the Implementation of the Rural Landless Employ- ment Guarantee Programme. New Delhi. - . 1986. Concurrent Evalution of IRDP: The Main Findings of the Survey for October 1985-March 1986. New Delhi. India, Ministry of Labour. 1978. Rural Labour Enquiry, 1974-75: Summary Re- port. New Delhi: Labour Bureau. India, Ministry of Planning. 1961. "Tables with Notes on Some Aspects of Landholdings in Rural India, 16th Round (July 1960-June 1961)." Na- tional Sample Survey Organization, Department of Statistics. - 1976a. "Tables with Notes on Earnings, Indebtedness, Cultivated Holdings and Assets of the Weaker Section Households in Rural India, 25th Round: July 1970-June 1971." National Sample Survey Organization, Report 233. . 1976b. "Tables on Landholdings, 26th Round: July 1971-September 1972." All India and Statewide Reports for Punjab, Kerala, Gujarat, and Maharashtra. National Sample Survey Organization, Department of Statis- tics, February. - 1976c. "Some Summary Information on the Survey on Employment- Unemployment, 27th Round, October 1972-September 1973." National Sample Survey Organization, Report 255. Draft. Bibliography 403 . 1978. "Employment, Unemployment Situation at a Glance." Sarvekshana 2 (2) October. . 1982. "Tables on the Pattern of Land Ownership, 37th Round." Na- tional Sample Survey Organization, Department of Statistics. . 1983. "Tables on Unemployment, 38th Round." National Sample Survey Organization, Department of Statistics. . 1985. "Results of Indebtedness of Rural Labour Households, 32nd Round." Sarvehshana 7 (3-4) January-April. . 1986a. "Results of the Third Quinquennial Survey on Consumer Ex- penditure, 38th Round." Sarvekshana 9 (4) April: S1-S103. . 1986b. "Results on Assets and Liabilities of Rural and Urban House- holds: 37th Round, 1982." Sarvekshana 10 (1) July. . 1987. "Results on Indebtedness of Rural and Urban Households: 37th Round (January-December 1982)." Sarvekshana I I (1) July. . 1988. Economic Survey, 1987-88. New Delhi. India, Planning Commission. 1968. Report on Evaluation of High-Yielding Varieties Programme, Kharif( 1968. Delhi: Programme Evaluation Organi- sation. . 1971. Evaluation Study of the High-Yielding Varieties Programme: Report for the Rabi 1968-69: Wheat, Paddy, andJowar. Delhi: Programme Evaluation Organisation. . 1976. The High-Yielding Varieties Programme in India, 1970-75. Part II. New Delhi: Programme Evaluation Organisation and Australian National University. . 1982. Integrated Rural Development Programme: A Quick Preliminary Evaluation. New Delhi: Programme Evaluation Organisation. . 1985a. Evaluation Report on Integrated Rural Development Programme. New Delhi. . 1985b. The Seventh Five-Year Plan, 1985-90. New Delhi. Indian Society of Agricultural Economics. 1972. Problems of Farm Mechaniza- tion. Seminar Series 9. Bombay. . 1974. Rural Developmentfor Weaker Sections. Seminar Series 12. Bom- bay. Institute of Food and Nutrition Sciences. 1978. Economic and National Effects of Food for Relief Work Projects. Dhaka: University of Dhaka. . 1981. Nutrition Survey of Rural Bangladesh. Dhaka: University of Dhaka. IRRI (International Rice Research Institute). 1972. Rice, Science and Man. Los Bafios. 1975. Major Research in Upland Rice. Los Bafios. . 1978a. Economic Consequences of the New Rice Technology. Los Bafios. 1978b. Changes in Rice Farming in Selected Areas of Asia. Los Bafios. 404 Bibliography Isenman, P. I., and H. W. Singer. 1977. "Food Aid Disincentive Effects and Their Policy Implications." Economic Development and Cultural Change 25 (2) January: 205-37. Islam, M. M. 1978. Bengal Agriculture 1920-1946: A Quantitative Study. Cam- bridge: Cambridge University Press. Islam, Nurul. 1978. Development Strategy of Bangladesh. Oxford: Pergamon Press. . 1980. Poultry Raising in a Comilla Village: A Survey of Existing Facilities and Problems. Comilla: Bangladesh Academy for Rural Development. Islam, Rizwanul. 1979. "What Has Been Happening to Rural Income Distri- bution in Bangladesh?" Development and Change 10 (3) July. Jabbar, M. A. 1977. "Relative Productive Efficiency of Different Tenure Classes in Selected Areas of Bangladesh." Bangladesh Development Studies 5 (1) January. Jain, Shail. 1975. Size Distribution of Income: A Compilation of Data. Baltimore, Md.: Johns Hopkins University Press. Jannuzi, E Tomasson. 1974. Agrarian Crisis in India: The Case of Bihar. Austin: University of Texas Press. . 1976. Preliminary Report Concerning the Need for Further Research on the Hierarchy of Interests in Land in Bangladesh. Dhaka: U.S. Agency for Interna- tional Development. Jannuzi, E Tomasson, and James T Peach. 1977a. Report on the Hierarchy of Interests in Land in Bangladesh. Washington, D.C.: U.S. Agency for Interna- tional Development. -. 1977b. Summary Report of the 1977 Land Occupancy Survey of RuralBan- gladesh. Dhaka: Bureau of Statistics. - 1979. "A Note on Land Reforms in Bangladesh: The Efficacy of Ceilings." Journal of Peasant Studies 6 (3) April: 342-47; comment by K. Siddiqui, pp. 348-50. 1980. The Agrarian Structure of Bangladesh: Impediments to Development. Boulder, Colo.: Westview Press. Jayaraman, T. 1984. "A Review of the Special Credit Programme in an Irriga- tion Project in India." Agricultural Administration 16: 1-16. Jayasuriya, J., and R. Chand. 1986. "Technical Change and Labor Absorption in Asian Agriculture: Some Emerging Trends." World Development 14 (3): 415-18. Jaynes, G. D. 1979a. "The Economics of Land Tenure." Paper presented at the Conference on Adjustment Mechanisms of Rural Labor Markets in De- veloping Areas, Agricultural Development Council and International Crop Research Institute for Semi-Arid Tropics, Hyderabad, August 22-24. - 1979b. Production and Distribution in Agrarian Economics. Discussion Paper 309. New Haven, Conn.: Economic Growth Center, Yale University. Jha, L., and others. 1984. Report of the Evaluation Committee on Operation Flood II. New Delhi: Ministry of Agriculture. Bibliography 405 Jodha, N. S. 1972. "A Strategy for Dry Land Agriculture." Economic and Politi- cal Weekly 7 (13) March. . 1973. "Special Programmes for the Rural Poor: The Constraining Framework." Economic and Political Weekly 8 (13) March. . 1979. "Agricultural Tenancy in Semi-Arid Tropical Parts of India." International Crops Research Institute for the Semi-Arid Tropics, Hydera- bad. Draft. . 1986. "Common Property Resources and Rural Poverty in Dry Re- gions of India." Economic and Political Weekly 21 (27): 1169-81. Johl, S. S. 1971. Mechanization, Labor Use, and Productivity in Indian Agriculture. Occasional Paper 23. Columbus: Department of Agricultural Economics and Rural Sociology, Ohio State University. . 1975. "Gains of the Green Revolution: How They Have Been Shared in Punjab." Journal of Development Studies 2 (3) April. Johnson, D. 1950. "Resource Allocation under Share Contracts." Journal of Political Economy 58: 111-23. Johnston, B., and J. Cownie. 1969. "The Seed Fertilizer Revolution and Labor Force Absorption." American Economic Review 59 (4) September. Johnston, B., and P. Kilby. 1975. Agriculture and Structural Transformation. New York: Oxford University Press. Jose, A. V. 1974. "Trends in the Real Wage Rates of Agricultural Workers." Economic and Political Weekly 30 (3). . 1978. "Real Wages, Employment and Income of Agricultural Labour- ers." Economic and Political Weekly, Review of Agriculture 12 (42) March. Joshi, P C. 1961. "Problems and Prospects of Ceilings on Land Holdings in India." Agricultural Situation in India 15 (5) August. . 1974. "Agrarian Structure and the Rural Poor: Review of Perspec- tives." Social Change 4 (September-December). 1975. Land Reforms in India. Bombay: Allied Publishers. 1979a. "Perspectives on Poverty and Social Change: The Emergence of the Poor as a Class." Economic and Political Weekly 14 (7-8) February: 355-66. . 1979b. "Technological Potentialities of Peasant Agriculture." Economic and Political Weekly 14 (June): A47-A57. Joshi P., D. Bahl, and D. Jha. 1981. "Direct Employment Effect of Technical Change in Uttar Pradesh Agriculture." IndianJournal of Agricultural Econom- ics 36 (4) October-December. Juergensmeyer, M. 1979. "Cultures of Deprivation: Three Case Studies in Punjab." Economic and Political Weekly 14 (7-8) February: 255-62. Jul, M. 1979. "Unexpected Benefit from a Dairy Project." Food and Nutrition Bulletin 1 (3) May. Junankar, P. N. 1977. Tests of the Profit Maximisation Hypothesis: A Study of In- dian Agriculture. Discussion Paper 96. Colchester: Department of Econom- ics, University of Essex. 406 Bibliography - 1978a. "Do Indian Farmers Maximise Profits?" Department of Eco- nomics, University of Essex, Colchester. May. Draft. - 1978b. Profit Maximization: Translog Functions Applied to Indian Agricul- ture. Discussion Paper 313. Kingston, Ontario: Institute of Economic Re- search, Queens University. Kahlon, A. S. 1970. "Farm Technology: Its Implications in Agricultural Eco- nomics." Presidential Address, Indian Society of Agricultural Economics, October 23. Kahlon A. S., K. Dhawan, and G. Gill. 1975. "Relative Profitability of Dairy Enterprise vis-a-vis Crop Cultivation in the Punjab." Indian Journal of Agri- cultural Economics 30 (3) July-September: 120-28. Kahlon, A. S., and S. S. Johl. 1964. "Nature and Role of Risk and Uncertainty in Agriculture." Indian Journal of Agricultural Economics 19 (January- March). Kahlon, A. S., and G. Singh. 1973. "Social and Economic Implications of Large-Scale Introduction of High-Yielding Varieties of Wheat in Punjab with Special Reference to the Gurdaspur District." United Nations Devel- opment Programme Global Project. Department of Economics and Sociol- ogy, Punjab Agricultural University, Ludhiana. Kalirajan, K. 1981. "The Economic Efficiency of Farmers Growing High- Yielding Irrigated Rice in India." American Journal of Agricultural Economics 63 (3) August: 566-607. Kaneda, Hiromitsu. 1969. "Economic Implications of the Green Revolution and the Strategy of Agricultural Development in West Pakistan." Pakistan Development Review 10 (2) Summer. - 1972. "Economic Implications of the Green Revolution and the Strat- egy of Agricultural Development in West Pakistan." In Keith Griffin and A. R. Khan, eds., Growth and Inequality in Pakistan. New York: St. Martin's. Kaneda, Hiromitsu, and M. Ghaffar. 1970. "Output Effects of Tubewells on the Agriculture of the Punjab: Some Empirical Results." Pakistan Develop- ment Review 10 (1) Spring: 71-72. Kansal, S. 1981. "Data on Income Distribution in Bangladesh." Working Paper 1981-8. EPD Income Distribution Project, World Bank, Washington, D.C. July. Kapteyn A., T. Wansbeek, and J. Buyze. 1979. "Maximizing or Satisficing?" Review of Economics and Statistics 61 (November): 549-63. Karanjkar, S. V., and S. N. Soni. 1980. "Economics of Poultry Enterprise in Jabalpur." Indian Journal of Agricultural Economics 35 (4) October-Decem- ber: 170. Katyal, R., and V. Sood. 1976. "Real Growth in Agriculture at State Level." Indian Association for Research in National Income and Wealth, Tenth General Conference. January. Kaul, J., and S. Kahlon. 1971. "Problems of Marginal Farmers and Agricul- tural Laborers in Hoshiarpur District of Punjab." Department of Econom- ics and Sociology, Punjab Agricultural University, Ludhiana. Bibliography 407 Keesing, D. 1975. "Economic Lessons from China."Journal of Development Eco- nomics 2: 1-32. Kelkar, G. A., and S. Subramanian. 1977. "Pattern of Assets of Rural House- holds, 1961-71." Indian Reserve Bank Staff Occasional Papers 2 (1) June: 1-57. Khan, Akhtar Hameed. 1971. Tour of 20 Thanas. Comilla: Bangladesh Acad- emy for Rural Development. Khan, A. R. 1972. The Economics of Bangladesh. London: Macmillan. Khan, A. R., and others. 1981. Employment, Income and the Mobilisation of Local Resources: A Study of Two Bangladesh Villages. Asian Employment Pro- gramme. Bangkok: International Labour Organisation. Khan, Ahmed Syed, A. M. Chaudhry, and M. Aslam. 1969. "Economics of Modern Poultry Production in West Pakistan." Faculty of Agricultural Economics and Rural Sociology, West Pakistan Agricultural University. October. Khan, D. A. 1978. Employment and Occupational Change in the Rural Punjab: Consequences of the Green Revolution. Lahore: Economic Research Institute. Khan, M. H. 1975. The Economics of the Green Revolution in Pakistan, New York: Praeger. . 1977. "Land Productivity, Farm Size and Returns to Scale in Pakistan Agriculture." World Development 5 (4) April: 317-23. . 1979a. "Agrarian Institutions and Change in Pakistan." Department of Economics, Simon Fraser University, Vancouver. Draft. . 1979b. "Farm Size and Land Productivity Relationships in Pakistan." Pakistan Development Review 18 (1) Spring: 9-76. . 1981. Underdevelopment and Agrarian Structure in Pakistan. Boulder, Colo.: Westview Press. Khan, M. H., and D. R. Maki. 1979. "Effects of Farm Size on Economic Effi- ciency: The Case of Pakistan." American Journal of Agricultural Economics February: 64-69. . 1980. "Relative Efficiency by Farm Size and the Green Revolution in Pakistan." Pakistan Development Review 19 (1) Spring: 51-64. Khan, S., and others. 1977. Agricultural Employment and Underdevelopment in Bangladesh: The Next Decade. UNDP/FAO Working Paper. New York: United Nations Development Programme and Food and Agriculture Organiza- tion. Khan, T. M., and S. Bose. 1968. "Report on Income of Agricultural Workers in Pakistan." Pakistan Institute of Development Economics, Islamabad. No- vember. Processed. Khan, Waheeduddin, and V. B. R. S. Somasekhara Rao. 1974. "On Ten- ancy, Rent Systems and Improved Technology: An Investigation in West Godavari District." Community Development and Panchayati Raj Digest, April. Khan, Waheeduddin, and R. N. Tripathy. 1972. "Intensive Agriculture and 408 Bibliography Modern Inputs: Prospects of Small Farmers-A Study in West Godavari Dis- trict." National Institute of Community Development, Hyderabad. Khandker, R. H. 1973. "Distribution of Income and Wealth in Pakistan." Paki- stan Economic and Social Review 11 (1) Spring. Khasnabis, Ratan. 1981. "Operation Barga: Limits to Democratic Reform- ism." Economic and Political Weekly, Review of Agriculture 16 (25-26) June 20-27: A43-A48. Khilji, N. 1986. "Optimum Resource Utilization in Pakistan's Agriculture." Pakistan Development Review 25 (4) Winter: 469-86. Khusro, A. M. 1964. "Returns to Scale in Indian Agriculture." Indian Journaul of Agricultural Economics 19 (3-4) July-December. Also in Khusro (1973). - 1965. "Land Reforms since Independence." In V. B. Singh, ed., Eco- nomic History of India, 1857-1956. Bombay: Allied Publishers. -. 1973. Economics of Land Reform and Farm Size in India. Institute of Eco- nomic Growth. Delhi: Macmillan. Kishore, D. 1986. "An Alternate Strategy for the Transfer of Technology with Special Reference to India." Agricultural Administration 21: 197-204. Korten, David C. 1980. "Community Organization and Rural Development: A Leasing Process Approach." Public Administration Review September- October: 480-511. Krishna, P. V., and S. C. Bandhopadhyay. 1975. "A Case Study of a Success- ful Dairy." Indian Journal of Agricultural Economics 30 (3) July-September: 145. Krishna, Raj. 1959. "Agrarian Reform in India: The Debate on Ceilings." Economic Development and Cultural Change 7 (April). 1 1967. "Government Operations in Foodgrains. Economic and Political Weekly 3 (37) September 16. - 1973a. "Unemployment in India." Economic and Political Weekly 8 (9) March 3. - 1973b. "Unemployment in India: Presidential Address." Indian Jour- nal of Agricultural Economics 28 (1) January-March. - 1974. "Unemployment in India." Teaching Forum 38 (March). Agricul- tural Development Council, New York. - 1975. "Measurement of the Direct and Indirect Effects of Agricul- tural Growth with Technical Change." In Lloyd G. Reynolds, ed., Agricul- ture in Development Theory. New Haven, Conn.: Yale University Press. - 1976. Rural Unemployment: A Survey of Concepts and Estimates for India. World Bank Staff Working Paper 234. Washington, D.C.. Krishna, Raj, and A. Chhibber. 1983. Policy Modelling of a Dual Grain Market: The Case of Wheat in India. Research Report 38. Washington, D.C.: Interna- tional Food Policy Research Institute. Krishnaji, N. 1971. "Wages of Agricultural Labour." Economic and Political Weekly, Review of Agriculture 6 (39) September 25: A35-A41. Bibliography 409 1987. "Agricultural Growth, Prices and Rural Poverty." Economic and Political Weekly 22 (26) June 27. Krishnamurty, N. S., and S. Subramanian. 1977. "Financial Assets of Rural Households." Indian Reserve Bank Staff Occasional Papers 2 (1) June: 84-135. Kumar, P., V. C. Mathur, and R. P. Singh. 1981. "Estimating Labor Demand Functions and Labor Absorption in Agriculture: A Case Study of Delhi Wheat Farms." Indian Journal of Agricultural Economics 36 (4) October- December: 7-13. Kurian, N. 1987. "IRDP: How Relevant Is It?" Economic and Political Weekly 22 (52) December 26: A161-A176. Kurien, C. T 1977. "Putting the Instruments of Rural Development into the Hands of Producers." Paper presented at the Ramon Magsaysay Award Workshop, National Dairy Development Board, Anand, February 7-9. . 1986. "Reconciling Growth and Social Justice Strategies and Struc- tures." In M. L. Dantwala and others, Rural Development: The Asian Experi- ence. New Delhi: Oxford University Press and IBH Publishing Co. Ladejinsky, Wolf. 1977. Agrarian Reform as Unfinished Business: The Selected Pa- pers of Wolf Ladejinsky. Louis J. Walinsky, ed. New York: Oxford University Press. Lakshminarayan, H. 1973a. "Small Farmers' Development Program: A Note." Economic and Political Weekly 8 (17) April 28. . 1973b. "The Social and Economic Implication of Large-Scale Intro- duction of High-Yielding Varieties of Wheat in Haryana." UNDP Global Project. Agricultural Economic Research Center, University of Delhi, New Delhi. . 1979. "Inter-Size Group and Inter-State Variations in Distribution of Agricultural Assets." Economic and Political Weekly 14 (26) June 30: A69-A74. Lakshminarayan, H., and S. Tyagi. 1976. "Some Aspects of Size Distribution of Agricultural Holdings." Economic and Political Weekly 11 (44) October 9: 1637-40. . 1977. "Inter-State Variations in Types of Tenancy." Economic and Polit- ical Weekly 12 (39) September 24. Lal, D. 1975. "The Agrarian Question." South Asian Review 8 (4) July-Octo- ber. .1976. "Agricultural Growth, Real Wages and the Rural Poor in India." Economic and Political Weekly, Review of Agriculture 11 (26) June 26. . 1977. "Agricultural Growth and Rural Real Wages: A Reply." Eco- nomic and Political Weekly 12 (20) May 14. Lau, L., and P. Yotopoulos. 1971. "A Test for Relative Efficiency and Applica- tion to Indian Agriculture." American Economic Review 61 (March): 94-107. Lawson, R. 1977. "New Directions in Developing Small-Scale Fisheries." Ma- rine Policy 1 (1) January: 45-51. 410 Bibliography Leaf, M. 1983. "The Green Revolution and Cultural Change in a Punjab Vil- lage, 1965-1978." Economic Development and Cultural Change 31 (2) January: 227-70. Lehman, D. 1977. "The Death of Land Reform." Paper presented at the World Bank Workshop on the Analysis of Distributional Issues in Develop- ment Planning, Bellagio, Italy, April. Lele, Uma. 1968. "The Traders of Sholapur." In John W. Mellor, T. L. Weaver, Uma Lele, and S. Simons, eds., Developing Rural India. Ithaca, N.Y: Cornell University Press. - 1971. Food Grain Marketing in India: Private Performance and Public Pol- icy. Ithaca, N.Y: Cornell University Press. - 1972. "Distributional Efficiency and Agricultural Price Policy: Food Grain Marketing in India." Teaching Forum Paper, Marketing Price Analy- sis and Trade no. 21. Agricultural Development Council, New York. Octo- ber. - 1974. "The Roles of Credit and Marketing in Agricultural Develop- ment." In Nurul Islam, ed., Agricultural Policy in Developing Countries. Pro- ceedings of a Conference of the International Economics Association, Bad Godesberg, Federal Republic of Germany. New York: Wiley. - 1976. "Considerations Related to Optimum Pricing and Marketing Strategies in Rural Development." Paper presented at the 16th Interna- tional Conference of Agricultural Economics, Nairobi, July-August. - 1977. "Cooperatives and the Poor: A Comparative Perspective." Paper presented at the Experts' Consultation on Cooperatives and the Poor, International Cooperative Alliance, Loughborough, U.K., July. Lele, Uma, and John W. Mellor. 1972a. "Jobs, Poverty and the Green Revolu- tion." International Affairs 47 (1): 20-32. . 1972b. Technological Change and Distributive Bias in a Dual Economy. Oc- casional Paper 43. Ithaca, N.Y: Department of Agricultural Economics, Cornell University. 1981. "Technological Change, Distributive Bias and Labor Transfer in a Two-Sector Economy." Oxford Economic Papers 33 (November): 426-41. Liedholm, Carl, and Enyinna Chuta. 1976. The Economics of Rural and Urban Small-Scale Industries in Sierra Leone. African Rural Economy Paper 14. East Lansing: Department of Agricultural Economics, Michigan State Univer- sity. Lipton, Michael. 1968. "The Theory of the Optimizing Peasant." Journal of Development Studies 4: 327-51. . 1974. "Towards a Theory of Land Reforms." In D. Lehman, ed., Peas- ants, Landlords and Governments: Agrarian Reform in the Third World. New York: Holmes and Meier. - 1976. "Agricultural Finance and Rural Credit in Poor Countries." World Development 4 (2) July. - 1977. Why Poor People Stay Poor: Urban Bias in World Development. Cam- bridge, Mass.: Harvard University Press. Bibliography 411 . 1978. "Inter-Farm, Inter-Regional and Farm/Non-Farm Income Dis- tribution: The Impact of the New Cereal Varieties." World Development 6 (3): 319-37. . 1979. "The Technology, the System and the Poor: The Case of the New Cereal Varieties." In Institute of Social Studies, Development of Societies: The Next Twenty-Five Years. Dordrecht, Netherlands: M. Nijhoff. . 1983a. Demography and Poverty. World Bank Staff Working Paper 623. Washington, D.C. . 1983b. Labor and Poverty. World Bank Staff Working Paper 616. Wash- ington, D.C. . 1983c. Poverty, Undernutrition, and Hunger. World Bank Staff Working Paper 597. Washington, D.C. . 1984. "Conditions of Poverty Groups and Impact on Indian Eco- nomic Development and Cultural Change: The Role of Labour." Develop- ment and Change 15: 473-93. . 1985. Land Assets and Rural Poverty. World Bank Staff Working Paper 744. Washington, D.C. .1988. Poor and the Poorest: Some Interim Findings. World Bank Discus- sion Paper 25. Washington, D.C. Lluch, Constantino, Alan A. Powell, and Ross A. Williams. 1977. Patterns in Household Demand and Saving. New York: Oxford University Press. Lockheed, Marlaine E., Dean T. Jamison, and Lawrence T Lau. 1980. "Farmer Education and Farm Efficiency: A Survey." Economic Development and Cultural Change 29 (1) October: 37-76. Lockwood, Brian. 1981. "Farm Mechanization in Pakistan: Policy and Prac- tice." Working Paper 17. The Consequences of Small Rice Farm Mechaniza- tion Project, Agricultural Development Council, Lahore. Lockwood, Brian, P. K. Mukherjee, and R. T. Shand. 1971. The High-Yielding Varieties Programme in India. Part I. Delhi: Planning Commission of India and Australian National University. Longhurst, R., and Michael Lipton. 1989. New Seeds and Poor People. Balti- more, Md.: Johns Hopkins University Press. Longhurst, R., and P. Payne. 1979. Seasonal Aspects of Nutrition: Review of Evi- dence and Policy Implications. Discussion Paper 5. Brighton: Institute of De- velopment Studies, University of Sussex. September. Lowdermilk, Max K. 1972. "Diffusion of Dwarf Wheat Production Tech- nology in Pakistan's Punjab." Ph.D. diss., Cornell University, Ithaca, N.Y. Lowdermilk, Max K., Alan C. Early, and David M. Freeman. 1978. Farm Irri- gation Constraints and Farmers' Responses: Comprehensive Field Survey in Paki- stan. Water Management Research Project, Technical Report 48A-E Fort Collins: Colorado State University, September. Lowdermilk, Max K., and David M. Freeman. 1977. "Organising Farmers to Improve Irrigation Water Delivery: The Problem and Prospects for Solu- tions in Pakistan." Pakistan Economic and Social Review 15 (Autumn-Winter): 152-73. 412 Bibliography Lowdermilk, Max K., R. Frey Scott, and David M. Freeman. 1979. "Cancian's 'Upper Middle Class Conservatism' Thesis: A Replication from Pakistan." Rural Sociology 44 (2): 420-30. Lucas, R. 1977. Sharing, Monitoring and Incentives: Marshallian Misallocation Re- assessed. Discussion Paper 10. Boston, Mass.: Department of Economics, Boston University. Ludden, D. 1978. "Patronage and Irrigation in Tamil Nadu: A Long-Term View." Social History Review 16 (3). - 1979a. "Dimensions of Agrarian Political Economy: Focus on Tamil Nadu, India." Peasant Studies 8 (4) Fall. - 1979b. "Ecological Zones and the Cultural Economy of Irrigation in Southern Tamil Nadu." South Asia: Jourmal of South Asian Studies 9. Mahendradev, S. 1986. "Growth of Labour Productivity in Agriculture." Eco- nomic and Political Weekly 21 (25-26) June 21-28. - 1987. "Growth and Instability in Foodgrains Production: An Inter- state Analysis." Economic and Political Weekly 22 (39) September 26: A82- A93. Mahmood, Moazam. 1977. "The Pattern of Adoption of Green Revolution Technology and Its Effects on Land Holdings in the Punjab." Pakistan Eco- nomic and Social Review 15 (1-2) Spring-Summer. Mahmud, W. 1979. "Foodgrain Demand Elasticities of Rural Households in Bangladesh: An Analysis of Pooled Cross-Section Data." Bangladesh Devel- opment Studies 7 (1) Winter. Mallum, D., and V. Gupta. 1977. Changing Pattern of Wheat Production and Marketing. CMA Monograph 67. Ahmedabad: Center for Management in Agriculture, Indian Institute of Management. Mandal, G. P., and M. G. Ghosh. 1976. Economics of the Green Revolution: A Study in East India. Muzaffarnagar: Asian Publishers. Marbro, R. 1971. "Employment and Wages in a Dual Agriculture." Oxford Economic Papers 23: 401-17. Marshall, A. 1961. Principles of Economics. 9th ed. Cambridge: Cambridge University Press. Masum, M. 1976. "Unemployment and Underemployment in Agriculture: A Case Study of Bangladesh." Ph.D. diss., School of Economics, Delhi Uni- versity. - -. 1979. "Unemployment and Underemployment in Bangladesh: A Micro Study." Bangladesh Development Studies 7 (2) Summer: 47-69. Mathai, P. 1972. "Rural Industrialization and the Maximization of Employ- ment Opportunities in India." Small Industry Bulletin for Asia and the Far East 9. Mathur, D. P., and V. K. Gupta. 1978. Management of Small Poultry Farms. CMA Monograph 73. Ahmedabad: Center for Management in Agriculture, In- dian Institute of Management. Mavalankar, N. 1971. "Economics of Consolidation in India: A Case Study Revised." Artha Vijnana 13 (4) December. Bibliography 413 Mavinkurve, B. S., and P. K. Sundernathan. 1975. The Small Farmers Develop- ment Agencies (1972-73): A Field Study. Bombay: Reserve Bank of India. Mazumdar, Dipak. 1963. "On the Economics of Relative Efficiency of Small Farms." Economic Weekly, Special Number (July): 1259-63. . 1965. "Size of Farm and Productivity: A Problem of Indian Peasant Agriculture." Economica 32 (126) May: 161-73. . 1975a. "The Theory of Sharecropping with Labor Market Dualism." Economica 42 (167) August: 261-71. . 1975b. The Theory of Urban Underemployment in Less Developed Countries. World Bank Staff Working Paper 198. Washington, D.C. . 1975c. The Urban Informal Sector. World Bank Staff Working Paper 211. Washington, D.C. . 1980. "A Descriptive Analysis of the Role of Small-Scale Enterprises in the Indian Economy." Development Economics Department, World Bank, Washington, D.C. Draft. McInerney, John, and Graham Donaldson. 1975. The Consequences of Farm Tractors in Pakistan. World Bank Staff Working Paper 210. Washington, D.C. Mead, C. 1981. Landless in India and Bangladesh: A Critical Review of Data Sources. Working Paper 71. New York: Population Council. Mehra, Shakuntala. 1976. Some Aspects of Labor Use in Indian Agriculture. Tech- nological Change in Agriculture Project. Occasional Paper 88. Ithaca, N.Y: Department of Agricultural Economics, Cornell University, June. Also In- dian Journal of Agricultural Economics 31 (4). Mellor, John W. 1976. The New Economics of Growth. Ithaca, N.Y.: Cornell Uni- versity Press. Mellor, John W., and Gunvant M. Desai, eds. 1985. Agricultural Change and Rural Poverty: Variations on a Theme by Dharm Narain. Baltimore, Md.: Johns Hopkins University Press. Mellor, John W., and Uma Lele. 1973. "Growth Linkages of the New Foodgrain Technologies." IndianJournal of Agricultural Economics 28 (1) Jan- uary-March: 35-55. Mellor, John W., and M. Mudahar. 1974a. Modernizing Agriculture, Employment and Economic Growth: A Simulation Model. Occasional Paper 75. Ithaca, N.Y: Department of Agricultural Economics, Cornell University. . 1974b. Simulating a Developing Economy with Modernizing Agriculture Sector: Implications for Employment and Economic Growth. Occasional Paper 76. Ithaca, N.Y.: Department of Agricultural Economics, Cornell Uni- versity. Mellor, John W., and others. 1968. Developing Rural India: Plan and Practice. Ithaca, N.Y: Cornell University Press. Mencher, J. 1974. "Conflicts and Contradictions in the Green Revolution: The Case of Tamil Nadu." Economic and Political Weekly 9 (February). . 1978. "Agrarian Relations in Two Rice Regions of Kerala." Economic and Political Weekly, Annual Number (February). 414 Bibliography Michie, Barry H. 1973. "Variations in Economic Behaviour and the Green Revolution: An Anthropological Perspective." Economic and Political Weekly 3 (26) June 30. Minhas, B. S. 1974. "Rural Development for the Weaker Sections: Experience and Lessons." In Rural Development for Weaker Sections. Seminar Series 12. Bombay: Indian Society of Agricultural Economics and Indian Institute of Management. Mishra, G. 1980. "Impact of Dairy Development in Rural Areas of Karnataka." Institute for Social and Economic Change, Bangalore. Pro- cessed. Mishra, V. N., and D. S. Tyagi. 1972. "An Economic Analysis of the HYV of Wheat: A Study of Kota District." Artha Vikas January 1. Mitra, Ashok. 1979. The Status of Women: Literacy and Employment. New Delhi: Indian Council of Social Science Research. Mitra, P. K. 1979. "A Theory of Interlinked Rural Transactions." Develop- ment Research Center, World Bank, Washington, D.C. August. Draft. Mohammad, Ghulam. 1965. "Private Tubewell Development and the Crop- ping Patterns in West Pakistan." Pakistan Development Review Spring: 25-26. Mohan, Rakesh. 1974. "Contribution of Research and Extension to Produc- tivity Change in Indian Agriculture." Economic and Political Weekly, Review of Agriculture 9 (39) September 18. Moore, J. R., S. S. Johl, and A. M. Khusro. 1973. Indian Foodgrain Marketing. New Delhi: Prentice Hall of India. Moorti, T 1971. "A Comparative Study of Well Irrigation in Aligaoli District, India." Cornell International Agricultural Development Bulletin, May 19. Moreland, W. 1929. The Agrarian Structure of Muslim India. Allahabad: Central Book Report. Moscardi, E. 1976. "A Behavioral Model for Decision under Risk among Small-Holding Farmers." Ph.D. diss., University of California, Berkeley. - 1979. "Methodology to Study Attitudes Towards Risk: The Puebla Project." In A. Valdes, G. S. Scobie, and J. Dillon, eds., Economics and the Design of Small-Farmer Technology. Ames: Iowa State University Press. Mudahar, M. S. 1973. Dynamic Models of Agricultural Development with Demand Linkages. Occasional Paper 59. Ithaca, N.Y: Department of Agricultural Economics, Cornell University. - 1974. Dynamic Analysis of Direct and Indirect Implications of Technological Change in Agriculture: The Case of Punjab, India. Occasional Paper 79. Ithaca, N.Y: Department of Agricultural Economics, Cornell University. Mukherjee, M. 1980. "Some Aspects of Agrarian Structure of Punjab 1925-47." Economic and Political Weekly 15 (26) June 28. Mukherjee, P. K. 1970. "The HYVP: Variables That Matter." Economic and Po- litical Weekly 5 (13) March 28: A15-A22. Mukherjee, Ramakrishna. 1961. "Rural Class Structure in West Bengal." In Bibliography 415 A. R. Desai, ed., Rural Sociology in India. Bombay: IBH Publishing Co. Mukherjee, Robin M. 1986. "Statistical Information on Final Consumption in India and the National Sample Survey." Economic and Political Weekly 21 (5) February 1: 206-09. Mukund, K. 1986. "Bank Credit and IRDP: Madhya Pradesh Experience." Economic and Political Weekly 21 (41) October 11. Muller, J. 1974. "On Sources of Measured Technical Efficiency: The Impact of Information." American Journal of Agricultural Economics 56 (4) Novem- ber: 730-38. Mundle, S. 1981. "Land, Labor and the Level of Living in Rural Bihar." Asian Regional Team for Employment Promotion, International Labour Organi- sation, Geneva. Processed. . 1982. "Land, Labor and the Level of Living in Rural Punjab." Asian Regional Team for Employment Promotion, International Labour Organi- sation, Geneva. June. Processed. . 1983. "Effect of Agricultural Production and Prices on the Incidence of Rural Poverty: A Tentative Analysis of Interstate Variations." Economic and Political Weekly, Review of Agriculture (June): A48-A53. Muqtada, M. 1974. "The Seed-Fertilizer Technology and Surplus Labor in Bangladesh Agriculture." Bangladesh Development Studies 3 (4): 403-28. Murty, C. 1987. "Influence of Socio-Economic Status on Contractual Terms of Tenancy: A Study in Two Delta Villages of Andhra Pradesh." Economic and Political Weekly 22 (39) September 26: A 111-A1 21. Muthiah, C. 1971. "The Green Revolution: Participation by Small vs. Large Farmers." Indian Journal of Agricultural Economics 26 (1) January-March. Myers, Desaix. 1973. "Comilla: Reassessment and Replication-The Cooper- ative under Stress." Small Farmer Credit in South Asia. USAID Spring Review of Small Farmer Credit. Washington, D.C.: U.S. Agency for International Development. Myrdal, Gunnar. 1968. The Asian Drama: An Inquiry into the Poverty of Nations. New York: Pantheon. Naidu, V. T. 1964. "Risk and Uncertainty in Agriculture in Relation to Credit." Indian Journal of Agricultural Economics 19 (January-March). Nair, K. 1979. In Defense of the Irrational Peasant. Chicago, Ill.: University of Chicago Press. Nakajima, C. 1970. "Subsistence and Commercial Family Farms: Some Theo- retical Models of Subjective Equilibrium." In C. R. Wharton, Jr., ed., Sub- sistence Agriculture and Economic Development. Chicago, Ill.: Aldine. Narain, Dharm, and P. Joshi. 1969. "Magnitude of Agricultural Tenancy." Economic and Political Weekly, Review of Agriculture, September 27. Narain, Dharm, and Shyamal Roy. 1980. Impact of Irrigation and Labor Availa- bility on Multiple Cropping: A Case Study of India. IFPRI Research Report 20. Washington, D.C.: International Food Policy Research Institute. 416 Bibliography Naseem, S. M. 1971. "Small Farmers and Agricultural Transformation in Pa- kistan Punjab." Ph.D. diss., University of California, Davis. - 1973. "Mass Poverty in Pakistan: Some Preliminary Findings." Paki- stan Development Review 12 (4) Winter. - 1979. "A National Profile of Poverty in Pakistan." Joint ESCAP-Quaid e Azam University project, Bangkok. Draft. Nayyar, R. 1976. "Wages for Agricultural Labourers in Uttar Pradesh." Eco- nomic and Political Weekly 6 (11). NCAER (National Council of Applied Economic Research). 1974. Credit Re- quirements for Agriculture. New Delhi. 1975. Changes in Rural Income in India, 1968-71. New Delhi. _ 1977. Resource Use, Productivity and Land Reforms in Uttar Pradesh. New Delhi. 1978. Fertilizer Demand Study: Interim Report, 1975-76. New Delhi. 1979. Survey Data on Patterns of Fertilizer Use on Selected Crops: 1975-76 and 1976-77. Fertilizer Demand Study, Final Report, vol. 10. New Delhi. NDDB (National Dairy Development Board, India). 1977. Operation Flood II: A Proposal by the National Dairy Development Board. Anand, June. Netherlands, Ministry of Foreign Affairs. 1978. Bangladesh: Rural Development in Four Tharas in Kushtia District: Study on the Problems of the Rural Poor. The Hague: International Technical Assistance Department. Newbery, David. 1974. "Cropsharing Tenancy in Agriculture: A Comment." American Economic Review 64: 1060-66. - 1975a. "The Choice of Rental Contract in Peasant Agriculture." In Lloyd G. Reynolds, ed., Agriculture in Development Theory. New Haven, Conn.: Yale University Press. - 1975b. "Tenurial Obstacles to Innovation." Journal of Development Studies 11 (3) July: 263-77. - 1977. "Risk Sharing, Sharecropping and Uncertain Labor Markets." Review of Economic Studies 44: 585-94. Noronha, Raymond. 1985. "A Review of the Literature on Land Tenure Sys- tems in Sub-Saharan Africa." Agriculture and Rural Development Depart- ment, World Bank, Washington, D.C. Nulty, L. 1972. The Green Revolution in West Pakistan: Implications of Technologi- cal Change. New York: Praeger. Nyholm, Klaus, Henrik Schaumburg-Muller, and Kirsten Westergaard. 1974. "Socio-Economic Aspects of Dairy Development: Report from the Banga- lore Milkshed Area." Economic and Political Weekly, Review of Agriculture (December 17): A127-A136. Oberoi, A., and I. Ahmed. 1981. "Labour Use in Dynamic Agriculture: Evi- dence from Punjab." Economic and Political Weekly 16 (13) March 28. Oberoi, A., and S. Manmohan. 1980. "Migration, Remittances and Rural De- velopment: Findings of a Case Study in the Indian Punjab." International Labor Review, March. Bibliography 417 Ojha, Gyaneshwar. 1970. "Small Farmers and HYV Programmes." Economic and Political Weekly 5 (14) April 4. Oomen, M. 1971a. "Green Revolution and Agrarian Conflict." Economic and Political Weekly, Review of Agriculture 6 (26) June 26. . 1971b. Land Reforms and Socioeconomic Change in Kerala. Madras: Christian Literature Society. . 1975. "Agrarian Legislations and Movements as Sources of Change: The Case of Kerala." Economic and Political Weekly 10 (40) October. Pacey, C., R. Longhurst, and R. Chambers. 1981. Why Seasons Matter: Seasonal Dimensions of Rural Poverty. London: Croom Helm. Padki, M. 1965. "Consolidation of Holdings." Agricultural Situation in India 19 (5) August. Paglin, M. 1965. "Surplus Agricultural Labor and Development: Facts and Theories." American Economic Review 55 (September). Paine, Suzanne. 1977. "Agricultural Development in Less Developed Coun- tries (Particularly South Asia): An Introduction to Bhaduri." Cambridge Journal of Economics 1 (4) December: 335-39. Pakistan, Government of. 1951. Census of Pakistan. Islamabad: Government Press. Pakistan, Ministry of Agriculture and Works. 1960a. Agricultural Census of Pa- kistan: A Summary of East Pakistan Data. Lahore: P. C. and M. G. Press. . 1960b. Agricultural Census of Pakistan: A Summary of West Pakistan Data. Karachi. . 1968. Farm Management Research in Pakistan. Planning Unit. Islama- bad: Government Press. Pakistan, Ministry of Food and Agriculture. 1975. Census of Agriculture, 1972: All Pakistan Report. Lahore: Agricultural Census Organization. Pakistan, Ministry of Food, Agriculture, and Rural Development. 1978. Year- book of Agricultural Statistics. Islamabad. Pakistan, Water and Power Development Authority (WAPDA). 1977. Extended Agroeconomic Survey of the Indus Basin. Lahore. Palmer, Ingrid. 1972. Food and the New Agricultural Technology. Geneva: United Nations Research Institute for Social Development. . 1975. The New Rice in the Philippines. Geneva: United Nations Research Institute for Social Development. . 1976. The New Rice in Asia: Conclusions from Four Country Studies. Ge- neva: United Nations Research Institute for Social Development. Pandey, R. K., and T. S. Bhogal. 1980. "Prospects of Increasing Income and Employment on Mixed Farms." Indian Journal of Agricultural Economics 35 (4) October-December: 144-51. Pandey, R. K., H. P. Singh, and B. C. Saxena. 1981. "Economic Impact of Dairy Development Programmes in the Milk Collection Areas of Madhararam Milk Supply Scheme, Chinglepur (Tamil Nadu)." Indian Jour- nal of Agricultural Economics 35 (4) October-December: 172. 418 Bibliography Pandey, S. M., ed. 1976. Rural Labor in India: Problems and Policy Perspectives. New Delhi: Sri Rami Center for Industrial Relations and Human Re- sources. Pandit, M. L. 1978. "Some Less Known Factors Behind Recent Industrial Changes in Punjab and Haryana." Economic and Political Weekly November 25: 1935-40. Pant, Chandrasekhar. 1980. "Tenancy and Cropping Patterns in India." Ph.D. diss., Indian Statistical Institute, New Delhi. - 1981. Tenancy in Semi-Arid Tropical Villages of South India: Determinants and Effects on Cropping Patterns and Input Use. Economic Program Report 29. Andhra: International Crops Research Institute for the Semi-Arid Tropics. Pant, S. P., and S. V. Karanjkar. 1965. "Economics of Dairy Enterprises in Jabalpur with Special Reference to the Scale of Enterprise." IndianJournal of Agricultural Economics 20 (1). Papanek, Gustav. 1977. Economic Growth, Income Distribution and the Political Process in Less Developed Countries. Discussion Paper 9. Boston, Mass.: De- partment of Economics, Boston University. - 1979a. Methodological and Statistical Appendix. Discussion Paper 30. Boston, Mass.: Department of Economics, Boston University. 1979b. Real Wages, Growth, Inflation, Income Distribution and Policies in Pakistan, India, Bangladesh, Indonesia. Discussion Paper 29. Boston, Mass.: Department of Economics, Boston University. Pardeshi, J. M., and D. V. Kazar. 1981. "An Economic Evaluation of Land Allotment to the Landless Poor in Ahmednagar District of Western Maharashtra: A Case Study." Indian Journal of Agricultural Economics 36 (1) January-March: 109-18. Pardhy, M. K. 1968. "Rural Projects: Assessment of Achievements and Fail- ures." Yojana, February 18. Parthasarathy, Gogula. 1973. "Changes in Rice Farming and Their Economic and Social Impact: Case Study of a Delta Village." Department of Coopera- tion and Applied Economics, Andhra University, Waltair, Andhra Pradesh. - 1977. "Land Reform, Rural Dynamics, and Rural Poverty: Review of Indian Experience." Paper presented at the International Seminar on Agrarian Reform, Institutional Innovation, and Rural Development: Major Issues in Perspective, Land Tenure Center, University of Wisconsin, Madison. July. - 1979. "Land Reform and the Changing Agrarian Structure." In C. H. Shah and C. N. Vakil, eds., Agricultural Development of India. New Delhi: Orient Longmans. - 1985. "Reorientation of Rural Development Programmes: A Note on Some Basic Issues." Economic and Political Weekly 20 (48) November 30: 2125-29. - 1987. "Changes in the Incidence of Rural Poverty and Recent Trends in Some Aspects of the Agrarian Economy." Indian Journal of Agricultural Economics 42 (1) January-March: 1-22. Bibliography 419 Parthasarathy, Gogula, and D. H. Babu. 1970. "HYVP, Variables That Matter: A Comment." Economic and Political Weekly 5 (25) June 20: 980. Parthasarathy, Gogula, and D. S. Prasad. 1974. "Responses to and Impact of HYV Rice according to Land Size and Tenure in a Delta Village, Andhra Pradesh, India." Developing Economies 12 (2) June. Parthasarathy, Gogula, and K. S. Raju. 1972. "Is There an Alternative to Rad- ical Land Ceilings?" Economic and Political Weekly 7 (27) July 1. Patel, R. K. 1980. "Rapporteur's Report on Economics of Livestock Enter- prises with Special Reference to Its Employment Potential." IndianJournal of Agricultural Economics 35 (4) October-December: 174-90. Patel, S. M., and M. K. Pandey. 1976. Economic Impacts of Kaira District Coopera- tive Milk Producers' Union (Anand Dairy) in Rural Areas of Kaira District (Guja- rat State). Ahmedabad: Institute of Cooperative Management. Patel, V. 1987. "Social Forestry: A Farmer's Viewpoint." Economic and Political Weekly 22 (32) August 8: 1365-66. Patnaik, U. 1987. Peasant Class Differentiation. Oxford: Oxford University Press. Perkins, Dwight H., and Shahid Yusuf. 1984. Rural Development in China. Bal- timore, Md.: Johns Hopkins University Press. Perrin, R., and D. Winkelman. 1976. "Impediment to Technical Progress on Small vs. Large Farms." AmericanJournal of Agricultural Economics 58 (5) De- cember. Phukan, Umananda. 1972. A Study of Double Cropping in Sibsagar District, Assam (1968-69) India. Jorhat: Agro-Economic Research Center for Northeast India, Assam Agricultural University. Pinckney, Thomas C. 1986. "Stabilizing Pakistan's Supply of Wheat: Issues in the Optimization of Storage and Trade Policies." Pakistan Development Review 25 (Winter): 451-67. Pingali, Prabhu, Yves Bigot, and Hans P. Binswanger. 1987. Agricultural Mechanization and the Evolution of Farming Systems in Sub-Saharan Africa. Bal- timore, Md.: JQhns Hopkins University Press. Planning, Research and Action Institute. 1961. "Impact of Consolidation of Holdings: An Evaluation Study." Lucknow. Pollnac, Richard B. 1981. Sociocultural Aspects of Developing Small-Scale Fisher- ies: Delivering Services to the Poor. World Bank Staff Working Paper 490. Washington, D.C. Prasad, P. 1976. "Poverty and Bondage." Economic and Political Weekly 11 (31-33) August. . 1979. "Caste and Class in Bihar." Economic and Political Weekly 14 (7-8) February: 481-83. . 1987. "Toward a Theory of Transformation of Semi-Feudal Agricul- ture." Economic and Political Weekly 22 (31) August 1: 1287-91. Pray, Carl E., and Jock R. Anderson. 1986. Bangladesh and the CGIAR Centers: A Study of Their Collaboration in Agricultural Research. CGIAR Study Paper 8. Washington, D.C.: World Bank. 420 Bibliography Punjab, Economic and Statistical Office. 1971. The Evaluation of the Rural In- dustrialization Programme in the Punjab. Report 109. Chandigarh. Qayyum, A. 1977. "Land Reforms (Pakistan): Review of Implementation." Land Tenure Center, University of Wisconsin, Madison. July. Quasem, M. A. 1978. "Factors Affecting the Use of Fertilizers in Bangladesh." Bangladesh Development Studies 6 (3) Monsoon. Quasem, M. A., and Mahabub Hossain. 1979. "Fertilizer Use in Two Selected Areas of Bangladesh." Bangladesh Development Studies 7 (4) Autumn. Quibria, M. 1976. "A Note on Farm Size, Efficiency and the Socioeconomics of Land Distribution." Bangladesh Development Studies 4 (January). Quiz6n, Jaime, and Hans P. Binswanger. 1986. "Modeling the Impact of Ag- ricultural Growth and Government Policy on Income Distribution in India." World Bank Economic Review 1 (1) September: 103-48. Rabbani, G. 1966. "Surplus Labour in Padi Cultivation in East Pakistan in 1964-65." Bureau of Statistics, Government of East Pakistan. Radhakrishna, R., G. V. S. N. Murthy, and N. C. Shah. 1979. Models of Con- sumer Behaviour for the Indian Economy. Ahmedabad: Sardar Patel Institute of Economic and Social Research. Radhakrishna, R., and K. N. Murty. 1978. Food Demand Modelfor India. Ah- medabad: Sardar Patel Institute of Economic and Social Research. Radhakrishna, R., and A. Sharma. 1979. "Analysis of Sectoral Price Move- ments in a Developing Economy: Effects of Movement in Agricultural Prices and Production on Industrial Prices, Demand Pattern and Income Distribution." Paper presented at the Seventh International Conference on Input/Output Techniques, Innsbruck, April. Radhakrishnan, S. A., and M. Sivanandham. 1975. "Dairying as a Subsidiary Enterprise for Farmers: A Micro Level Analysis." IndianJournal of Agricul- tural Economics 30 (3): 159. Rahman, M. A. 1975. "Farm Size, Efficiency and the Socioeconomics of Land Distribution." Bangladesh Development Studies 3 (3). Rahman, R. 1. 1978. "Measurement of Rural Unemployment: A Dis- aggregative Approach." Bangladesh Development Studies, Winter. - 1981. "New Technology in Bangladesh Agriculture: Adoption and Its Impact on Rural Labor Markets." Asian Employment Programme Working Paper. International Labour Organisation, Geneva. Raj, K. N. 1970. "Ownership and Distribution of Land." Indian Economic Re- view 1 (5 n.s.) April: 1-42. - 1976. "Trends in Rural Unemployment in India: An Analysis with Reference to Conceptual and Measurement Problems." Economic and Politi- cal Weekly 11 (31-33) August. Rajaraman, I. 1975. "Poverty, Inequality, and Economic Growth: Rural Pun- jab, 1960-61 to 1970-71." Journal of Development Studies, July. Rajasthan, Government of. 1977. Crop Estimation Study under Agricultural Ex- tension Program in Chambal Command Area. Vols. I and 2. Kota. Bibliography 421 Rajat, D., and R. B. L. Bhardwaj. 1974. Systems of Multiple Cropping to Maximize Economic Returns and Employment. Proceedings of the First FAO/SIDA Seminar on Improvement and Production of Field Crops for Plant Scientists from Africa and the Near East, Cairo, September 1-20, 1973. Rome: Food and Agriculture Organization. Rajpurohit, A. R. 1972. "Study of HYV Programme in Mysore (1967-68): Mexican Wheat in Bijapur District." Gokhale Institute Series 17. Poona. Processed. . 1975. "Bovine Feed Availability and Requirements in Karnataka with Reference to Dairy Development Programmes." Indian Journal of Agricul- tural Economics 30 (3) July-September: 111-20. Ranade, C., and R. Herdt. 1978. Shares of Farm Incomes from Rice Production. Los Bafios: International Rice Research Institute. Ranadive, B. T. 1979. "Caste, Class and Property Relations." Economic and Po- litical Weekly 14 (7-8) February: 337-48. Randhawa, M. S. 1974. Green Revolution. Delhi: Vikas. Rani, Usha. 1971. "Size of Farm and Productivity." Economic and Political Weekly 6 (26) June 26: A85-A89. Rao, A. P. 1967. "Size of Holdings and Productivity." Economic and Political Weekly, November 11. Rao, B. P. 1973. The Economics of Agricultural Credit Use in Southern Brazil. An- dhra: Andhra Pradesh University Press. Rao, C. H. H. 1966. "Alternative Explanations of the Inverse Relationship between Farm Size and Output per Acre in India." Indian Economic Review 1 (2) October. . 1970. "Farm Size and Credit Policy." Economic and Political Weekly 5 (52) December 26. . 1971. "Uncertainty, Entrepreneurship and Sharecropping in India." Journal of Political Economy 79: 580. . 1972. "Ceiling on Agricultural Land Holdings: Its Economic Ra- tionale." Economic and Political Weekly, Review of Agriculture 7 (26) June 24. . 1975. Technological Change and the Distribution of Gains in Indian Agricul- ture. Delhi: Macmillan. . 1976. "Rapporteur's Report on Changes in the Structural Distribu- tion of Land Ownership and Use (since Independence)." Indian Journal of Agricultural Economics 31 (3) July-September: 56-62. . 1977a. "Agricultural Growth and Rural Poverty: Some Lessons from Past Experience." Economic and Political Weekly 12 (33-34). . 1977b. Caste and Poverty: A Case Study of Scheduled Castes in a Delta Vil- lage. Malikipuram: Savithri Publications. . 1986. "Changes in Rural Poverty in India: Implications for Agricul- tural Growth." Mainstream January 11. Rao, K., and A. Sirohi. 1986. "Risk of Fertilizer Application to Paddy in West 422 Bibliography Godavari District: Use of Contingency Matrix Approach." IndianJournal of Agricultural Economics 41 (1) January-March: 42-51. Rao, M. 1981. "Commercialisation and Agricultural Growth in Asia: A Com- parative Study of South Korea and Coastal Andhra." Paper presented at the International Workshop on Rural Transformation in Asia, New Delhi, October 2-4. Rao, P V. G. K. 1969. Economic Aspects of High-Yielding Wheat and Punjab (Spe- cial Reference to Amritsar District Rabi 1968-69. Delhi: Agro-Economic Re- search Center. Rao, V. M., and R. S. Deshpande. 1986. "Agricultural Growth in India: A Review of Experiences and Prospects." Economic and Political Weekly 21 (38-39) September 20-27: AIOI-Al 13. Rao, V. V. Bhanoji. 1982. "Data on Income Distribution in India." Working Paper 1980-2. EPD Income Distribution Project, World Bank, Washington, D.C. June. - 1987. "Changing Village Structure: Impact of Rural Development Programmes." Economic and Political Weekly 22 (13) March 28: A2. Rao, V. V. Bhanoji, and T Chotigeat. 1981. "The Inverse Relationship be- tween Size of Land Holdings and Agricultural Productivity." AmericanJour- nal of Agricultural Economics 63 (3) August: 571-74. Rath, N. 1988. "The 1988-89 Budget: Does It Tackle Any of India's Prob- lems?" Economic and Political Weekly 23 (15) April 9: 739-45. Rath, R. K. 1979. "Towards a Functional Land Redistribution Programme." Economic and Political Weekly 14 (2) June 13. Raup, P. 1963. "The Contribution of Land Reforms to Agricultural Develop- ment: An Analytical Framework." Economic Development and Cultural Change 12 (1) October. Rawski, Thomas G. 1979. Economic Growth and Employment in China. New York: Oxford University Press. Ray, A. K., B. R. Atteri, A. C. Sen, and P. N. Mathur. 1979. "Quantitative and Qualitative Impact of Training and Visit Systems on Different Groups of Farmers: A Case Study of Hooghly District, W. Bengal." IndianJournal of Agricultural Economics 34 (4) October-December. Ray, Amal. 1979. "Organizational Problems of Small Farmers Development Administration." Economic and Political Weekly December: A161-A165. Reddy, Y., and others. 1987. "Dryland Farming: Constraints to Improved Technologies." Margin 19 (3) April-June. Reid, J. 1974. "Sharecropping as an Understandable Response: The Post- Bellum South." Journal of Economic History 33: 106-30. Reserve Bank of India. 1975. Small Farmers Development Agencies: A Field Study, 1972-73. Bombay. - 1976. Assets of Rural Households, as onJune 30, 1971 (All-India Debt and Investment Survey, 1971-72). Bombay. Reynolds, N., and P. Sunder. 1977. "Maharashtra's Employment Guarantee Bibliography 423 Scheme: A Program to Emulate?" Economic and Political Weekly 12 (29) July 16: 1149-58. Rice, E. B., ed. 1973. "Summary of the Spring Review of Small Farmer Credit." Small Farmer Credit Summary Papers. Washington, D.C.: U.S. Agency for International Development. Ridker, R. G., and H. Lubell, eds. 1971. Employment and Underemployment Prob- lems of the Near East and South Asia. New Delhi: Vikas. Riskin, C. 1978. "Intermediate Technology in China's Rural Industries." World Development 6 (11-12): 1297-1311. Robinson, W. C. 1969. "Disguised Unemployment Once Again, 1951-1961." American Journal of Agricultural Economics, August. Robinson, W. C., and N. Abbasi. 1979. "Underemployment in Pakistan." Paki- stan Development Review 18 (4) Winter. Rochin, Refugio I. 1971. "A Micro-Economic Analysis of Small Holder Re- sponse to High-Yielding Varieties of Wheat in West Pakistan." Ph.D. diss., Department of Agricultural Economics, Michigan State University, East Lansing. . . 1972. "Responsiveness of Subsistence Farmers to New Ideas: Dwarf Wheat on Unirrigated Small Holdings in Pakistan." Reprint 88. Land Ten- ure Center, University of Wisconsin, Madison. May. . 1973. "A Study of Bangladesh Farmers' Experiences with IR-20 Rice Variety and Complimentary Production Inputs." Bangladesh Economic Re- view I (1) January. Rosenstein-Rodan, Paul N. 1981. "The New International Economic Order or Relations between the Haves and Have Nots (North-South)." University Lecture 16. Boston University, Boston, Mass. March. Rosensweig, Mark M. 1978. "Rural Wages, Labor Supply and Land Reform: A Theoretical and Empirical Analysis." American Economic Review 68: 874- 961. Roumasset, J. A. 1976. Rice and Risk: Decision Making among Low-Income Farm- ers. Amsterdam: North-Holland. . 1978. "The Case against Crop Insurance in Developing Countries." Philippine Review of Business and Economics, March. . 1979. "Sharecropping, Production Externalities and the Theory of Contracts." American Journal of Agricultural Economics 61 (4) November: 640-47. Roumasset, J. A., J. M. Boussard, and 1. J. Singh, eds. 1979. Risk and Uncer- tainty in Agricultural Development. New York: Agricultural Development Council. Roy, Prannoy. 1981. "Transition in Agriculture: Empirical Indicators and Results (Evidence from Punjab, India)." Journal of Peasant Studies 8 (2) January. Roy, Shyama. 1987. "The Tilona Model: A Successful Indian Grass-Roots De- velopment Strategy?" Canadian Journal of Development Studies 8 (2): 355. 424 Bibliography Roy, Shyama, and Fred Sanderson. 1972. The Green Revolution in India: An Appraisal. Brookings Institution Discussion Paper. Washington, D.C. Rudra, A. 1968. "Farm Size and Yield per Acre." Economic and Political Weekly (Special Number) July. - 1969. "Big Farmers of Punjab: Second Installment of Results." Eco- nomic and Political Weekly, Review of Agriculture (December): A213-A219. - 1975a. "Loans as a Part of Agrarian Relations: Some Results of a Pre- liminary Survey in West Bengal." Economic and Political Weekly 10 (28) July 12. - 1975b. "Sharecropping Arrangements in West Bengal." Economic and Political Weekly 10 (39) September 27. - 1976. "Hiring of Labour by Poor Peasants." Economic and Political Weekly 11 (1-2) January 10: 33-36. - 1978a. "Class Relations in Indian Agriculture." Parts I, II, and III. Economic and Political Weekly 13 (24) June 3: 916-23; 13 (25) June 10: 963-68; and 13 (26) June 17: 998-1004. - 1978b. "Organization of Agriculture for Rural Development: The In- dian Case." CambridgeJournal of Economics 2: 381-406. - 1981. "One Step Forward, Two Steps Backward." Economic and Politi- cal Weekly 16 (25-26) June 20-27: A61-A68. - 1987. "Share of Agriculture in the Labor Force." Economic and Political Weekly 22 (50) December 12: 2157. Rudra A., A. Majid, and B. Talib. 1969. "Big Farmers of Punjab." Economic and Political Weekly 4 (39) September 27. Rudra, A., and A. Sen. 1980. "Farm Size and Labor Use: Analysis and Policy." Economic and Political Weekly 15 (5-7): 391-94. Ruttan, Vernon W 1971. "Tenure and Productivity of Philippine Rice Pro- ducing Farms." Philippine EconomicJournal 79: 587-95. - 1973. Induced Technical and Institutional Change and the Future of Agri- culture. New York: Agricultural Development Council. - 1977. "The Green Revolution: Seven Generalizations." International Development Review 19 (4) December. Rutten, M. 1986. "Social Profile of Agricultural Entrepreneurs: Economic Behaviour and Life-Style of Middle-Large Farmers in Central Gujarat." Economic and Political Weekly 21 (13) March 29: A15. Sagar, S., and others. 1987. "Poverty among Identified Weaker Sections in Rajasthan." Economic and Political Weekly 22 (26) June 27. Sahota, G. S. 1968. "The Efficiency of Resource Allocation in Indian Agricul- ture." American Journal of Agricultural Economics 50 (August): 584-605. Sain, K. 1977. "Changes in Ownership and Operational Holdings in West Bengal: An Analytical Study." Bangladesh Development Studies 5 (2) April: 201-10. Saini, A. S. 1975. "Impact of Dairy Enterprises on Farm Incomes in Punjab." Indian Journal of Agricultural Economics 30 (3) July-September: 159. Bibliography 425 Saini, G. R. 1969. "Resource-Use Efficiency in Agriculture." IndianJournal of Agricultural Economics 24 (April-June): 1-18. . 1971. "Holding Size, Productivity, and Some Related Aspects of India's Agriculture." Economic and Political Weekly 6 (26) June: A79-A85. . 1976. "Green Revolution and the Distribution of Farm Incomes." Eco- nomic and Political Weekly 11 (13) March 27. . 1979. Farm Size, Resource Use Efficiency and Income Distribution. New Delhi: Allied Publishers. . 1980. "Farm Size, Productivity and Some Related Issues in India's Ag- riculture: A Review." Agricultural Situation in India February: 777-83. Saith, A. 1981. "Production, Prices and Poverty in Rural India."Journal of De- velopment Studies 17 (2) January. Saith, A., and A. Tankha. 1972. "Agrarian Transition and the Differentiation of the Peasantry: A Study of a West U.P. Village." Economic and Political Weekly 7 (13) April 1. Salam, Abdul. 1976. "Resource Productivity in the Punjab's Agriculture." Pa- kistan Development Review 15 (2) Summer. Salimullah, S., and Shamsul Islam. 1976. "A Note on the Conditions of the Rural Poor in Bangladesh." Bangladesh Development Studies 4 (2) April. Sanghvi, P. 1969. Surplus Manpower in Agriculture and Economic Development. Bombay: Asia Publishing House. Sankhayan, P. L., and others. 1978. "An Analysis of Income and Expenditure Patterns of Punjab Farmers." Department of Economics and Sociology, Punjab Agricultural University, Ludhiana. Sanyal, S. K. 1969. "Size of Holding and Some Factors Related to Produc- tion." Economic and Political Weekly August 16: 1345-47. . 1972. "Has There Been a Decline in Agriculture Tenancy?" Economic and Political Weekly 7 (19) May 6. . 1976. "A Review of the Conceptual Framework of Land Holding Sur- veys." Indian Journal of Agricultural Economics 31 (3) July-September: 2-10. 1977a. "Long Haul to Agrarian Reform." Economic and Political Weekly 8 (5) December. . 1977b. "Trends in Some Characteristics of Land Holdings: An Analy- sis for a Few States." Parts I and II. Sarvekshna 1 (1) July: 1-13 and 1 (2) October: 65-76. 1979. "Changes in the Structural Distribution of Land Ownership and Use: A Few Comments." Indian Journal of Agricultural Economics 34 (1) January-March: 101-07. Sanyal, S., and S. Sinha. 1976. "Methodological Problems in Large-Scale Sam- ple Surveys: Experiences from the National Sample Survey." Paper pre- sented at the Symposium on Recent Developments in Survey Methodology, Indian Statistical Institute, Calcutta, March 22-27. Sau, R. 1971. "Agricultural Revolution by Production Function." Economic and Political Weekly 6 (26) June 26. 426 Bibliography - 1973. "Political Economy of Indian Agriculture: What Is It All About?" Economic and Political Weekly 7 (20). . 1988. "The Green Revolution and Industrial Growth in India." Eco- nomic and Political Weekly 23 (16) April: 789-96. Sawant, S. 1983. "Investigation of the Hypothesis of Deceleration in Indian Agriculture." Indian Journal of Agricultural Economics 38 (4) October-De- cember. Saxena, A. 1987. "Concurrent Evaluation of IRDP: Selected Aspects for Ad- ministrative Follow-up." Economic and Political Weekly 22 (39) September 26: A121-A124. Scandizzo, Pasquale L. 1979. "Implications of Sharecropping for Technology Design in Northeast Brazil." In A. Valdes, G. Scobie, and J. Dillon, eds., Economics and the Design of Small-Farmer Technology. Ames: Iowa State Uni- versity Press. Schaefer, R. 1978. "What Are We Talking about When We Talk about Risk? A Critical Survey of Risk and Risk Preference Theories." Research Memo- randum 78-69. International Institute for Applied Systems Analysis, Laxenberg, Austria. December. Schluter, M. 1971. Differential Rates of Adoption of the New Seed Varieties in India: The Problem of Small Farms. Occasional Paper 47. USAID Employment and Income Distribution Project. Ithaca, N.Y: Department of Agricultural Economics, Cornell University. . 1973. The Role of Cooperative Credit in Small Farmer Adoption of the New Cereal Varieties in India. Occasional Paper 64. USAID Employment and In- come Distribution Project. Ithaca, N.Y: Department of Agricultural Eco- nomics, Cornell University. - 1974. The Interaction of Credit and Uncertainty in Determining Resource Allocation and Income on Small Farms: Surat District, India. Occasional Paper 68. USAID Employment and Income Distribution Project. Ithaca, N.Y: De- partment of Agricultural Economics, Cornell University. Schluter, M., and R. Longhurst. 1972. Some Aspects of the Suitability of High- Yielding Rice and Bajra Varieties for the Small Farm: Thanjavur and Mehsana Dis- tricts, India. Occasional Paper 57. Ithaca, N.Y: Department of Agricultural Economics, Cornell University. Schluter, M., and J. Mellor. 1972. "New Seed Varieties and the Small Farm." Economic and Political Weekly, Review of Agriculture 7 (13) March 25. Schultz, Theodore W. 1964. Transforming Traditional Agriculture. New Haven, Conn.: Yale University Press. - 1977. "Farm Entrepreneurs, Incentives, and Economic Policy." Agri- cultural Economic Research Paper 77-20. University of Chicago, August. Draft. - 1979. "The Economics of Being Poor." Nobel Lecture, Stockholm, December 8. Schumacher, E. E 1973. Small Is Beautiful. New York: Harper & Row. Schutjer, W., and M. Van der Veen. 1977. Economic Constraints on Agricultural Bibliography 427 Technology Adoption. Occasional Paper 5. Washington, D.C.: Economic and Sector Planning Division, U.S. Agency for International Development. Selden, M. 1971. The Yenan Way in Revolutionary China. Cambridge, Mass.: Harvard University Press. Sen, Amartya K. 1962. "An Aspect of Indian Agriculture." Economic and Polit- ical Weekly, Annual Number, February. . 1964. "Size of Holdings and Productivity." Economic and Political Weekly, Annual Number, February. . 1966. "Peasants and Dualism with or without Surplus Labor."Journal of Political Economy 74: 425-50. . 1973. "Poverty, Inequality, Unemployment: Some Conceptual Issues in Measurement." Economic and Political Weekly 8: 31-33. . 1975. Employment, Technology and Development. London: Oxford Uni- versity Press. . 1980. Levels of Poverty: Policy and Change. World Bank Staff Working Paper 401. Washington, D.C. . 1981. Poverty and Famines: An Essay on Entitlement and Deprivation. Ox- ford: Clarendon Press. Sen, Bandhudas. 1970. "Opportunities in the Green Revolution." Economic and Political Weekly 5 (13) March 28. . 1974. Green Revolution in India: A Perspective. New York: John Wiley. Sen, S. 1974. A Richer Harvest. Bombay: Tata McGraw-Hill. Sengupta, S. 1981. "West Bengal Land Reforms and the Agrarian Scene." Economic and Political Weekly, Review of Agriculture 16 (25-26) June 20-27: 169-75. Shah, C. H. 1972. "Impact of Tenancy Reform on Level of Technology (Guja- rat and Maharashtra)." Artha Vikas, July. Shah, S. L., and R. Singh. 1970. "The Impact of New Agricultural Technol- ogy on Rural Employment in North-West U.P." IndianJournal of Agricultural Economics, July-September. Shah, T., A. K. Tripathi, and M. Desai. 1980. "Impact of Increased Dairy Pro- ductivity on Farmers' Use of Feedstuffs." Economic and Political Weekly 15 (33) August: 1407-12. Shanmugasundram, V. 1973. "Economic and Social Implications of High- Yielding (Paddy) Varieties Programme." Department of Economics, Uni- versity of Madras. Sharma, A. C. 1973. "Influence of Certain Economic and Technological Fac- tors on the Distribution of Cropped Area under Various Crops in the Lu- dhiana District."Journal of Research of the Punjab Agricultural University, June. Sharma, R. K. 1974. "Green Revolution and Farm Employment: An Analysis of the Experience in Punjab." IndianJournal of Industrial Relations, January. Shavell, S. 1979. "Risk Sharing and Incentives in the Principal and Agent Re- lationship." BellJournal of Economics Spring: 55-73. Shigemochi, H. 1978. The Structure of Disparity in Developing Agriculture: A Case 428 Bibliography Study of the Pakistan Punjab. Occasional Papers Series 16. Tokyo: Institute of Development Economics. Shivakumar, S. S. 1978, "Aspects of Agrarian Economy in Tamil Nadu: A Study of Two Villages. Part I, Class Structure." Economic and Political Weekly 13 (18) May 6: 762-70. Shukla, V. P., and S. V. Karanjkar. 1975. "Economic Analysis of Government Milk Supply Scheme in Jabalpur." Indian Journal of Agricultural Economics 30 (July-September): 144. Shyam, R., and D. P. Gupta. 1975. "Study of Marginal Farmers and Agricul- tural Labourers in Ballia (1973-74)." Agro-Economic Research Center, Motilal Institute of Research, University of Allahabad. Sidhu, D. S. 1975. "Rapporteur's Report on Dairy Development and the Bo- vine Economy." IndianJournal of Agricultural Economics 30 (3) July-Septem- ber: 161-71. - 1977. Wheat Production, Marketable Surplus and Price Policy in India. CIMMYT Monograph 13. Mexico City: Centro Internacional de Mejor- amiento de Maiz y Trigo. Sidhu, D. S., and S. S. Grewal. 1981. Agricultural Growth and Employment Shifts in Punjab: Findings of a Field Survey. New Delhi: Birla Institute of Scientific Research. . 1987. "The Changing Land Holdings Structure in Punjab." Indian Journal of Agricultural Economics 42 (3) July-September: 294-301. Sidhu, D. S., and others. 1979. "An Analysis of Income and Expenditure Pat- terns of Punjab Farmers (Report for the Year 1977-78)." Department of Economics and Sociology, Punjab Agricultural University, Ludhiana. Sidhu, S. S. 1974a. "Economics of Technical Change in Wheat Production in the Indian Punjab." American Journal of Agricultural Economics 56 (May). - 1974b. "Relative Efficiency in Wheat Production in the Indian Pun- jab." American Economic Review 64 (September): 742-51. Sidhu, S. S., and C. A. Baavante. 1979. "Farm Level Fertilizer Demand for Mexican Wheat Varieties in the Indian Punjab." American Journal of Agricul- tural Economics 61 (3) August: 455-62. - . 1980. The Environmental Factors and Farm Level Input Demand and Wheat Supply in the Indian Punjab: An Application of the Translog Profit Function. Mus- cle Shoals, Ala.: International Fertilizer Development Center. Sigurdson, Jon. 1972. "Rural Industry: A Traveller's View." China Quarterly, April-June. . 1977. Rural Industrialization in China. Cambridge, Mass.: Harvard University Press. - 1978. "Rural Industrialization: A Comparison of Development Plan- ning in China and India." World Development 6: 667-80. Singh, A. J., S. S. Miglani, and Tirath Singh. 1979. "Changes in the Pattern of Cost, Productivity and Farm Income Distribution in Punjab: A Case Study of Ferozepur District." Economic Affairs October-December: 287-92. Bibliography 429 Singh, Amarjit, and A. S. Kahlon. 1976. "The Impact of New Technology on Farm Size, Tenurial Relations, and Land Rents in the Punjab." Indian Journal of Agricultural Economics 31 (3) July-September. Singh, Amrik, and R. V. Singh. 1977. "Impact of Dairy Enterprises on Pro- ductivity and Employment." Agricultural Situation in India 32 (3) June: 139-42. Singh, Daulat, V. K. Singh, and R. K. Singh. 1981. "Changing Patterns of Labor Absorption on Agricultural Farms in Eastern U.P.: A Case Study." Indian Journal of Agricultural Economics 36 (4) October-December: 39-44. Singh, Gopal. 1984. "Socio-Economic Bases of the Punjab Crisis." Economic and Political Weekly 19 (1) January 7. Singh, Harpal. 1976. "Structural Changes in the Size Distribution of Hold- ings: A Macro View." Indian Journal of Agricultural Economics 31 (3) July- September. Singh, Inderjit. 1982. Small Farmers and the Landless in South Asia. World Bank Staff Working Paper 320. Washington, D.C. . 1988a. Land and Labor in South Asia. World Bank Discussion Paper 33. Washington, D.C. . 1988b. Small Farmers in South Asia: Their Characteristics, Productivity, and Efficiency. World Bank Discussion Paper 31. Washington, D.C. . 1988c. Tenancy in South Asia. World Bank Discussion Paper 32. Wash- ington, D.C. Singh, Inderjit, and R. Day. 1975a. "Factor Utilization and Substitution in Economic Development: A Green Revolution Case Study." Journal of Devel- opment Studies 2 (3) April. . 1975b. "A Microeconomic Chronicle of the Green Revolution." Eco- nomic Development and Cultural Change 23 (4) Spring: 661-86. Singh, Inderjit, R. H. Day, and S. S. Johl. 1968. Field Crop Technology in the Punjab, India. Madison: Social Systems Research Institute, University of Wisconsin. Singh, Inderjit, Lyn Squire, and John Strauss. 1986. "A Survey of Agricul- tural Household Models: Recent Findings and Policy Implications." World Bank Economic Review 1 (1): 149-79. Singh, J. 1986. The Role of Institutional Finance in Agriculture. New Delhi: Ashish. Singh, Kartar. 1973. "The Impact of New Technology on Farm Income Dis- tribution in the Aligarh District of U.P." Indian Journal of Agricultural Eco- nomics, April-June. . 1981. "Rapporteur's Report on Labor Absorption in Agricultural and Rural Development." Indian Journal of Agricultural Economics 36 (4) October- December: 57-65. Singh, Prem, and S. D. Choubey. 1969. "For Jabalpur an Intensive Cropping Schedule." Indian Farming, November. Singh, R. P., and H. G. Goswani. 1975. "Study of Small Farmers Development 430 Bibliography Agency, Purnea, North Bihar, 1973-74." Ad Hoc Study A-19, Publication 57. Agro-Economic Research Center, University of Allahabad. Singh, Roshan. 1973. The Social and Economic Implications of the Large-Scale In- troduction of High-Yielding Varieties in Foodgrains (Wheat) in Muzaffarnagar (U. P.). Agra: Raja Balwant Singh College. Singh, Srinath. 1977. "Agricultural Development and the Small Farmer." Land Tenure Center, University of Wisconsin, Madison. July. Singh, Surinder. 1979. "Operation Flood II: Some Constraints and Implica- tions." Economic and Political Weekly 14 (41) October 27: 1765-74. Singh, Surinder, and R. K. Sharma. 1981. "Some Implications of Area Choice in the Operation Flood II Programme." Economic and Political Weekly 16 (13) March 28. Sinha, J. N. 1978. "Rural Development Planning: Dimensions and Con- straints." Economic and Political Weekly, Annual Number. Sinha, S. K. 1976. "Land Reforms and the Emerging Agrarian Structure in Bihar." Indian Journal of Agricultural Economics 31 (3) July-September: 23-27. Sirohi, A. S., G. S. Ram, and C. B. Singh. 1976. "Interstate Disparities in the Structural Distribution of Land Holdings in Rural India." IndianJournal of Agricultural Economics 31 (3) July-September: 14-23. Sirohi, A. S., and others. 1980. "Role of Dairy-Poultry Enterprises for In- creasing Income and Employment in Farms in the Union Territory of Delhi." Indian Journal of Agricultural Economics, October-December. Sivanandan, P. 1979. "Caste, Class and Economic Opportunity in Kerala: An Empirical Analysis." Economic and Political Weekly 14 (7) February: 475-79. Smith, Adam. 1976. The Wealth of Nations. Edwin Cannan, ed. Chicago, Ill.: University of Chicago Press. Solaiman, M., and A. Huq. 1973. "Small Farmer Credit in Bangladesh." Small Farmer Credit in South Asia. UJSAID Spring Review of Small Farmer Credit. Washington, D.C.: U.S. Agency for International Development, June. Somjee, A., and G. Somjee. 1978. "Cooperative Dairying and Profiles of So- cial Change in India." Economic Development and Cultural Change 26 (3) April: 577-90. Squire, Lyn. 1979. Labor Force, Employment and Labor Markets in the Course of Economic Development. World Bank Staff Working Paper 336. Washington, D.C. Srinivasan, T. N. 1972. "Farm Size and Productivity: Implications of Choice under Uncertainty." Sankhya: The IndianJournal of Statistics, ser. B, 34 (4): 409-18. - . 1974. "Income Distribution: A Survey of Policy Aspects." Sankhya: The IndianJournal of Statistics, ser. C, 36 (2-4) June-December: 369-96. 1976. The Remembered Village. Delhi: Oxford University Press. 1977. "Poverty: Some Measurement Problems." Conference Proceed- ings of the 41st Session of the International Statistical Institute, New Delhi, December. Bibliography 431 . 1979a. "Agricultural Backwardness under Semi-Feudalism: A Com- ment." Economic Journal, June. . 1979b. "Bonded Labor Contracts and Incentives to Adopt Yield- Raising Innovations in Semi-Feudal Agriculture." Development Research Center, World Bank, Washington, D.C. March. Draft. . 1979c. "Malnutrition: Some Measurement and Policy Issues." Devel- opment Research Center, World Bank, Washington, D.C. Draft. Srinivasan, T. N., and P. K. Bardhan. 1974. Poverty and Income Distribution in India. Calcutta: Statistical Publishing Society. Srinivasan, T N., and A. Bhaduri. 1979. "Agricultural Backwardness under Semi-Feudalism: An Interchange." Economic Journal 35 (4) June: 89. Srivastava, U. K., and E. 0. Heady. 1973. "Technological Change and Rela- tive Factor Shares in Indian Agriculture: An Empirical Analysis." American Journal of Agricultural Economics, August. Staub, W. J., and Melvin Blase. 1974. "Induced Technological Change in De- veloping Agriculture: Implications for Income Distribution and Agricul- tural Development." Journal of Developing Areas 8 (4) July. Steele, David B. 1975. "Rural Industrialization and Employment." Rural and Urban Employment Policies Branch, International Labour Organisation, Geneva. November. Stepanek, J. E 1973. "Comilla Cooperative Production Loans: A Note on the Cost of Capital." Small Farmer Credit in South Asia. USAID Spring Review of Small Farmer Credit. Washington, D.C.: U.S. Agency for International De- velopment. . 1979. Bangladesh: Equitable Growth? New York: Pergamon. Stevens, R. D., H. Alavi, and P J. Bertocci, eds. 1976. Rural Development in Bangladesh and Pakistan. Honolulu: University of Hawaii Press. Stiglitz, J. 1974. "Incentives and Risk Sharing in Sharecropping." Review of Economic Studies 61 (April): 219-56. Stokes, Eric. 1978. The Peasant and the Raj. Cambridge: Cambridge University Press. Stone, B., ed. 1987. Fertilizer Pricing Policy. Washington, D.C.: International Food Policy Research Institute. Streifkerk, H. 1981. "Too Little to Live on-Too Much to Die on: Employ- ment in Small-Scale Industries in Rural South Gujerat." Economic and Politi- cal Weekly 16 (15) April 11. Subbarao, K. 1980. "Institutional Credit Uncertainty and Adoption of HYV Technology: A Comparison of East U.P. with West U.P." Indian Journal of Agricultural Economics 35 (1) January-March. . 1985a. "Regional Variations in the Impact of Anti-Poverty Pro- grammes: A Review of Evidence." Economic and Political Weekly 20 (43) Octo- ber 26. . 1985b. "State Policies and Regional Disparity in Indian Agriculture." Development and Change 16: 523. 432 Bibliography Sukhatme, P V 1965. Feeding India's Growing Millions. London: Asia Publish- ing House. Sundrum, R. 1986. Growth and Income Distribution in India. London: Sage. Swaminathan, M. S. 1972. "Recent Research on Multiple Cropping." Re- search Bulletin. New Delhi: Indian Agricultural Research Institute. Swaminathan, M. S., and S. S. Bains. 1970. "Latest Technology for Multiple Cropping: Principles, Practices, and Problems." Report of the National Seminar on Multiple Cropping, Ministry of Agriculture, New Delhi, May. Swenson, G. 1976. "The Distribution of Benefits from Increased Rice Produc- tion in Thanjavur District, S. India." IndianJournal of Agricultural Economics 31 (January-March). Tendulkar, S. 1983. "Rural Institutional Credit and Rural Development: A Review Article." Indian Economic Review 18 (1) January-June. Terhal, Piet, and Martin Doornbus. 1983. "Operation Flood: Development and Commercialization." Food Policy 8 (August): 235-39. Thakur, D. S. 1975. "Impact of Dairy Development through Milk Coopera- tives: A Case Study of Gujarat." Indian Journal of Agricultural Economics 30 (3) July-September. Thiesenbusen, W. 1974. "What Changing Technology Implies for Agrarian Reform." Land Economics 50 (1) February. Thomas, John W., and R. M. Hook. 1977. Creating Rural Employment: A Man- ualfor Organizing Rural Works Programs. Washington, D.C.: U.S. Agency for International Development. Thorner, D. 1956. The Agrarian Prospect in India. Delhi: Delhi University Press. Timmer, C. Peter. 1971. Some Comments on the Mellor-Lele Labor Supply Theory of Economic Development and Its Applications. Discussion Paper 71-76. Stan- ford, Calif.: Stanford Food Research Institute, Stanford University. Tinnermeir, R., and C. Dowsell. 1973. "Small Farmer Credit." Workshop Re- port. Agricultural Development Council, Research and Training Network, New York. March. Tyagi, D. 1982. "How Valid Are the Estimates of Trends in Rural Poverty?" Economic and Political Weekly 17 (26) June 26: A54-A61. - 1987. "Domestic Terms of Trade and Their Effects on Supply and Demand in the Agricultural Sector." Economic and Political Weekly 22 (13) March 28: A30-A37. Tyers, R. 1977. Small Farmers in Bangladesh: A Study of Their Prospects. Cam- bridge, Mass.: Center for Population Studies, Harvard University. Vaidyanathan, A. 1986a. Agricultural Development and Rural Poverty. Madras: Institute of Development Studies. - 1986b. "Labour Use in Rural India: Spatial and Temporal Varia- tions." Economic and Political Weekly 21 (52) December: A 130. - 1986c. "On the Validity of NSS Consumption Data." Economic and Polit- ical Weekly 21 (3) January 18. Bibliography 433 Verma, R. C., and D. C. Pant. 1978. "Potentialities of Increasing Farm Income and Employment through Dairying." Indian Journal of Agricultural Economics 33 (3) July-September. Visaria, Pravin. 1970a. "Farmers' Preference for Work on Family Farms." In Report of the Committee of Experts on Unemployment. New Delhi: Planning Commission, Government of India. . 1970b. "Unemployment in India in Perspective." Economic and Political Weekly 5 (29-31). . 1972. "Rapporteur's Report on Rural Unemployment." IndianJournal of Agricultural Economics 27 (4) October-December: 179-89. . 1975. "A Survey of Research on Employment in India." In Survey of Research on Agriculture. Delhi: Indian Council of Social Science Research. . 1977a. "Living Standards, Employment and Education in Western India, 1972-73." Development Research Center, World Bank, Washington, D.C. Draft. . 1977b. "Trends in Rural Unemployment in India." Economic and Polit- ical Weekly 12 (5) January 29: 139-45. . 1978. "Size of Holding, Living Standards and Employment in Rural Western India, 1972-73." Development Research Center, World Bank, Washington, D.C. October. Draft. . 1979. "Demographic Factors and the Distribution of Income: Some Issues." Economic and Demographic Change: Issues for the 1980s. Proceedings of the Conference. Liege: International Union for Scientific Study of Popula- tion. Vol. 1. Also in World Bank Reprint Series 129. Washington, D.C. . 1980. Poverty and Unemployment in India: An Analysis of Recent Evidence. World Bank Staff Working Paper 417. Washington, D.C. . 1981. "Poverty and Unemployment in India: An Analysis of Recent Evidence." World Development 9 (23): 277-300. Visaria, Pravin, and L. Visaria. 1973. "Employment Planning for the Weaker Sections in Rural India." Economic and Political Weekly 8 (4-6). Von Blackenburg, K. 1972. "Who Leads Agricultural Modernization? A Study of Some Progressive Farmers in Mysore and Punjab." Economic and Political Weekly 40 (7). Vyas, V. S. 1970. "Tenancy in a Dynamic Setting." Economic and Political Weekly 5 (26) June: A73-A80. . 1971. "Mainsprings of Agricultural Growth in India." Pause Memo- rial Lecture, New Delhi, August 9. . 1975. India's High-Yielding Varieties Programme in Wheat, 196611967- 197111972. Mexico City: Centro Internacional de Mejoramiento de Maiz y Trigo (CIMMYT). . 1976. "Structural Change in Agriculture and the Small Farm Sector." Economic and Political Weekly 11 (1-2) January 10. . 1979. "Some Aspects of Structural Change in Indian Agriculture: Presidential Address." Indian Journal of Agricultural Economics 34 (1) January-March. 434 Bibliography - 1987. "India's Rural Development Strategies: Lessons in Agricultural Growth and Poverty Alleviation." Paper presented at the International As- sociation of Agricultural Economists Symposium, Beijing, October. Vyas, V. S., Anil Bhatt, and S. M. Shah. 1985. Decentralised Planning in India. Bombay: Oxford University Press and IBH Publishing Co. Vyas, V. S., and N. Jagannathan. 1985. "Cooperation in Farm Production: Conditions of Viability." In T. Bergmann and T Ogura, eds., Cooperation in World Agriculture. Tokyo: Food and Agriculture Policy Research Centre. Vyas, V. S., and G. Mathai. 1978. "Farm and Non-Farm Employment in Rural Areas: A Perspective for Planning." Economic and Political Weekly 13 (6-7) February. Vyas, V. S., and H. V. Shivamaggi. 1983. "Agricultural Labour in India." In- dian Institute of Management, Ahmedabad. Processed. Wade, Robert. 1975. Irrigation and Income Distribution: Three Papers. Paper 85. Brighton: Institute of Development Studies, University of Sussex. November. - 1980a. "Water to the Fields: India's Changing Strategy." South Asian Review 8 (4) July-October. - 1980b. "On Substituting Management for Water in Canal Irrigation: A South Indian Case." Economic and Political Weekly, Review of Agriculture, 15 (52) December 27: A147-A160. - 1982. "The System of Administrative and Political Corruption: Canal Irrigation in South India." Journal of Development Studies 18 (April): 287-328. - 1984. "Irrigation Reform in Conditions of Populist Anarchy: An In- dian Case." Journal of Development Economics 14 (April): 285-303. - . 1987a. "Management of Common Property Resources: Collective Ac- tion as an Alternative to Privatisation or State Regulation." Cambridge Jour- nal of Economics 11 (June): 95-106. - 1987b. "Management of Common Property Resources: Finding a Co- operative Solution." World Bank Research Observer 2 (July): 219-34. Wade, Robert, and Robert Chambers. 1980. "Managing the Main System: Canal Irrigation's Blind Spot." Economic and Political Weekly 15 (39) Septem- ber 27. Wadhwa, D. C. 1973. Agrarian Legislation in India, 1793-1966. Vols. 1 and 2. Poona: Gokhale Institute of Politics and Economics. Warriner, Doreen. 1973. "Results of Land Reforms in Asian and Latin Amer- ican Countries." Food Research Institute Studies 12 (2): 115-33. Westley, J. 1986. Agriculture and Equitable Growth: The Case of Punjab and Haryana. Boulder, Colo.: Westview Press. Wharton, C. R., Jr. 1963. "The Economic Meaning of Subsistence." Malayan Economic Review 8: 46-58. Bibliography 435 Whitcombe, E. 1973. "The New Agricultural Strategy in Uttar Pradesh, India, 1968-70: Technical Problems." In R. T. Shand, ed., Technical Change in Asian Agriculture. Canberra: Australian National University Press. Williamson, Cheryl R. 1980. "The Nutritional Impact of Operation Flood: A Framework for Analysis." Institute for International Development, Har- vard University, Cambridge, Mass. May. Draft. Wills, I. R. 1972. "Projection of Effects of Modern Inputs on Agricultural Income and Employment in a C.D. Block, U.P., India." American Journal of Agricultural Economics 54 (3). Winrock International Livestock Research and Training Center. 1983. Sheep and Goats in Developing Countries: Their Present and Potential Role. World Bank Technical Paper 15. Washington, D.C. Wood, D. G. 1973. "From Raiyat to Rich Peasant." South Asian Review 7 (1) October. World Bank. 1975. Agricultural Credit. Sector Policy Paper. Washington, D.C. 1978. Forestry. Sector Policy Paper. Washington, D.C. 1979. Bangladesh: Current Trends and Development Issues. Washington, D.C. .1980. World Development Report 1980. New York: Oxford University Press. 1982. Fishery. Sector Policy Paper. Washington, D.C. 1984. Pakistan: Review of the Sixth Five-Year Plan. World Bank Country Study. Washington, D.C. . 1986. Poverty and Hunger: Issues and Optionsfor Food Security in Develop- ing Countries. World Bank Policy Study. Washington, D.C. . 1987a. Agricultural Mechanization: Issues and Options. World Bank Pol- icy Study. Washington, D.C. . 1987b. Bangladesh: Promoting Higher Growth and Human Development. World Bank Country Study. Washington, D.C. . 1988a. World Development Report 1988. New York: Oxford University Press. . 1988b. World Tables 1987. Washington, D.C. Wunderlich, G. 1970. Land Reforms in India. USAID Spring Review Country Paper. Washington, D.C.: U.S. Agency for International Development. Yap, Lorene Y. L. 1975. Internal Migration in Less Developed Countries: A Survey of the Literature. World Bank Staff Working Paper 215. Washington, D.C. Yelamanchili, Satyanarayana. 1981. "Wage Trends in India, 1830 to 1976: An Analysis, Discussion and Compilation of Data." Studies in Employment and Rural Development 74. Employment and Rural Development Division, World Bank, Washington, D.C. October. Yotopoulos, P. A. 1972. "Economic Rationality and Efficiency through the Production Function, Darkly." Paper presented at the Purdue Conference on Small Farm Agriculture, Purdue University, West Lafayette, Indiana. November. 436 Bibliography Yotopoulos, P. A., and L. J. Lau. 1973. "A Test for Relative Economic Effi- ciency: Some Further Results." American Economic Review 63 (1) March: 214-23. Yotopoulos, P. A., L.J. Lau, and K. Somel. 1970. "Labour Intensity and Rela- tive Efficiency in Indian Agriculture." Food Research Institute Studies in Agri- cultural Economics, Trade, and Development 9 (1): 43-55. Zaman, M. A. 1976. "Land Reform in Bangladesh to 1970." Research Paper 66. Land Tenure Center, University of Wisconsin, Madison. February. Zaman, M. Raquibuz. 1973. "Share-cropping and Economic Efficiency in Bangladesh." Bangladesh Economic Review 1 (2) April: 149-72. Index Abraham, S., 137 Alderman, H., 214 Adams, D. W., 145 Ali, A., 93 Agarwal, B., 189 Amjad, Rashid, 141, 170 Agrarian structure: agrarian change Anand Dairy Cooperative, 4, 212-14, theories and, 48-51; changes in, 1, 215, 216, 225 3; definition of, 47; landholdings Anderson, Dennis, 87 and, 51-62; landless and, 49 Antipoverty programs. See Poverty Agricultural growth: in Bangladesh, (programs to fight) 28-29; employment and, 242-44; Appa Rao, V., 215 income and, 2; in India, 26-28; in Asaduzzaman, M., 163 Pakistan, 28; programs spurring, 2- Attwood, D. W., 27, 49 7; reduction of poverty and, 2, 23, Auden-Laerina, T., 194 26-29. See also Small-farm sector Agricultural output, 8; in Bangladesh, Bahl, D., 189 19, 23-26; employment linkages Bangladesh, 4, 6; agricultural growth and, 234; fertilizers and, 166-76; of and poverty in, 28-29; agricultural food grains, 18-19; Hw use and, 157, output in, 19, 23-26; agricultural 165-76; income and, 19; in India, wage rates in, 29-33; credit in, 141, 19-23; net domestic product per 147; dairying in, 206,214; farm size capita and, 18-19, 22; in Pakistan, in, 134, 157, 162-63; fertilizer use 19-23; of small farms, 101, 108-14; in, 170-76; fisheries in, 228, 229; trends in, 18-26, 41 food prices in, 196; irrigation in, Agricultural wages, 29-35 128, 129, 134-35; labor demand Aquaculture, 228-29 increase in, 181; landholding in, 38, Ahluwalia, Montek, 12, 13, 26, 27, 28, 52-56, 65-69; landlessness in, 73; 185, 199 land redistribution in, 300,301,302, Ahmad, B., 193 307-08,309,314,325,329,330,331; Ahmed, I., 170 livestock farming in, 205, 207, 226; Alagh, Y K, 22 marketing systems in, 150; mech- Alamgir, M., 17 anization in, 193; multiple cropping Albrecht, Herbert, 39 in, 115, 117, 118, 123, 124, 125; 437 438 Index population in, 8; poverty patterns 216; farm and nonfarm linkages in, 17-18; public works programs in, and, 234-35; of food grains, 239-40; 259, 260, 262, 263, 264; tenancy in, industrial development and, 234, 278, 284, 285, 286, 293, 297, 298; 237-41; landless and, 237, 240 training and visit system in, 114; Contracts (tenancy), 285, 291, 292 unemployment in, 82, 85; wage Cooperatives: credit and, 137, 140, labor in, 77, 91-93 141, 147; dairy, 4, 212-14, 215, 216, Banks, 137, 147. See also Credit 221, 225; forestry, 230-31; mar- Bardhan, Pranab K, 28, 33-34, 197, keting and, 150 284 Cows. See Dairying; Livestock farming Barker, K, 167, 181, 194 Crash Scheme for Rural Employment Basant, R., 189, 191 (csRE), 5, 262, 266 Baxter, Michael, 108 Credit: access to, 146-48; antipoverty Bell, Clive, 284 programs and, 5; in Bangladesh, Benami (illegal transfer of holdings), 141, 147; Chinese rural develop- 295 ment and, 253; cooperatives and, Benor, Daniel, 108, 111 137, 140, 141, 147; dairying and, Bhalla, G. S., 22, 198 224, 225; default, 143; HYVs and, Bhalla, Sheila, 50-51 140, 142, 143, 146; in India, 136, Bhalla, Surjit, 140, 142 140, 142, 146, 148; interest rates Binswanger, Hans P., 188 and, 140, 144-45; irrigation and, Birth rate, 39 129-30, 132, 143, 146; in Pakistan, Blair, H., 135, 230 136, 140, 141, 142, 146; poultry Blyn, George, 181, 199 raising and, 228; rural industrial Bose, S. R., 14, 17, 29, 141, 189 development and, 234, 258, 275; Bruce, R.,Jr., 27 small-farm sector and, 135-48; Buffaloes, 205, 206, 207, 209, 215, 217 sources of, 136-42; technology and, Burki, S.J., 262, 264 142-44; uncertainty and, 145-46. See also Subsidies Caste system, 40 Crop insurance program, 146 Cereal technologies, 3 Cropping, 3; dairying and, 217-18, Chadha, G. K, 198 221-24; intensities, 101. See also Chandhry, M. G., 181, 185, 189, 199 Multiple cropping Chandrashekhar, D., 229 Crop yields. SeeAgricultural output Change theories (agrarian), 48-51 Crotty, R., 207, 217 Cheung, S. N. S., 284 China, rural industrialization in, 251- Dairying, 4; Anand Dairy Cooperative 57 and, 212-14, 215, 216, 225; in Clark, E. M., 189 Bangladesh, 206, 214; constraints Class-polarization theory, 48-51, 52 on, 224; credit and, 224, 225; crop Clay, EdwardJ., 181 production and, 217-18, 221-24; Comilla cooperative credit model, 141. employment in, 218-21; grazing See also Credit and, 207; income from, 210, 215, Command area development (CAD), 216-17, 217-18, 221-24, 225; in 131 India, 210,212-14,217; keyfeatures Commodity expenditures, 238 of, 205-12; milk yields and, 207, Consumer goods, 237-38, 238-39 225; nutrition and, 216-17; Consumption, 26, 148; dairy product, Operation Flood and, 212, 226; in Index 439 Pakistan, 206,212,214; poverty and, sector and, 109-10, 114, 155; 214-216, 224-26. See also Livestock subsidies for, 23 farming Fisheries, 4, 228-29 Dandekar, V. M., 35, 307, 328, 329,330, Flood control, 2 331 Food grains: consumption of, 239-40; Dantwala, M. L., 6, 50 output of, 18-19 Dasgupta, Biplab, 181, 193 Food prices. See Prices Day, R. H., 191 Food for Work Program (FWP), 5, 260 De Kruijk, H., 28, 76 Forestry, 229-31 Desai, Gunvant M., 6 Fuel, 229 Desert Development Program (DDP), 5 "Full package" concept, 167 Donaldson, Graham, 189, 193 Doornbus, Martin, 214 Gangadharan, T. P., 215 Drought Prone Areas Program (DPAP), George, P. S., 214 5, 262-63 Ghaffar, M., 181 Dung (used as fuel), 229 Ghose, A. K., 194, 197, 199 Goats, 226 Education, 3, 7, 40, 250, 271, 272 Gotsch, Carl H., 140, 146, 163, 170, Employment: agricultural growth and, 189, 197, 200 26, 28; in agricultural sector, 2; in Grameen Bank, 147 dairying, 218-21; farm and Green Revolution, 51, 101, 239, 248; nonfarm, 234-35, 241-44; H\V use agricultural growth and, 26, 28; and, 157, 160,177-84, 240, 242; benefits and beneficiaries of, 176- income and, 88-93; industrial 77, 197-200; farm size and, 157-65, development and, 233, 241-44; 197-98, 200; fertilizer use and, 165- mechanization and farm, 176, 177, 76; food prices and, 157, 195-97; 188-95, 200; multiple cropping Hws and, 155,156,157-65; in India, and, 124, 125; occupational 12; landless and, 176-97; mechan- distribution of, 87-93; rural wage ization and, 176, 177, 188-95, 200; labor analysis and, 77-81, 93; rural poverty and, 12; technology schemes for, 5, 6, 7, 262, 265-70. See and, 2, 3, 26; tenancy and, 191-94, also Wages 278, 285, 297, 299; wage rates and, Employment Guarantee Scheme 33, 34 (EGS), 6, 265-70 Griffith, Keith, 194, 197, 199 Exports, 235 Gross domestic product (GDP) per Extension services, 3; "full package" capita, 8, 9 concept of, 167; industrialization Guisinger, S., 28, 185 and, 234, 258; training and visit Gupta, T, 230 system and, 108-14 Gupta, V. K, 228 Family planning, 5, 7, 273 Haider, A. S., 91 Feder, Gershon, 110, 111 Harrison,James Q., 108, 111 Feed (for cattle), 207, 210, 224, 225 Health, 5, 39, 40, 217, 252, 271, 272 Fertilizers: in Bangladesh, 170-76; Herdt, R., 167, 181, 184, 194 crop yields and, 165-76; dung and, Hicks, N., 28, 185 229; in India, 167, 169; loans for, High-yielding varieties: adoption of, 140; multiple cropping and, 118, 19, 155, 156, 157-65; agricultural 123; in Pakistan 169-70; small-farm growth and, 2-3; benefits and 440 Index beneficiaries of, 176-77, 197-200; 159-62; fertilizer use in, 167, 169; crop yields and, 157, 165-76; food prices in, 196; forests in, 229, employment and, 157, 160, 177-84, 230; grain distribution in, 197, 198- 240, 242; food prices and, 157, 195- 99; Green Revolution in, 12; uw 97; Green Revolution and, 155, 156, use in, 185; industrial development 157-65; income and, 155, 157; in in, 244-50, 255-57; irrigation in, India, 185; landless and, 176-77; 128, 129, 131, 132, 133-134; labor landlords and, 48; loans for, 140, demand increase in, 181, 184; 142, 143, 146; mechanization and, landholding in, 35-38, 50, 52, 56, 176, 177, 188-95, 200; multiple 57,62-65,67,70-71; landlessness in, cropping and, 118, 123; rural 72-73; land redistribution in, 300- nonfartn employmentand, 240,242; 01, 302-03, 306, 309, 325, 328, 329, small-farm sector and, 101,108,109; 330, 331-32; livestock farming in, tenancy and, 157, 286. 292; wage 205, 227; mechanization in, 189, effects of, 157, 184-87 191, 194; multiple cropping in, 115, Ho, Samuel P. S., 250 116, 117, 118, 122, 124, 125; Hossain, Mahabus, 56, 147, 170 population in, 8; poverty patterns Household size and composition, 39, in, 12-13; rural nonfarm pro- 40 duction in, 235, 236, 244-50, 255- HYVs. See High-yielding varieties 57; tenancy in, 277-78, 284, 285, 286, 293, 297, 298, 299; training Imports, 225, 235, 236 and visit system in, 108, 109; Income: agricultural output and, 19; unemployment in, 82, 83-85; wage agricultural growth and, 2; Chinese labor in, 77, 88-91 rural development and, 253; Indonesia, 167, 194 consumption and, 234, 237-38, Industrial development (rural): broad 238-39; dairying, 210, 215, 216-17, elements of, 233, 257-59; China 217-18, 221-24, 225; fishery, 229; and, 251-57; consumption linkages iHxv use and, 155, 157; inequalities and, 234, 237-41; credit and, 234, in, 95; of landless, 89; land- 258, 275; employment and, 233, ownership and, 51; livestock and, 241-44; extension services and, 234, 210, 226; means of raising, 87; 258; farm and nonfarm linkages multiple cropping and, 124-25; and, 234-35; India and, 244-50, noncrop, 87-88, 204, 231; nonfarm, 255-57; production linkages and, 87-88; rural employment and, 88- 234, 235-36; public works programs 93; rural nonfarm growth and, 259; and, 260; Taiwan and, 250-51 rural nonfarm production and, Infrastructure: agricultural growth 235-36; tenancy and, 298; training and, 2-3; irrigation and, 131, 270; and visit system and increase in, 111 rural development and, 93, 258, India: agricultural growth and poverty 263-64, 271 in, 26-28; agricultural output in, Insurance (crop), 146 19-23; agricultural wage rates in, Integrated Rural Development 33; antipoverty programs in, 5-6, Program (IRDP), 5, 7, 273-75 262, 271-75; commodity expen- Interest rates, 140, 144-45 ditures in, 238; credit in, 136, 140, International Development 142, 146, 148; dairying in, 210, Association (IDA), 134 212-14, 217; employment schemes Investment: agricultural growth and, in, 262, 265-70; farm size in, 157, 2-3; in land improvements, 286, Index 441 294; rural nonfarm development Landless, 7, 26; agrarian theory and, and, 234, 258 49; analysis of, 73-77; in Ban- Irrigation, 51, 156, 159; agricultural gladesh, 73; consumption and, 237, growth and, 2; agricultural output 240; definition of, 72; Hyv use and, and, 22, 23; in Bangladesh, 128, 176-77; income sources of, 89; in 129, 134-35; credit and, 129-30, India, 72-73; livestock and dairying 132, 143, 146; in India, 128, 129, and, 210, 212, 215, 216, 217, 218, 131, 132, 133-34; mechanization 221, 224, 225; poultry and, 228 and, 188, 189-91; multiple Landlords, 48; Hw use and, 157; land cropping and, 118-23, 125; in improvements by, 294; operational Pakistan, 129,134; sewage and, 115; land and, 63. See also Tenancy Shallow Tubewells project, 134; Landownership. See Landholdings. small-farm sector and, 101, 109,113, Land redistribution, 56; in Bangla- 128-35; Thana Irrigation Program, desh, 300, 301, 302, 307-08, 309, 135 314, 325, 329, 330, 331; fragmented Islam, Faridul, 163 holdings and, 308-14, 316; historical review of (since indepen- Jajmani system, 184 dence), 300-03; in India, 300-01, Jha, D., 189 302-03, 306, 309, 325, 328, 329, 330, Johl, S. S., 146 331-32; overview of, 289-91, 299- Johnston, B., 235 300, 317-18; rural poverty and, 1, 3- Joshi, P., 189 4; schemes for, 303-08, 324-31 Jul, M., 217 Land tenancy. See Tenancy Lau, L., 111 Kahlon, A. S., 146 Leaf, M., 28, 199 Kaneda, Hiromitsu, 170, 181 Leiserson, Mark, 87 Karanjkar, S. V., 227 Lele, Uma, 148, 149, 150, 196, 240 Khan, A. H., 135 Lipton, Michael, 155, 161, 167, 200 Khan, A. R., 13-14, 17, 29, 33, 38, 193, Literacy, 40 322, 323 Livestock farming, 4, 307; in Ban- Khan, D. A., 93 gladesh, 205, 207, 226; economic Khan, M. H., 69, 141, 170, 181, 193, features of, 205-12; feed and, 207, 330 210,224,225; fisheries and, 228-29; Kilby, P., 235 income and, 210, 226; in India, 205, Krishna, R. R., 215 227; in Pakistan, 205, 206, 226; poultry and, 227-28; small stock Labor, demand for, 181, 184 development and, 226-27. See also Lal, D., 34, 185 Dairying, Poultry Land: cost of, 225; rural nonfarm Loans. See Credit production and, 235 Lowdermilk, Max K., 142 Landholdings: agrarian structure and, 51-62; agrarian theory and, 48-51; McInerney, John, 189, 193 in Bangladesh, 38, 52-56; ceilings Malaysia, 167, 194 on, 299-300, 317-18; income and, Manufacturing, 4, 88, 256. See also 51; in India, 35-38, 50, 52, 56, 57, Industrial development (rural) 62-65, 67, 70-71; inequality and, Marginal Farmers and Landless Labor 71-72, 95; operational, 62-71; rural Agencies (MELA), 5 poor and, 35-39, 277 Marketing systems, 148-51 442 Index Marshall, A., 284 Net domestic product (NDP) per Mathai, G., 89, 236 capita, agricultural output and, 18- Mathur, D. P., 228 19, 22 Mazumdar, Dipak, 244 Newbery, David, 284 Mechanization, 33, 196; in Bangla- Nutrition, 272; dairying and, 216-17; desh, 193; employment and, 176, Minimum Needs Program and, 271, 177, 188-95, 200; Hyv use and, 176, 172 177, 188-95, 200; in India, 189, 191, Nyholm, Klaus, 214 194; irrigation and, 188, 189-91; multiple cropping and, 124, 128; in Ojha, Gyaneshwar, 168 Pakistan, 189, 191-93, 194; tenants Operation Barga, 295-96 and, 191-94 Operation Flood, 212, 226 Mellor,John W., 196, 234, 240, 243 Output. SeeAgricultural output Mergos, G., 214 Oxen, 206, 210 Migration, 40, 191, 234, 250 Milk: consumption, 217; imports, 225; Pakistan, 4, 6, 7; agricultural growth prices, 207-10; products, 225, 238; and, 28; agricultural output in, 19, production, 216-218; unions, 215; 23; agricultural wage rates in, 33; yields, 207, 225 credit in, 136, 140, 141, 142, 146; Minhas, B. S., 35, 38, 269, 314 dairying in, 206, 212, 214; farm size Minimum Needs Program (MNP) 5, in, 157, 163; fertilizer use in, 169- 270, 271-73 70; grain distribution in, 198, 199- Mishra, G., 214, 215 200; irrigation in, 129, 134; labor Mixed farming, 217-28, 221-24 demand increase in, 181; land- Mohan, D., 230 holding in, 38-39,50,56-57,69-71; Moneylenders, 136, 142, 144, 147. See landlessness in, 76-77; land also Credit redistribution in, 300,302,303,309, Mukherjie, Robin M., 13, 27 313, 317, 325, 330, 331; livestock Multiple cropping: in Bangladesh, farming in, 205, 206, 226; market- 115, 117, 118,123,124,125; ing access in, 150, 151; mechani- employment and, 124, 125; zation in, 189,191-93,194; multiple fertilizers and, 118, 123; Hyvs and, cropping in, 115, 117,118, 122, 123, 118, 123; income and, 124-25; in 125; population in, 8; poverty in, India, 115, 116, 117, 118, 122, 124, 13-16, 28, 321, 322, 323; public 125; irrigation and, 118-23, 125; works programs in, 259, 262, 263; mechanization and, 124, 128; in rural nonfarm production in, 235, Pakistan, 115, 117, 118, 122, 123, 236; tenancy in, 278-81, 284, 285, 125; small-farm sector and, 115-28 293-94, 297, 298, 299; training and Mundle, S., 198-99 visit system in, 114; unemployment Muqtada, M., 18 in, 82, 85-86; wage labor in, 77, 78, Murti, K, 229 91-93 Myers, Desaix, 135 Pakistan, M. G., 189 Pandey, M. K, 215, 216, 217 Naseem, S. M., 14-16, 38, 170, 189, Pandit, M. L., 248 321-22, 323 Parthasarathy, Gogula, 34, 169, 297 National Rural Employment Program Patel, S. M., 215, 216, 217 (NREP), 5, 6, 7, 270 People's Works Program, 264 Index 443 Perrin, R., 168 259, 262, 263; politics and, 264, 269; Personal responsibility system (PRS), wages and, 262, 263, 265 252, 253 Pesticides, 113, 123 Quasem, M. A., 170 Philippines, 167 Pigs, 226 Ramaswamy, S., 229 Pilot Intensive Rural Employment Ram, G. S., 330, 331 Program (PiREP), 5 Rao, C. H. H., 27, 62, 140, 161, 168 Politics, public works programs and, Rath, N., 35, 307, 328, 329, 330, 331 264, 269 Rent regulations, 293, 296-99, 316 Population: agricultural output and, Rice, 163-65, 166, 169 8, 19, 23; of India, Pakistan, and Rochin, Refugio, 141, 170 Bangladesh, 8; rural employment Rudra, A., 284 and growth in, 243 Rural Works Program, 259, 264 Population pressure theory, 48, 49, 50, Ruttan, Vernon, 157 51 Poultry, 4, 227-28 Saith, A., 197 Poverty, 1, 8, 40-41; agricultural Salam, Abdul, 169, 170 growth and, 2, 23, 26-29; Sanyal, S. K, 65 agricultural wages and, 35; in Savings, 234 Bangladesh, 17-18; dairying and, Scarcity Relief Program, 259, 262 214-16, 224-26; Green Revolution Schaumburg-Muller, Henrik, 214 and, 12; in India, 12-13; land- Schluter, M., 137 holding and, 35-39, 277; land "Semifeudal" paradigm, 286 redistribution and, 1, 3-4; myths Sen, Amartya K, 6, 82 about, 8-9; in Pakistan, 13-16, 28, Sen, Bandhudas, 137 321, 322, 323; patterns of, 12-18; Sewage, 115 programs to fight, 4-7, 262, 271-75; Shah, C., 50 reliability of data on, 9, 23, 321-23; Sharecropping, 278, 283, 284, 285, socioeconomic characteristics of 286, 287, 298 rural poor and, 39-40, 41; tenancy Sheep, 226 and, 277, 283, 288-89, 290 Sidhu, D. S., 181 Prasad, D. S., 169 Sigurdson,Jon, 255 Prices: of food in Bangladesh and Singh, C. B., 330, 331 India, 196; food-grain consumption Singh, Inderjit, 51, 57, 67, 191, 277 and, 239-40; government Sirohi, A. S., 330, 331 intervention and, 23, 146; Hw use Slade, Roger, 110, 111 and food, 157, 195-97; marketing Small Farmers Development Agencies systems and, 149-50; milk, 207-10; (SEDA), 5 rural development program and, Small-farm sector: agricultural output 262 and, 101, 108-14; credit and, 135- Production, 234-236. See also 48; cropping intensity and, 101; Agricultural output definition of, 100-01; fertilizer use Public policy, 3, 23 and, 109-10, 114, 155; Hw use and, Public works: in Bangladesh, 259, 260, 101, 108, 109; irrigation and, 101, 262, 263, 264; Food for Work 109,113, 128-35; marketing and, Program and, 5-6; industrial 148-51; multiple cropping and, development and, 260; in Pakistan, 115-28; risk and, 101; training and 444 Index visit system and, 108-14. See also regulations and, 296-99, 316; scope Agricultural growth of, 291-92; security of tenure and, 4, Soni, S. N., 227 294-96, 316; technology and, 286- Stiglitz,J., 284 87, 288, 289, 290, 292, 299; Subbarao, K, 23, 161 zamindari system and, 285, 293. See Subsidies: credit, 5, 23, 141, 144; also Land redistribution, Landlords, fertilizer, 23; interest rate, 145; Tenancy Integrated Rural Development Terhal, Piet, 214 Program and, 5, 273; small farms Thailand, 167, 194 and, 145; tractors and, 195 Thakur, D. S., 216, 217 Subsistance farmers, 100, 148 Township and village enterprises (TvEs), 253-55, 256, 257 Taiwan: rural industrialization in, 250- Training and visit system (T&V), 108- 51; rural nonfarm production in, 14 235, 236 Tyagi, D., 241 Technology: credit and, 142-44; deriving maximum benefit from, Underemployment, 125, 266 108-09; Green Revolution and, 2,3, Unemployment, 81-86, 233, 242, 262, 26; tenancy and, 286-87, 288, 289, 266; in Bangladesh, 82, 85; in India, 290, 292, 299 82, 83-85; in Pakistan, 82, 85-86 Tenancy: in Bangladesh, 278, 284,285, 286, 293, 297, 298; as barrier to Vyas, V. S., 50, 89, 236 innovation, 286-87; contracts, 285, 291, 292; data on, 277-81; Wage labor (rural), 77-81, 93 economic relationships and nature Wages: agricultural, 29-35; in of, 287-91; efficiency of, 284-86, Bangladesh, 29-33; Green 288, 292; functions of, 283, 292, Revolution and, 33,34; Hyv use and, 316-17; HYV use and, 157, 286, 292; 157, 184-87; in India, 33, 185; income and, 298; in India, 277-78, public works programs and, 262, 284, 285, 286, 293, 297, 298, 299; 263, 265 land improvements and, 286, 294; Westergaard, Kirsten, 214 mechanization and, 191-94; myths Wills, I. R., 140 and realities of, 291-93; operational Winkelman, D., 168 landholdings and, 63; in Pakistan, 278-81, 284, 285, 293-94, 297, 298, Yelamanchili, Satyanarayana, 34 299; poverty and, 277, 283, 288-89, 290; reform, 291-99; rent Zamindari system, 285, 293 I The World Bank Poverty is the grim reality for some 400 million people-mostly small farmers and agricultural laborers-in India, Pakistan, and Bangladesh. To remedy the problem, South Asian governments and international agencies have focused oh raising the productivity of small farmns and increasing opportunities for rural employment. This strategy, however, has long been criticized for doing the poor more harrn than good. Inderi it Singh challenges that pessimistic view by critically reviewing a wealth of evidence from recent academnic literature and the World Bank's operational experience. He shows that rapid agricultural growth has benefited all classes of the poor and that the "great ascent" from poverty to a more materially rewarding life has begun. A variety of programs intended to help the poor directly are examined in detail. Research, extension, and training activities are evaluated for their effectiveness in promoting the adoption of high-yielding varieties of cereal, spreading new farming technology, encouraging multiple cropping, and increasing the cultivation of high-value crops. Singh also considers programs in dairying, poultry farming, commercial fishing, and forestry and argues that policymakers have neglected these potentially profitable activities. Finally, he discusses the dismal failure of land reforms in reducing poverty. Inderjit Singh has held numerous posts at the World Bank since he joined the staff in 1975. He is currently the principal economist in the Socialist Economic Reform Unit of the Country Economics Department. OF RELATED INTEREST FROM THE WORLD BANK Land and Labor in South Asia (Discussion Paper 33) Small Farmers in South Asia: Their Characteristics, Productivity, and Efficiency (Discussion Paper 31) Tenancy in South Asia (Discussion Paper 32) World Development Report 1990: Poverty The Johns Hopkins University Press Couer design by Joyce Petruzzelli ISBN 0-8018-3954-8