Have agricultural economists neglected poverty issues?

Thiesenhusen, William C.

1. INTRODUCTION

In the short term one can be pessimistic about the collective progress of the Third World and its interactions with industrial countries. There is plenty of bad news. With one-quarter of the world's population, industrialized countries consume about 80 percent of the world's goods. With three-quarters of the world's population, developing countries command less than one-quarter of the world's resources. And the imbalance is growing worse. (1) Of the 2.7 billion people in the tropical and subtropical regions outside of China, 40 percent live in poverty; more than 14 million of their children under 5 years of age starve to death or die of disease each year. (2) Furthermore, at the same time as an increasing proportion of the population of Africa is composed of young people (65 percent of its population is now under age 25), education budgets are being cut--from $10.8 billion in 1980 to $5.8 billion in 1986. (3) In an article assessing the globalization of economies, Richard J. Barnet writes: "Poverty, population pressures, civil war, and repression are turning Sub-Saharan Africa--black Africa minus South Africa and Namibia--into a giant disaster zone, and in countries in South America, such as Colombia and Peru, the civil society is dissolving. In the Philippines more than seventy percent of the population is poor by any human standard. With the end of the Cold War, the increasing marginalization of the Third World appears likely." (4) The predictions are ominous. Barnet concludes his article, written before the crisis in Iraq, by speaking to an industrial-country audience: "There is no real north-south dialogue, and politicians in the industrial world feel little pressure to begin one. The end of the third global conflict of the twentieth century offers a unique moment to re-examine basic assumptions about the political organization of the planet. Unless the people of Asia, Africa, and Latin America are part of a new democratic order, we will all lose the chance for peace the world has won." (5)

But taking a longer and more disaggregated view--looking at progress since the Second World War--can engender a buoyant feeling that the future does carry promise. Some countries which all but defined the underdeveloped world forty years ago are Newly Industrial Countries (NIC) today. And there has been forward movement even where progress was less dramatic, in countries whose eventual NIC status is confidently forecast. Agricultural productivity has taken quantum leaps. Some diseases which were sapping industrial productivity years ago are now eradicated and the productive life spans of populations have been stretched. Despite the negative overall figures cited, education has become more available to the masses of mankind, even rural dwellers. The Inter-American Development Bank, after chronicling a disastrous decade--one which saw Latin America in 1989 with the same per capita income it had in 1977--and even though expenditures on health fell in real terms by 5 percent in Costa Rica, 15 percent in Chile, 10 percent in Brazil, 22 percent in Venezuela, and 30 percent in Uruguay, (6) concluded that life expectancy, illiteracy rates, and infant mortality rates continued to improve in the 1980s. (7)

In a sense, these long-term positive indicators show a path for laggard countries today and are especially inspiring since some of the countries making the most progress toward NIC status were unlikely candidates in the 1950s. It is safe to say that most of those who live in the Third World are considerably better off today than they were then. Indeed, with all the problems that still attend development in the Third World, future historians may read these last four or five decades as a time when there was the beginning of a true global concern for the environment, for economic and social development, for the amelioration of societal inequities--an era marked by more security of expectations than any in the past. (8) Although it is scant comfort to the hungry, perhaps the most important progress has been made on attitudes. We may be at the beginning of an era when poverty, ignorance, and disease are not to be accepted as an inescapable lot but as conditions that can and ought to be removed by policy and effort. Of course, we are a long way from that goal in low-income countries, those at the bottom of the GNP per-capita list. And whether new awareness and concern will translate into action is a matter for speculation.

Since so many of the root problems in Third World countries are with the rural sectors, agricultural economists have a special responsibility in development concerns: they are, by definition, applied social scientists, and they investigate the empirical bases for real world economic problems and devise suitable alternative policies, The contributions of agriculture to economic development are: (1) to produce a surplus and transfer it to nonagriculture, (2) to provide markets for urban-based industry, (3) to maintain a labour reservoir, (4) to assist in capital formation, and (5) to accumulate foreign exchange. Agricultural economists exert much effort researching these matters.

Yet, as I read the economics journals and examine the career trajectories of young agricultural economists, I have become concerned that there is a diminished amount of intellectual activity devoted to coping with rural poverty in the Third World. Frankly, I worry about a recent disciplinary penchant of researching very narrow problems with considerable rigour and often little data to the exclusion of empirical examination of some of the more sweeping economic concerns confronting oar societies.

In this paper, I deal with five questions:

(1) Why do agricultural economists neglect rural poverty?

(2) Why does the presence of large amounts of rural poverty imperil development?

(3) Should rural poverty be treated in situ?

(4) What is the shape of the current growth-equity debate as concerns poverty reduction?

(5) Are there any new and promising methods for coping with rural poverty?

My plea is for more research into neglected problems of Third World poverty, particularly in the countryside. The World Bank estimates that a billion or so contemporary people--a good one-fifth of mankind--can still be considered to be in the grip of poverty. Poverty is predominantly a rural problem, accounting for 60 percent of the so-called hungry-poor in Latin America, 80 percent in Asia, and 90 percent in Africa. (9)

Suggesting poverty research in the Third World is not novel or even original. And, happily, it does seem to be of renewed current concern, mainly on the part of international organizations as the 1990s commence, perhaps because structural adjustment has wrought harsh, though, it is hoped, short-run difficulties to some categories of the poor. The Asian Development Bank devoted a recent issue of the Asian Development Review (10) to the matter; the World Bank featured it in its most recent World Development Report. (11) The draft "State of Food and Agriculture", to be published by FAO in 1991, contains a section on rural poverty and structural adjustment. Daedalus, the journal of the American Academy of Arts and Sciences, devoted much of its winter 1989 edition to the issue. A good bit of the 1990 report of the Inter-American Development Bank, Economic and Social Progress in Latin America, is devoted to poverty problems of women in Latin America's work force, and you were addressed on it in your January meetings in 1989 and again last year. (12)

2. ADDRESSING RURAL POVERTY

FAO (13) sets out some characteristics of the Third World rural poor, defined as those poor who either work on farms or do nonfarm work that depends in part on agriculture:

(1) They are heterogeneous, including small-scale farmers, the landless, nomads, pastoralists, and fisherfolk;

(2) The landless and the near-landless constitute the largest single group of those afflicted by rural poverty: some 30 million agricultural households are landless and 138 million are near landless; landlessness is especially serious in South Asia, but numbers in this category are growing in all of the Third World;

(3) Poverty affects an average of half the rural population in the Third World, with Africa exhibiting the highest incidence (though the situation is somewhat better in the Near East);

(4) Close correlates with extreme poverty are undernutrition, high child mortality, and illiteracy;

(5) Resources to accommodate the poor are shrinking. In the 1975-80 period, absolute numbers in the agricultural population expanded by 103 million; the comparable 1980-85 figure was 96 million. At the same time, expansion of arable land in the former period was 20.6 million hectares as opposed to 11.4 million hectares in the latter period;

(6) The poor turn to nonfarm jobs when they can. There is an inverse relationship between nonfarm employment of the rural labour force and farm size, and nonfarm employment is especially utilized by the rural poor during slack seasons in the countryside;

(7) Rural poverty stems from underemployment, which is rapidly being translated into unemployment in urban areas. Between 1950 and 1980, the population of Africa's largest cities increased more than sevenfold while those of major Asian and Latin American cities either tripled or quadrupled;

(8) Poverty is closely correlated with high rates of sectoral economic growth, and economic downturns affect countries with a higher percentage of people in poverty more adversely than those with fewer poor people; that is, economic slowdowns are more precipitous in countries with larger groups in poverty. There is some tendency for low or falling wages to be associated with low or falling rates of agricultural growth;

(9) Rural poverty is associated with countries experiencing the greatest fluctuation in the price index for food;

(10) The Inter-American Development Bank adds that rural women and children are more severely affected by poverty than rural men, and it is often the work of rural women that pulls families out of a "poverty" category; (14) and

(11) A recent volume, Rural Poverty in South Asia, (15) points out that international migration and the repatriation of remittances have recently proved essential for rural poverty alleviation in some countries.

Why Do Economists Avoid Rural Poverty?

The first reason why agricultural economists shun studies of rural poverty comes from the complexities which these characteristics introduce to any analysis of the rural poor:

(1) Since the poverty-stricken combine rural and urban as well as domestic and international strategies to raise their incomes, the sectoral and national limitation of any analysis is difficult, if not impossible. To do their jobs, agricultural economists find themselves working out of their sector and in situations which transcend national territories. We continually are being made aware of how interconnected all of our countries have become--through our factor and product markets, through our common environment, through technology, through improved communications. The worldwide trend to free trade areas, which will have a profound effect on rural labour and rural poverty, merits more careful thought and attention. How rural poverty will be affected by the global economy is scarcely mentioned except to assume that "a rising tide raises all ships";

(2) Since poverty impact on gender and on age groups is differential, analyzing family income without these considerations is fraught with problems;

(3) Since some aspects creating poverty are transitory, such as structural adjustment, short- and long-term analyses and projections are required;

(4) Since some aspects of poverty may be solvable only with allocation of new resources, economics is entangled with more broadly social issues which also must be addressed. By definition, coping with rural poverty implies that growth and equity issues be addressed, and this means a square confrontation with politics. In sum, problems introduced by rural poverty tend not to obey the boundaries imposed upon them by academic disciplines; and

(5) Since the rural poor are such a heterogeneous lot, a government policy favourable to one group may wreak havoc on another. Raising producer prices, for instance, may help the land-poor but worsen the situation of the landless.

In sum, the "simple model" is not readily adaptable to problems which are this encompassing, though models will be absolutely essential to analyzing component parts of the issue.

The second reason that causes agricultural economists to avoid the issue of rural poverty has philosophical roots: Can we address poverty objectively? Some think not and stay away from the problem because of ideological convictions. However, for the Third World, one of the refreshing by-products of the recent liberating events in the USSR and Eastern Europe and the passing of the Reagan era in U. S. politics is that discussions on such issues can now be less polarized. The debate on where within the spectrum from free markets to Marxist countries the Third World ought to be has the promise of being more open and pragmatic now than in the recent past. There is scope for both private action--where private initiative needs to be unleashed, entrepreneurship increased, and investment rewarded-and, when needed, governmental redress.

On privatization and laissez faire, John Kenneth Galbraith warns that it is not true that there is an essential therapy that comes from human deprivation and unemployment and that "out of unemployment and hunger will come a new and revitalized work ethic" (16) and enhanced economic growth. Thus, even in an economy that is "privatizing" and in other ways making a transition to freer markets, there is a place for social expenditures on matters such as poverty alleviation. Galbraith reminds us of the blinded vision that free-marketeers sometimes exhibit when poverty is the subject: "Nothing over the centuries has more often been urged than the social reward of hardship by those who will not have to suffer it". (17) Galbraith concludes, "In a very real sense, both East and West, our task is the same: it is to seek and find the system that combines the best in market motivation and socially motivated action". (18)

A third reason for agricultural economists to avoid rural poverty issues is that some of the reforms required for poverty alleviation, such as land reform, result in lower labour productivity on specific pieces of reformed property in the short run, even though there is higher production per unit of land almost immediately. Then again, some agricultural economists have become satisfied over the years with "policies" that raise labour productivity in agriculture at the expense of enormous waves of urban migration.

Fourth, agricultural economists are often committed to policies that benefit commercial farming at the expense of farm labour and small-scale agriculture.

And fifth, agricultural economists frequently undervalue the contributions to development made by the rural poor. It is sometimes assumed that land-poor rural people make only a marginal contribution to production and that when compared to large producers who are ordinarily favoured with more than their share of good land, technology, market facilities, and credit, small farmers are insignificant contributors to overall growth. In Brazil, as suggested by a strong inverse relationship between farm size and production per hectare, this is not true. The contributions of small farmers to production in Brazil are substantial. (19) With Melmed-Sanjak, I examined the 1980 agricultural census, the latest available in 1990, to determine the contributions to agricultural production of farms under 50 hectares, which occupy only 12 percent of the agricultural land in that vast country. These small farms utilize 28 percent of agricultural capital (in value terms) but employ 71 percent of agricultural labour and occupy 39 percent of the country's cropland. This cohort of small farms produces half of the income of the agricultural sector and 40 percent of gross income. Around 80 percent of this product is marketed, mostly for domestic consumption (see Table 1). A U. N.-sponsored study of the 1970s corroborates our work, showing that 41 percent of Latin America's domestically consumed agricultural products were produced by small farmers and that this sector also accounted for 32 percent of the region's exportable agricultural products. Indications are that the percentages were similar in the 1980s, even though 62 percent of these farm families were below the poverty line. (20) In other words, the commodity contribution of the land-poor is substantial; they contribute more than proportionally to agricultural output, the vast bulk of which is transferred via the market as agricultural surplus. Furthermore, the production process of small-scale farmers is factor-rational. In a worker-plentiful, capital-scarce economy, small farms utilize more than their share of labour and save on capital.

Why Does the Presence of Large Amounts of Rural Poverty Imperil Development?

There are economic as well as moral penalties to be paid by countries that do not work to alleviate their rural poverty problems. If these issues are not addressed, growing poverty hobbles the chances of reaping rewards from globalization. Can a country like Brazil, for example, reach her potential on the world stage with fully one-third of her population--the rural landless; the urban homeless and marginals; and many of the tertiary, informal sector of the economy--participating only marginally in the market economy? If the number taking part in the growth process is substantially below the potential, will not growth itself be jeopardized?

At another level, the Bretton Woods institutions emphasize that countries with an outward-oriented growth pattern have fared much better than those which, behind high tariff walls, produce primarily for internal markets. For the past several decades, in fact, the progress of exporters has been teaching this lesson, one only recently learned in Latin America, where, led first by Chile and joined now by Mexico, there appears to be a shift toward freer trade.

The poverty rate is presently low in those East Asian countries which serve as models for would-be export promotions, those which attained NIC status in the 1980s. Unemployment in these countries is low, literacy is high, and health conditions are much improved, even for those at the very bottom of the income distribution spectrum. Why this occurred has escaped many analysts: while great progress in poverty reduction has taken place because of the NICs' export orientation, chroniclers of their histories often fail to give adequate attention to how their internal markets were built, their institutions reshaped, and their land tenure patterns reformed so that the formerly excluded poor enjoyed at least some immediate growth benefits. In the NICs (especially Taiwan but also South Korea--and now Thailand and Malaysia), most of the peasantry seem to benefit, though, of course, not to the same extent as the rich.

These countries teach that growth, distribution, and amelioration of poverty are importantly linked. Reforming institutions early in the development process, perhaps through land reforms or other distributional policies [like the New Economic Policy (NEP) in Malaysia], is a measure to be ardently sought. Central America in the 1970s illustrates the perils of economies which grow rapidly but are institutionally biased in such a manner that the poorer groups are excluded.

A second major economic reason for alleviating rural poverty is its link to environmental degradation, clearly made by the World (Brundtland) Commission on Environment and Development: "Those who are poor and hungry will often destroy their immediate environment in order to survive: they will cut down forests; their livestock will overgraze grasslands; they will overuse marginal land; and in growing numbers they will crowd into congested cities". (21) The solution lies in the intermediate (but, as yet, not very operational) concept of "sustainable development", which the report defines as "development that meets the needs of the present without compromising the ability of future generations to meet their own needs"." The report concludes, "[This] is not a prediction of ever increasing environmental decay, poverty and hardship in an ever more polluted world among ever decreasing resources. We see instead the possibility for a new era of economic growth ... absolutely essential to relieve the greater poverty that is deepening in much of the developing world". (23)

In support of this idea, Martinez-Alier documents several ways in which poverty is a cause of environmental degradation: it promotes overuse of firewood and the eventual use of dung for heat instead of fertilizer; it results in land overuse and eventual deterioration; it necessitates off-farm work such that semi-proletarians do not have time for conservation practices. (24)

There is an unfortunate side effect to linking the poor with environmental damage where there is great inequity in resource distribution, for it seemingly results in blaming the victim for environmental degradation. (25) Keyfitz provides a recent example: "In Brazil peasants from overpopulated regions have destroyed millions of acres of rain forests in an attempt to eke out a living from soil that is essentially unsuitable for farming". (26) Even Martinez-Alier notes, "The social differentiation of a peasantry creates marginal cultivation, which in turn leads to soil erosion since they are concerned with immediate survival". (27) And again, "In some poor countries, sheer population pressure has made soil erosion worse by driving the landless onto marginal lands such as hillsides and forests. The growing shortage of firewood often means an ugly cycle of denuded hillsides, flash floods, silted rivers". (28)

In fact, the peasant is only the most immediately responsible party for environmental damage; if blame must be apportioned, it is equally, if not more, appropriate to charge the rural land-tenure system, which allows rich landlords to monopolize the best resources in the region and often, as reported in literature on the inverse relationship between productivity and farm size, use them wastefully. Illustrating by reference to peasants who have moved to the Brazilian frontier, Mahar reasons, "It would be wrong to place all of the blame on the settlers. Pushed by poverty and skewed land distributions in their region of origin, the settlers have merely responded to incentives created by the government". (29)

The threat to natural resources from inequalities has frequently been asserted. Martinez-Alier notes that "environmental degradation and poverty become tied together ... because of inequality of distribution". (30) Hecht has concluded: "It is ludicrous to describe environmental degradation as only a function of demographics. Rather the situation is due to the extraordinary maldistribution of land ...". (31) Mahar writes that "the real causes [of tropical deforestation] are poverty, unequal land distribution, and low agricultural productivity combined with rapid population growth". (32) The World Resources Institute claims: "The real causes of deforestation are poverty, skewed land distribution (due to historical patterns of land settlement and commercial agricultural development), and low agricultural productivity". (33)

Painter claims that "the critical issue underlying environmental destruction is the gross inequity in access to resources". (34) The Brundtland Commission also draws attention to the importance of greater equity in land distribution to improving environmental conservation. (35) Foy and Daly elaborate, arguing that environmental problems are caused primarily by maldistribution of resources and total human resource use relative to the carrying capacity of the environment and only secondarily by misallocation of resources due to government policy or market failure. (36) Haiti, El Salvador, and Costa Rica, they say, "stand as warnings to large countries" like Brazil. Foy and Daly report that in the three countries, environmental degradation is not a relatively simple technical matter of "getting the prices right"; rather, it relates to more complex issues involving the political structure itself. (37)

Examining the relation of unequal resource distribution to the environment involves a recognition that groups of users place differential demands on resources; the poor have a proximate, direct effect on them, while the impact of those who own ample land and water is indirect but just as decisive, if not more so, in determining resource destruction. Indeed, it appears that if the environmental problem is ever to be alleviated, the lock which large landlords have on most of the productive resources in some Third World areas must be substantially modified or broken through land reform. Problems of deforestation, for example, probably need to be solved in the major farming regions of the countries involved, far from the site of actual resource damage.

Should Rural Poverty Be Treated in Situ?

Urban poverty focuses more attention on itself than rural poverty; hence, it is more quickly transformed into a political problem, thus becoming a target for government policy. As poverty is transplanted to urban areas when rural areas modernize, those concerned with development are presented with a dilemma: Should they recommend institutional change and more appropriate technology that will be labour absorbing and will enable more intensive use of the available resources in rural areas, thus treating the poverty problem in situ, or should they permit labour productivity to rise in farming and allow poverty to become an explosive urban issue when workers move en masse to cities?

The issue of the locus where poverty will be addressed comes up in the Brazilian case, and whether economic policy is addressed to rural or urban problems shapes its cost. In the last several decades, Brazil has seen an unprecedented migration to towns in advance of economic opportunity, often resulting in increased unemployment and homelessness in the urban areas. Because much of the land in the country is tied up in large estates and plantations, successive waves of agricultural commercialization and mechanization have pushed labour off the land, with resident farm labour and even sharecropping becoming outmoded institutions because of the infrequency of their occurrence. Yet there remain about 10 million landless farm labourers who depend mainly on agriculture pursuits. Remarkably, there are some 35 million hectares of idle agricultural land tied up in private estates, the census reveals, enough to give each landless peasant 3.5 hectares--a king's ransom in Asian terms, and an event about as politically likely as Brazil's granting its Amazon rain forest to Bolivia.

In fact, if land transfers to needy peasants were possible, some $ 3,500 (in U. S. dollars) per rural job created would be needed. That amount would pay for the land at current prices and provide a modest number of inputs to new owners. In contrast, to accommodate the 10 million labourers in town would involve a cost per job established that is severalfold that amount. (38)

The lesson here is that it is possible to alleviate poverty in the countryside by utilizing idle capacity and the production process. However, linking poverty to production is virtually impossible in the urban setting, especially where internal migration has been substantial. In towns, poverty alleviation is usually attempted with rudimentary welfare systems and spending programmes, which quickly run aground for lack of funds and political will. Since no urban safety net is adequate for the job, city services are strained, income polarization increases, urban homelessness proliferates, and repression mounts.

Most development technicians would opt for policies that would alleviate poverty on the farm not only because they are cheaper but also because acute concern for the polluted urban environment has arisen during the last decade. But the political barriers which mitigate against a countryside solution are formidable. As rural-to-urban migration reaches some sort of balance point, however, and the challenged middle classes of the city gain some political ascendancy over the elite rural classes, policy-makers should recognize that the alleviation of poverty through rural redistribution and reform may well become possible.

Also, rural poor people in Third World countries are obtaining a more articulate political voice. Complaints from those in rural poverty are increasingly enunciated with the works of rural area religious groups and nongovernmental organizations (NGOs), better methods of communications, political parties anxious for the rural vote, rural unionization, and the cooperative movement.

If rural poverty problems should be increasingly addressed in situ, how should they be treated? There are five major ways to ameliorate rural conditions in the Third World; specific poor-group targeting is possible with all of them: (1) increase productivity of those with some land; (2) couple more peasants with land in settlements or in agrarian reforms; (3) increase the health and skill levels of those in poverty; (4) pull rural poverty groups into more farm, nonfarm, agroindustrial, urban, or international employment; (5) channel government spending to the poor. The last is, in general, not a viable option in the Third World; Pakistan is known for some degree of success in the first and the last of these; (39) and currently, agrarian reform is an obvious but politically difficult option. (40) On the fourth method, Adelman cautions, "Great Britain, Belgium, France, Switzerland, Germany, Denmark, Sweden, Norway and the U. S. all experienced declines in the shares of income accruing to the poor and increases in absolute and relative poverty during early phases of their Industrial Revolutions". (41) She notes that those periods of maldistribution were long lasting--that is, from two to six or seven decades--but estimates that they would probably be confined to one generation in the contemporary Third World. Yet industry and agriculture are quite different in this respect. Adelman argues that agricultural development utilizes precisely the factors that the farm sector has in abundance--unskilled labour. For example, "planting of HYVs [high-yield varieties] on irrigated land can produce an increase in the demand for unskilled labour by a factor of 7.5". (42) In Pakistan, one can argue "agriculture first" in the course of economic development primarily because of the sector's efficient use of labour and its production of food. The latter, if imported, can be a lavish user of foreign exchange. Hard currency, however, is best directed elsewhere, bringing in technology and key industrial inputs.

What is the Shape of the Current Growth-equity Debate as Concerns Poverty Reduction?

Poverty has two roots: one is lack of economic growth; the other, inequitable income, wealth, and resource distribution. Major components of the latter are concentration of assets in the hands of a few, inadequate schooling, lack of social services and infrastructure, high fertility, and employment problems.

These days there is little debate on the point that agricultural growth is the indispensable component to rural poverty alleviation. John Mellor has noted that "the problem of decreasing poverty and improving the nutritional status of the poor must be seen as one of bringing about broad-based development with agriculture as the leading edge". (43) To accomplish this, he calls for (1) well-operating markets, (2) an indigenous agricultural research system, and (3) massive investment in rural infrastructure.

The World Bank, in its analysis of a dozen or so countries in which data have been relatively good over a fairly long contemporary period, argues that changes in the incomes of the poor can be explained by decomposing them into that part attributable to overall economic growth and that part ascribed to differences in the inequality of income distribution. (44)

In this analysis, unfortunately, rural and urban poor were not disaggregated, a task which must be accomplished some day. The World Bank displays the change in poverty as measured by a head-count index and compares it with a simulated change in poverty, or the change that would have occurred if inequality had either remained constant or moved in a favourable direction; if, that is, everyone received the same percentage increase in income which growth had generated or the percentage gain was somehow "progressively" distributed. This implies that if the actual decrease in poverty is less than the simulated decrease, then growth increases inequality: the poor gain less than the nonpoor. If the actual decrease in poverty is bigger than the simulated decrease, the opposite is true. The Bank's data show that in India, the actual and the simulated changes in poverty were almost exactly the same--that is, the poor enjoyed about the same percentage increase as everyone else over the 1972-83 period--and that, therefore, income attributable to growth was distributed in about the same way as initial income.

There is no case in the World Bank data where the effect of growth was exactly offset by increasing inequality, but there were many cases in which inequality worsened the poverty problem.

Brazil, for example, was enjoying a very high rate of growth during the study period, but concomitantly inequality was becoming a more severe problem. Given the growth it had experienced between 1960 and 1980, a drop of poverty incidence of 34 percent was forecast by the World Bank model. But since inequality increased, there was a drop of only 29 percent in poverty incidence. And if distribution in Brazil had proceeded to the same progressive extent as Malaysia, given Brazil's growth rate, poverty reduction would have been in the order of 43 percentage points. So the pattern of growth and the way the income effects of growth are distributed as well as its rate are important determinants of changes in the incidence of poverty. In Malaysia, growth was not as rapid as in Brazil, but income effects from it were distributed more equitably, and while a 19 percent reduction in poverty was forecast, actual reduction was 23 percentage points. High growth does not indicate progress in solving social problems. Pakistan is a high-growth, least-developed country, and Brazil, a high-growth, middle-income country. Yet Brazil has one of the highest mortality rates for children under 5 years of age among the LDCs and Pakistan has one of the lowest primary school enrollment rates for children. In the latter case, failure to improve the skills of the labour force limits the extent to which the poor can benefit from growth, the World Bank concludes. (45)

Levels and rates of growth of GNP do not always limit the progress of social indicators. One peculiar characteristic of development over the last several decades is the fact that social indicators have tended to run ahead of economic ones. And in some low-income countries, like China, Sri Lanka, and the Indian state of Kerala, per-capita income levels are very low, but other indicators--such as infant mortality, life expectancy, literacy, and nutrition--exhibit surprising progress, though the sustainability of that progress is in question, especially when growth rates slow and expenditures for nonsocial programmes rise. A country's solution to problems attendant upon poverty, in other words, depends on its GNP (a rough indication of its capacity to act) but also on its distribution of wealth (a sign of its public will). [One important limitation to poverty alleviation is that the Third World spends half of its total annual expenditures for maintenance of military establishments and servicing the national debt.] (46)

Focusing on absolute and dire poverty, that is, the poorest tenth of the population (which is most likely to be rural), all countries in the World Bank sample enjoyed larger-than-average percentage increases to this group except Brazil and Costa Rica, where relative inequality worsened over the period studied. Even at that, the poorest tenth enjoyed an absolute gain in Brazil but not in Costa Rica.

No simple pattern for poverty reduction emerges from the World Bank analysis. In low-growth Colombia, a country which also demonstrated some reduction in income inequality during the study period, the poorest tenth of the population did worse than the population at large, for mean income growth dropped by 11 percent in the decade ending with 1988 while income for the poorest tenth of the population fell by 20 percent. The opposite was true for Malaysia, a relatively high-growth country which also showed some improvement in income distribution in the period examined.

Countries in the World Bank sample which experienced the fastest reduction in poverty showed relatively rapid economic growth. In more egalitarian countries, a somewhat lower rate of economic growth served to reduce poverty, but more egalitarianism coupled with rapid economic growth was the surest recipe for poverty reduction.

Growth without a fairly equitable distribution is a very blunt instrument for poverty alleviation. Even so, one researcher has recently concluded that "in terms of satisfying basic needs, average real income growth is much more decisive than changes in income shares". (47) But growth can be honed with institutional change and public policy so that improvements in distribution channel income gains to lower strata of society, perhaps the most enduring manner of "targeting" a solution for groups in poverty. On the other hand, redistribution without growth is equally suspect: in the absence of growth, it is a once-and-for-all event. Fortunately, since Hollis Chenery's classic monograph (48) and the real life experiences of the four East Asian tigers, we know that in due time both equity and distribution can be achieved with some positive relation with one another and not as a tradeoff.

Are there any New and Promising Methods for Coping with Rural Poverty?

The trend to privatization over the 1980s and the resultant gaps in government services have introduced a new feature which is potentially important in alleviating rural poverty in many Third World countries.

This novel element amounts to a virtual proliferation of nongovernmental organizations, which have appeared to act in concert with an array of grass-roots organizations and are now serving to organize and focus the efforts of the rural poor on solving their own problems in certain communities. They also may serve to counteract local elites and the tendency toward centralization of resource management (characteristic of many Third World countries); to petition the government or private banks for resources; and to divide, use, and monitor inputs that are forthcoming. The function of NGOs in providing some surveillance over factor and technology use may prove an invaluable supplement to government extension services; for these organizations may provide an esprit de corps for economic activity which local groups of farmers often find missing in programmes managed by government technicians.

While often short on capital and nontraditional in style, these innovative organizations may stretch existing capital to more recipients or over projects of increasingly varied risk. At the same time, NGOs can make that expenditure more efficient by focusing it on specific poverty groups, expanding it to new and perhaps controversial areas, and possibly even adding a catalytic feature to the investment package. These NGO/grass-roots organizations may indeed function as quasi-capital, to use a term of Simon Kuznets, or even quasi-technology. They may make it easier to mobilize and even to maximize local economic activity because of their democratic, grass-roots, egalitarian, and self-help characteristics.

These NGOs/grass-roots organizations tend to work on some fairly welled-fined areas of rural development, usually among those farmers with some resources: peasants with some land, fisherfolk with some crafts, artisans with some skills and talent. Their efforts tend to concern increasing farm production, adding commonly managed community projects, augmenting women's income, enhancing environmental protection, serving to disburse productive resources like credit, and so on. The existing asset endowment of the poor is taken as a given, but the participants often receive rural credit, technology, basic information on how to access other resource pools, and so forth. Credits rather than grants are emphasized. These programmes, projects, and organizations frequently are sponsored by church groups, foundations, political parties, activist assemblies of one type or another, and even private enterprise itself.

Omissions of some groups in rural poverty in this network--the landless peasant, especially--are also noteworthy; these latter tend to be covered more by their inclusion in advocacy rather than resource-enhancing areas of NGO endeavour (for example, the landless pressing for agrarian reforms or higher wages and benefits; the underprivileged pressing for human rights considerations, rights for indigenous peoples, benefits for women and children, and the like). The very fact that they are left out of these programmes re-emphasizes the need to operate in a way that raises the incomes of those who lack basic resources, those who, in other words, constitute the very poorest of the rural poor.

Some elements of these programmes are appropriate for economic analysis and, in the main, their potential for alleviation of rural poverty has been neglected by scholars and policy-makers, though it is significant that an article in the World Bank journal, Finance and Development, for December 1990, called for a more serious consideration of the issue. This report concluded, "In 1983, NGOs from developed countries provided grants of some $ 3.6 billion (equal to 13 percent of official development assistance), mostly raised by themselves, to developing countries. By 1987, such NGOs were providing about $ 5.5 billion--nearly $1 billion more than credits from the International Development Association in 1988". (49) Kenya reports 130 NGOs interacting with 26,000 grass-roots organizations in 1988; Bangladesh has some 10,000 NGOs, 230 of which receive foreign funds. NGOs also were cooperating with 50 World Bank projects in 1990, and bilateral aid programmes also use them as convenient vehicles of assistance. Of all aid to developing countries, 64 percent is bilateral, 22 percent is multilateral, and 14 percent is nongovernmental. Growth of the activity of these organizations has been fairly rapid, but until lately the bulk of their activity was so small that they could be easily neglected by analysts.

An interest in the NGOs and in corresponding grass-roots organizations in the alleviation of rural poverty comes not only from the mounting size of these groups in terms of their contributions to foreign assistance but also from their unconventional manner of operating, from the flexibility they exhibit, from the charisma their employees often demonstrate, and from their success in arranging the institutional patterns necessary for small-scale development (for example, providing an integrated package of inputs or services, from health care to adult education and agricultural technology). The important questions: Are there some aspects of the work which NGOs do through their constant experimentation that are worthy of translation into larger nationally sponsored or international efforts? Or is this work so unique, so small, and so circumscribed that NGOs have a distinct sphere of influence which cannot transcend local effort? Do these organizations contain the seeds of their own destruction? Are they, in other words, by virtue of the democratic principles which characterize their essence, doomed to consume their scarce capital or to be riven by internal dissension? How can these eventualities be prevented?

I contend that the NGO/grass-roots-organization programmes--which illustrate varying, and possibly replicable, approaches to rural poverty and which do indeed link growth with distribution and social progress--are not sufficiently researched.

Such programmes often include the following, which deserve attention:

1. Successful mutual-liability group efforts at lending, where the groups themselves are the collateral;

2. Revolving credit funds which are community managed;

3. Attempts to reach the poor with education and agricultural technology but without asset redistribution;

4. Government "service vouchers", which the public sector distributes to target groups for technical assistance (this allows "consumers" who are unhappy with the extension help they are presently receiving to recur to other private groups which offer the same service); and

5. Efforts to move higher-value crops, technology, inputs, and credit, once open only to the richer farmers, to the poorer peasants.

To offer an example of the last-mentioned programme: traditionally, export crops have been the domain of elite, land-rich farmers in Central America. Recently these higher-value exportables have been moving, via micro-programmes on private company initiative, to peasant farms, where their cultivation raises with the following questions: (1) Are they environmentally sound, that is, do they promote too lavish use of chemical inputs? (2) Is contract farming, which sometimes is necessary for the incorporation of high-value exportables into peasant farming, so geared to the profits of large farms and/or private corporations that the small farmers are invariably shortchanged? (3) Does high-profit farming by small farmers so raise land prices that the gains of the few are swamped by the losses of the poorer nonparticipants? (4) Once the production of exportables has broken the nexus between peasants and subsistence, are the small farmers left on such unfamiliar territory that they end up more poorly than before?

What I have urged in this paper is something pretty homely: more study of rural poverty issues and empirical study of programmes aimed at helping the rural poor increase their incomes and the application of those lessons to national development.

Comments on "Have Agricultural Economists Neglected Poverty Issues?"

Professor Thiesenhusen has presented a radical paper. It is radical for two reasons. One, he links poverty to the concentrated land distribution in developing countries. And suggests we do something about it. Two, structural adjustment, women's development, and environmental degradation, all very pressing concerns on their own, are here linked to poverty very significantly.

I have two comments on the paper. One, to alleviate poverty there is a need to examine the causal mechanisms that generate poverty. So there is need for more detail here. Two, I disagree on the potential of NGOs in alleviating poverty in the long run. Let me take the points of agreement and disagreement in turn.

Inequal Land Distributions

The argument for land reforms is simple and compelling. 60 to 90 percent of the hungry poor live in rural areas of less developed countries, (LDCs). The landless and near landless are the single largest group of the rural poor. Their underemployment and unemployment leads to poverty. So land reforms have two arguments. Land for the land hungry will generate a permanent income stream for them, rather than leaving them as permanent basket welfare cases which in any case no LDC can afford.

Land redistribution will increase the number of small farms and reduce the number of large farms. Small farms absorb more labour per acre. This will increase employment. And Professor Thiesenhusen shows through the case study of Brazil, that this employment generation in the rural areas is more cost effective than a similar effort in the urban areas.

We do have a common worry that small farms have a higher output per acre but a lower output per labour compared to large farms. So land reforms will increase output per acre but reduce output per labour. But I suggest that there are two plus points here and only one minus. Total output will increase. And output per scarce resource, which is productive land will increase. Labour is not a scarce resource, so if output per labour drops this is a secondary problem to be dealt with through technical change in the long run.

I will add that lack of credit is critical for poverty alleviation. Imperfect credit markets bias income gains against the poor.

Structural Adjustment, Women, and Conservation Link to Poverty

Professor Thiesenhusen's survey of surveys highlights a number of interesting findings. He emphasises that growth is not necessarily linked to poverty alleviation. A redistribution policy is needed as in Malaysia. He argues that poverty affects rural women and children more than men. And that rural women's work pulls families out of poverty.

His most radical stand is that a survey of studies shows that poverty causes environmental degradation in a variety of ways. The argument is that a skewed distribution of productive assets drives the poor to strip productive assets with future potential for present consumption, e.g., forests, uplands erosion etc.

Causal Mechanisms of Poverty

For poverty alleviation we need as detailed scenarios of poverty generation as possible, otherwise poverty alleviation programmes will merely treat the symptoms rather than the root causes. Taking the example of Pakistan, I would like to emphasize one major causal mechanism that has been at work.

In canal colonies and irrigated areas, the green revolution technology has enhanced profitability. But landlords sharecropping out land have only had half the increase in profitability compared to self-cultivators. This has created a powerful incentive to evict sharecroppers, where share rents could not be increased. This eviction has been facilitated by large horsepower tractorization. So the landless population has increased significantly over time in the canal colonies.

In the more concentrated South and West of the country, land market imperfections have led to an inability to evict tenants. This has led large landlords to change the traditional sharecropping contract beyond all recognition. This has reduced the share, income and security of the sharecropper significantly.

NGOs

There is a Clint Eastwood film called 'The Good, the Bad and the Ugly'. Appraisal of Pakistan's poverty alleviation NGOs shows a few good ones, many inefficient bad ones, and mostly corrupt and ugly ones.

The strong points of the good ones are: one, that they target groups, and not individuals. Two, they deliver an integrated package of inputs, rather than meeting single needs like health, or education or food, because single needs all constrain one another.

The weak points are, one, ideological welfare motivation on the behalf of individuals cannot be a long-run factor in eliminating poverty. And two, corruption is a simple proportion of the size of the budget. Presently public programmes are large, so public corruption is high. When NGOs become large, their corruption will become high as well.

So I favour institutional structural change like redistributive land reforms and improvement of factor markets, credit and labour.

Moazam Mahmood

Pakistan Institute of Development Economics, Islamabad.

Comments on "Have Agricultural Economists Neglected Poverty Issues?"

I find myself in broad agreement with what Professor Thiesenhusen has stated despite our somewhat different backgrounds and experiences. We are both structuralists in the sense that any development economist must be a structuralist. But the structural elements that we emphasize are very different. That is because we come from different backgrounds of geographic emphasis. His emphasis and experience has been much more in Latin America whereas mine has been much more in Asia and Africa.

Given my general agreement I would like to make three brief complementary statements about the paper. One has to do with a philosophical issue. The second has to do with the nature of the growth processes that reduce poverty. The third concerns the circumstances in which the poverty reducing growth may not be possible without the kind of structural change that Professor Thiesenhusen emphasizes.

We have an immense moral problem in the world with something like a billion people falling quite clearly under an absolute poverty line that leaves them, in the Rawlsian sense, in an unjust, an unfair, and morally unacceptable circumstance. This is a moral problem of such immensity that its removal must be a basic objective of the development process. However, there is a very troubling aspect to this moral problem. It is here that surface some of the differences in approach, derived from differences in background, between Professor Thiesenhusen and myself.

If we face a moral problem that is to be dealt with primarily by redistribution of assets and income within developing countries then it is not particularly a problem of moral concern to people in the higher income countries. However, if it must be dealt with substantially through growth and through the broad processes of growth which require a very substantial set of incremental resources, such that the capability of providing those resources determines the time path over which they are provided, then it becomes a moral problem of the rich developed countries in which 95 to 100 percent of the population is above that poverty line. I maintain that such growth is essential. Thus, it is very much a moral problem of the rich countries of the world and one that they do not face up to particularly.

Why do the developed countries, the rich countries, not face up to this moral problem? I would assert that the vast bulk of the people in the rich countries of the world do recognize this moral problem and would like to deal with it. Thus, I largely dismiss the first possible explanation. That is that people in rich countries of the world have a nationalistic view of the world, such that poverty in other countries or within other cultural groups, is of less consequence to them than poverty in their own country. Since the rich countries have very little absolute poverty by the kind of definitions that are used in this paper they really do not see much of problem. I think that this is not true. The polls in the United States suggest this not to be the case. For example, 60 percent or more of the public support foreign aid in order to reduce poverty, although much less than a majority support foreign aid per se.

The reason for this discrepancy is the serious problem in the developed countries of a sense of hopelessness about measures to reduce poverty that is akin to the Malthusian view. The Malthusian view is a hopeless view about poverty in the world. It concludes that poverty cannot be eliminated because of processes, in the Malthusian case, of population growth. One sees that represented in the early years of the industrial revolution. We now have a sense of hopelessness about developing countries ever eliminating poverty. That sense of hopelessness is promulgated by the way the poverty issue is presented to the developed country audience. We are constantly told that the number of poor are expanding despite the substantial efforts and substantial foreign assistance in the past. We are constantly having it driven home to us that the poverty problem is a redistribution problem, not a growth problem, and, therefore it is a problem to be solved within the developing countries. So if one is going to have a recognition of absolute poverty in its massive occurrence in poor countries of the world, i.e. the one billion people, and if we are going to have a more rapid dealing with the problem and the elimination of it through major resource transfers in the world, one has to see it not as a hopeless problem but one which eventually will have to be dealt with by self-sustaining growth; a self-sustaining growth which is reached more rapidly by resource transfers.

That brings me to the second set of comments which are about the nature of the growth processes which can play a role in eliminating poverty of an absolute nature over the long run. I constantly emphasize that I am dealing with absolute poverty and not with distribution as represented by the Gini Coefficient. The people concerned with poverty in the foreign assistance community and even to a substantial extent within developing countries themselves virtually never focus on this issue of the details of the rural development strategy which would bring about a rapid diminution in poverty. Till one focusses on that one is dealing with rather a hopeless situation. I have tried to simplify these down to three elements.

First is agricultural technology which requires an immense institutional effort and institutional priorities which we do not often see either in the multi-lateral financial institutions or even within the bulk of the developing countries. Second is provision of rural infrastructure, which requires massive investment, I have made a back of the envelop estimate that in order to provide an acceptable level of rural infrastructure for eliminating poverty will require on the order of 300 billion dollars investment. That is obviously not an investment that can be made in one or even in a few years. Spread over 15 years that comes to $ 20 billion per year. Third, we have similar financial requirements for investments of a similar magnitude to those for infrastructure. Thus, there are very substantial investments to be made. The flow of resources can greatly affect the time path by which one moves along that route.

The third comment that I have to make about this paper deals with the redistributive side, very much emphasized by Professor Thiesenhusen. He is more on the redistributive side, and I somewhat more on the production side. As an aside, when he mentions the views of agricultural economists, I would not include myself. I am an economical agriculturalist which I say only partly in jest. But I am very production oriented. I do want to close on the note, however, that it is perfectly clear that there are countries in which you cannot do the sorts of productivity increasing things that an economical agriculturalist wants done until you have a major redistribution of assets. One of the critical issues to face up to is where is that line? Which are the countries in which we must have a major redistribution of assets to make growth progress. Frankly, I do not find it very helpful to see Bangladesh, where there is hardly a holding of larger than 3-4 hectares, as a country that cannot make progress in rural development until after a major redistribution of rural assets. I can understand that position in an imperial period Ethiopia or in pre-Sandanista Nicaragua. So we have a serious issue of where that line can be drawn so we can talk about the countries where we can make a good deal of progress with sustainable growth in eliminating poverty and the countries where we cannot.

Thank you very much.

John W. Mellor

IFPRI, USA.

REFERENCES

Adelman, Irma (1989) Industrialization and Poverty. Address to the Sri Lanka Economics Association, Colombo.

Barnet, Richard (1990) Defining the Moment. New Yorker 16 July.

Bautista, Germelino M. (1990) The Forestry Crisis in the Philippines: Nature, Causes, and Issues. Developing Economies 28 : 67-94.

Bonilla, Elssy (1990) Working Women in Latin America. Economic and Social Progress in Latin America. Washington, D. C.: Inter-American Development Bank.

Brundtland Commission (World Commission on Environment and Development) (1987) Our Common Future. New York: Oxford University Press.

Burki, Shahid Javed (1988) Poverty in Pakistan. In T. N. Srinivasan and P. K. Bardhan (eds) Rural Poverty in South Asia. New York: Columbia University Press. 69-88.

Chenery, Hollis et al. (1974) Redistribution with Growth. New York: Oxford University Press.

de Janvry, Alain, and Elisabeth Sadoulet (1989) Investment Strategies to Combat Rural Poverty: A Proposal for Latin American. World Development 17 : 1203-1221.

de Tray, Dennis (1989) Reducing Poverty in Developing Countries--Some Basic Principles for Policy Design. The Pakistan Development Review 28 : 4 437-460.

Dixon, John A., and Louise A. Fallon (1989) The Concept of Sustainability: Origins, Extensions, and Usefulness for Policy. Society and Natural Resources 2 : 73-84.

Economist (1989) The Environment: A Survey. 2 September.

FAO (Food and Agriculture Organization) (1988) ACC Task Force on Rural Development. Rural Development. Special Issue, February.

Foy, George and Herman Daly (1989) Allocation, Distribution and Scale as Determinants of Environmental Degradation: Case Studies of Haiti, El Salvador and Costa Rica. Environment Department Working Paper No. 19. Washington, D. C.: World Bank, Policy Planning and Research Staff.

Gaibraith, J. K. (1990) The Rush to Capitalism. New York Review of Books 25 October 51-52.

Hecht, Susanna B. (1985) Environment, Development, and Politics: Capital Accumulation and the Livestock Sector in Eastern Amazonia. World Development 13 : 663-684.

Interpaks Interchange (1990) Fitting Green Revolution to Needs of 'Hungry Poor'. 7:5.

Inter-American Development Bank (1990) Economic and Social Progress in Latin America. Washington, D. C.: IADB.

Keyfitz, Nathan (1989) The Growing Human Population. Scientific American September 118-126.

MacNeill, Jim (1989) Strategies for Sustainable Development. Scientific American September 154-165.

Mahar, Dennis J. (1989) Government Policies and Deforestation in Brazil's Amazon Region. Washington, D. C.: World Bank. (Mimeographed.)

Malik, Muhammad Hussain (1988) Some New Evidence on the Incidence of Poverty in Pakistan. The Pakistan Development Review 27 : 4 509-515.

Martinez-Alier, J. (1989) Poverty as a Cause of Environmental Degradation. Draft. Washington, D. C.: (Mimeographed.)

Mellor, John (1988) Agricultural Development in the Third World: The Food, Development, Foreign Trade Nexus. Paper prepared for the 1988 World Food Conference, 6-8 April.

N'Mi, Kebir (1990) Africa's Troubled Schools [Le Nouvel Afrique Asie]. World Press Review December.

NRC (National Research Council) (1990) Working Group on Population Growth and Economic Development Population Growth and Economic Development. Washington, D. C.: National Research Council.

Painter, Michael (1988) Comanagement with Whom? Conservation and Development in Latin America. Paper presented to symposium, Culture: The Missing Component in Conservation and Development, held in Washington, D. C.: 8-9 April.

Pearce, David (1989) An Economic Perspective on Sustainable Development. Development: The Journal of the Society for International Development 2/3

Pearce, David, Edward Barbier and Anil Markandya (1990) Sustainable Development: Economics and Environment in the Third World. Hants, Eng.: Edward Elgar Publishing.

Postel, Sandra (1984) Protecting Forests. In Lester R. Brown (ed) State of the World 1984. New York: W. W. Norton for the Worldwatch Institute.

Redclift, Michael (1987) Sustaining Development: Exploring the Contradictions. London: Methuen.

Srinivasan, T. N., and Pranab K. Bardhan (eds) (1988) Rural Poverty in South Asia. New York: Columbia University Press.

Stein, Leslie (1989) Third World Poverty, Economic Growth and Income Distribution. Canadian Journal of Development Studies 10 : 225-240.

Sutton, Francis X. (1989) A World to Make: Development in Perspective. Daedalus 118 : 11-17.

Thiesenhusen, William C. (ed) (1989) Searching for Agrarian Reform in Latin America. Boston: Unwin Hyman.

Thiesenhusen, William C., and Jolyne Melmed-Sanjak (1990) Brazil's Agrarian Structure: Changes from 1970 through 1980. World Development 18 : 393-415.

Tisdell, Clem (1988) Sustainable Development: Differing Perspectives of Ecologists and Economics, and Relevance to LDCs. World Development 16: 373-84.

UNICEF (United Nations Children's Fund) (1990) The State of the World's Children. New York: Oxford University Press.

Williams, Aubrey (1990) A Growing Role for NGOs in Development. Finance and Development 27 : 31-33.

World Bank (1990) Worm Development Report 1990. Washington, D. C.: The World Bank.

World Resources Institute (1985) Tropical Forests: A Call for Action, Part 1. Report of the International Task Force convened by the World Resources Institute, the World Bank, and the United Nations Development Programme. Washington, D.C.: WRI.

(1) [MacNeill (1989), p. 156.]

(2) Agenda 2000 (1988).

(3) N'Mi (1990).

(4) [Barnet (1990), p. 60.]

(5) Ibid.

(6) de Janvry and Sadoulet (1989).

(7) [Inter-American Development Bank (1990), p. 26.]

(8) Sutton (1989).

(9) [Interpaks Interchange (1990), p. 5.]

(10) Vol. 8, No. 1 (1990).

(11) World Bank (1990).

(12) [Malik (1988) and de Tray (1989).]

(13) FAO (1988).

(14) [Bonilla (1990), pp. 221-32.]

(15) Srinivasan and Bardhan (1988).

(16) [Galbraith (1990), p. 51.]

(17) Ibid.

(18) Ibid.

(19) Thiesenhusen and Melmed-Sanjak (1990).

(20) [Bonilla (1990), p. 225.]

(21) [Brundtland Commission (1987), p. 28.]

(22) [Ibid., p. 28.] Redclift seems to agree with this definition and also quotes G. Conway's definition: "'Sustainability refers to the system's ability to maintain productivity in the face of a major disturbance, such as that caused by soil erosion, farmer indebtedness, or unanticipated drought or a new pest. The loss of sustainability is then expressed through declining productivity or a sudden collapse in the system" [Redclift (1987), p. 191. The matter is again considered by Clem Tisdell (1988).

Pearce at al. (1990) define sustainability as a vector of development characteristics that is nondecreasing over time, where elements included in the vector (Pearce et al. (1990) include as "elements" increases in real income per capita, improvements in health and nutritional status, educational achievement, access to resources, a "fair distribution of income", increases in basic freedoms) are open to ethical debate, and where "the relevant time horizon for practical decision-making is similarly indeterminate outside of agreement on intergenerational objectives" [Pearce et al. (1990), p. 3]. A key necessary condition for achieving that definition is "non-negative change in the stock of natural resources and environmental quality" [Ibid., p. 4]. As Pearce says elsewhere, "Sustainable development is about being fair to the future. It is about leaving the next generation a similar, or better, resource endowment than that which we inherited" [Pearce (1989), p. 7].

In some contrast, Dixon and Fallon (1989) note that tradeoffs are essential to any definition. One such exchange is regional versus national development such as may be illustrated with a logging programme at Bacuit Bay in the Philippines, which would generate needed foreign exchange but reduce the fish kill and hamper tourism. While claiming that they prefer to protect the local industries because chances of national-level leakages are great, they remind that the more unsustainable alternative, logging, does finance development elsewhere in the country. Other questions are pertinent: Which alternative leaves most resources for the next generation? How far into the future does our concern extend--to next month, the next generation, or the next hundred years? Then there are matters of equity and of how much net social welfare we are ready to sacrifice for, as an example, one species that would be rendered extinct? Are there nonnegotiable alternatives? Dixon and Fallon show that sustainability is not self-evident and that it involves calculations and projections; this may even mean weighing of short-term productivity gains (and survival) against long-term conservation.

The National Academy of Sciences reminds that the rate at which a stock of resources is depleted depends on the rate of population growth, income levels, and, perhaps most important, the success of the price-induced search for more efficient ways to extract and use the resource. "The rate of population growth, in itself, bears no necessary relationship to the rate of depletion. Indeed, the fact that exhaustible resource consumption is highest in economies with high income levels means that the trends in demand for resources in developed countries may be much more important in determining the rate of global resource use than the trends in developing countries" [NRC (1990), p. 15].

(23) [Brundtland Commission (1987), p. 1.]

(24) Martinez-Alier (1989).

(25) [Mahar (1989), pp. 2-3.]

(26) [Keyfitz (1989), p. 120.] In the Philippines, peasants tend to be forced into uplands where damage may ensue. [See Bautista (1990), especially n. 24.] Postel quotes a study by the Indian Institute of Science, which estimated that over half the forest losses in India between 1951 and 1976 were from conversions to agricultural land, where primarily the landless were "encroaching on good forest to bring land under cultivation" [Postel (1984), p. 77]. Some countries, notably Indonesia, clear forested land to encourage the population of more populated islands to move to islands where pressure there is less on resources.

(27) [Martinez-Alier (1989), p. 3.]

(28) [Economist (1989), p. 9.]

(29) [Mahar (1989), p. 46.]

(30) [Martinez-Alier (1989), p. 3.]

(31) [Hecht (1985), p. 679.]

(32) [Mahar (1989), p. 3.]

(33) [World Resources Institute (1985), p. 3.]

(34) Painter (1988).

(35) [Brundtland Commission (1987), p. 141.]

(36) Foy and Daly (1989).

(37) Ibid.

(38) Author's interview with T. Miguel Pressburger, 28 November 1990.

(39) Burki (1988).

(40) Thiesenhusen (1989).

(41) [Adelman (1989), p. 2.]

(42) [Ibid., p. 5.]

(43) [Mellor (1988), p. 2.]

(44) [World Bank (1990), pp. 46-55.]

(45) [World Bank (1990), p. 51.]

(46) [UNICEF (1990), p. 1.]

(47) [Stein (1989), p. 237.]

(48) Chenery et al. (1974).

(49) [Williams (1990), p. 31.]

William C. Thiesenhusen is Professor of Agricultural Economics, University of Wisconsin-Madison, USA.

Table 1
Small Farms * in Brazil: Input and Output, as Percent of Total

 1970 1980

I. Ipput Indicators
 Farms 84 82
 Agricultural Land 15 12
 Cropland 52 39
 Capital Equipment 42 28
 Labour 76 71

II. Output Indicators
 Agricultural Production that is Marketed ** 80
 Farm Net Income 54 50
 Farm Gross Income 46 40

Source : Adapted from William C. Thiesenhusen and Jolyne
Melmed-Sanjak, "Brazil's Agrarian Structure: Changes from 1970
through 1980", World Development 18 (1990), Table 3, p. 398, which
was calculated from Censo Agropecuario, 1970, Tables 9, 27, 31, 34;
Censo Agropecuario, 1980, Tables 18, 29, 32, 33, and 34, and Table
4, p. 399.

* Under 50 Hectares.

** No data.