Land and Tenancy Markets

An overwhelming majority of the world’s poor live in rural areas and depend on agriculture, either as tenant farmers with or without some land of their own or as landless laborers (in agriculture and in rural non-farm activities). Clearly, the inequality in the distribution of land owned and the ability to access land for cultivation as a tenant influence the extent of poverty. The functioning of land and tenancy markets also matters. Indeed, in poor countries the cost of transactions in selling or buying land is often very high. A major reason is the difficulty in establishing a claim of ownership because of the absence of any formal land-titling system or because of poor maintenance of records of land titles and the corruption of officials who have the authority to certify ownership.² High transactions costs could lock in the poor who own small amounts of land, preventing them from selling their land and leaving agriculture to pursue more rewarding opportunities elsewhere. At the same time, those (including some of the poor) who can profitably use more land than they own are deterred from buying land because of the high costs of transactions.

These high transactions costs for sales and purchases of land need not lock in potential sellers and buyers if land tenancy markets functioned efficiently, with low transactions costs. Thus a potential seller, instead of selling land, could rent it out to others, and a potential buyer could rent land from others without having to purchase it. Further, if conditions in tenancy markets preclude long-term tenancy contracts from being entered into, the resulting lack of security of tenure would inhibit productivity-enhancing long-term investments by both tenants and landowners.

Credit, Insurance, and Financial Markets

In addition to being dependent on agriculture, the rural poor have to cope with uncertainties, some of which relate to the environment for production and consumption (such as weather and disease) and others of which are idiosyncratic (health and mortality rates among humans and livestock). Further, the agricultural production process is one in which inputs have to be committed in advance of the realization of an uncertain harvest, whereas consumption is more certain and more evenly paced over time. It is clear that even if production and consumption were free of any risk and uncertainty, the lack of synchronization between the two would still require a way to smooth consumption over time. It is also clear that it would be less expensive to achieve such smoothing if access to smooth and efficient credit markets were available than if each individual had to hold inventories of inputs and consumption goods. The need for credit is enhanced if purchased inputs (such as fertilizers and pesticides, energy and fuels, and hired labor) account for a large share of production costs, as in the case of cultivating Green Revolution varieties of crops. With well-functioning insurance markets, insurable risks would be addressed. However, uninsurable risks (or, more precisely, risks that are insurable only at a high cost) are also significant in the rural areas of poor countries.

For well-known and well-understood reasons of moral hazard, the absence of collateralizable assets, poorly functioning legal systems for enforcement of contracts, and the seizure and sale of whatever collateral has been pledged, formal credit and insurance markets in poor countries are either virtually absent or costly for the poor, if not altogether out of their reach. On the other hand, informal arrangements substitute in part for transactions in formal markets (Townsend, 1994; and Udry, 1993). However, the cost of informal transactions is not necessarily low, and in any case, informal arrangements are nowhere near adequate to substitute fully for the incomplete and imperfect functioning of credit and insurance markets.

Even though the poor do not save enough to invest in financial markets (particularly equity markets), they do invest their meager financial savings in the form of deposits in commercial banks, purchase of life insurance policies, and lending in informal credit markets. Clearly, the returns they realize on such investments depend on the functioning of the financial sector, including the banking system.

Product Markets

The efficient functioning of product markets at home and abroad is vital for poor producers and consumers. Needless to say, the extent of integration of national markets and the competitiveness of exports in world markets depend in large part on whether or not transport and communications infrastructure exists—and functions efficiently—to minimize the costs of transportation and of acquiring market intelligence. Insufficient integration would mean that price differentials across markets exist that cannot be arbitraged away. Also, given the uncertainties not only about harvests but also about the prices that will prevail when the harvested output is to be sold, it matters whether national markets for forward transactions exist and how costly it is to store commodities for later sale. Another issue of concern for small farmers who are mostly poor is the often large difference between the price they receive for their product and the price the ultimate users of the product pay. This difference could reflect possible monopsony power of processors of primary commodities for sale to consumers. It could also reflect high costs of transportation, insurance, and inspection to ensure that the relevant product standards (including sanitary and phytosanitary standards) are met. To what extent each of these contributes to the price difference is not easily determined and, in any case, would differ from commodity to commodity.

Labor Markets

Most often the only asset that the poor have is their own labor. As in the case of commodity markets, the extent of national integration of labor markets is relevant in ensuring that workers receive the best return for their work. Unlike commodities, the cost of whose movement within and between countries is primarily determined by costs of transportation and insurance, the cost of mobility of labor involves social and legal as well as economic barriers. Another inadequately recognized aspect of labor markets in many developing countries (for example, in South Asia) is that only a small part of the labor force (20 percent or less) is in formal wage and salary employment—the overwhelming majority is in self-employment, often in subsistence farming, handicraft activities, and household-based production for local markets. For them it is not so much the functioning of labor markets but of product and credit markets that is more relevant.

Opportunities for and Returns from Accumulation

I have already referred to returns from investment (if any) by the poor in financial instruments. However, a large share (nearly 50 percent or more) of savings and investment by households in developing countries such as India is in the form of physical assets, which they finance on their own without involving financial intermediaries. These assets include mostly those related to their production activities and some dual-use assets (production and consumption). Two points are worth making. First, how large a share of households’ savings is used to finance direct investment in physical assets depends in large part on the functioning of the financial system and households’ access to it, which together influence the cost of financial intermediation. Second, leaving aside investment in housing, investment in physical assets involved in household production could be viewed as a rough analogue of investment financed by retained earnings by enterprises in industrial countries. In other words, considerations of cost of capital, which are extensively discussed in the literature on investment finance in developed countries, are relevant for analyzing the incentives for using one’s own savings to invest in the assets acquired by households in poor countries. Needless to add, these issues are relevant only for those poor households that own household enterprises (farm or nonfarm). Of course, the share of such enterprise-owning households among the poor is likely to be small.

Just as their labor is the major asset owned by the poor, so too their investment in accumulating human capital is likely to be the major component of their investment. Although their poverty limits their saving and investment in any form, it is particularly limiting when it comes to human capital accumulation. Indeed, a major reason that the incidence of child labor is very high in many poor countries of South Asia and sub-Saharan Africa is the poverty of the working children’s parents. Such parents cannot afford to forgo the income from a child’s work (whether from paid work or in terms of unpaid contributions to the household’s farm or nonfarm enterprise). Besides, the out-of-pocket costs to parents for sending their children to school are often substantial.

Thus, for poor parents, both the out-of-pocket and the opportunity costs of investing in human capital accumulation by educating their children are high. Even in the unlikely event that the out-of-pocket costs of sending children to school are negligible (for example, because the state provides quality educational services free of charge), if poverty-induced opportunity costs are high, children are less likely to be sent to school. Three serious consequences arise. First, the earning prospects of uneducated (or less-educated) children in their adult working lives will be reduced compared with their competitors in labor markets. Second, unless labor market conditions improve in their adult life compared with those that prevailed in their childhood, they are likely to end up as poor as their parents were and, as such, unlikely to educate their own children. The prospect of perpetuating poverty across generations in such circumstances cannot be ruled out. Third, since some minimal education is often needed for an individual to participate effectively in the political and social processes that make decisions affecting his or her social and economic prospects, the uneducated will in effect be unable to exercise their right to participate. I should add here that national and international attempts to eradicate child labor through restrictions on imports or consumer boycotts of goods produced by children are not likely to succeed unless the basic cause of child labor is addressed, namely, the parents’ poverty.

Institutions

I have discussed above the effects of the absence or poor functioning of one of the most important institutions, namely, the market. Two other institutions affect the prospects of all agents in the economy: institutions of governance (that is, governments at all levels in each nation as well as multilateral intergovernmental institutions and legal systems) and nongovernmental (so-called civil society) institutions. It is obvious that government policy interventions in the economy are major factors in influencing not only the extent of poverty at a point in time but, even more important, the trends in poverty over time. I will return to government policies in the next section. Endemic corruption often distorts the adoption, enforcement, and effectiveness of policy interventions. Tackling corruption is a major challenge of governance in developing countries. It is a phenomenon that has been with us for ages—in fact, the Arthasastra, a Sanskrit treatise on statecraft dating from the fourth century B.C., lists more than 50 ways in which officials could be corrupt. Of course, the harmful effects of corruption on inflows of foreign investment have attracted attention. Nonetheless, China, a country in which corruption is thriving, has attracted large flows of investment, particularly from overseas Chinese who apparently are better than non-Chinese at operating in the corrupt Chinese system. The use of corrupt practices by transnational enterprises, often with the connivance of governments in their own countries, has received considerable attention in the literature. Other than to reiterate the obvious fact that the poor suffer most from corruption in their societies, I will not discuss it any further.

The recent literature on social capital has paid a great deal of attention to the role of nongovernmental civil society institutions. The Protestant ethic and the rise of capitalism, Confucian ethics, the caste system, the extended family, the place of women in the family and society, communitarianism, individualism, and many other social norms have been analyzed for their economic effects. Interestingly, sometimes the very same institution that was once viewed as inhibiting economic progress is later seen as conducive to it. For example, the institution of the extended family, once considered an obstacle to increasing savings and a source of nepotism, was later judged to save the costs of monitoring nonfamily wage labor, and more generally to reduce the costs of family-based activities in which mutual trust is important.

There can be no doubt that nongovernmental institutions, which in part substitute for absent government institutions, as well as social and religious norms and injunctions that govern individual and household behavior, play a major role in all societies. Some of these constitute social capital. They can enable or frustrate the market mechanism in allocating resources efficiently at a point in time and over time just as much as can the presence or absence of legal mechanisms for defining property and contractual rights and their enforcement. Other than drawing attention to their importance and referring to particular institutions below, I will not discuss them further. In any case the literature offers little guidance as to how to accumulate social capital.

Tenancy Reform

It is unlikely that radical land reforms would come about in poor agrarian societies through nonrevolutionary domestic political processes. However, tenancy reforms are more frequently undertaken. In the Indian state of West Bengal, a left-wing government was elected to power in 1977. It launched “Operation Barga” to implement and enforce the long-dormant agricultural tenancy laws that regulated rents and the security of tenure of sharecroppers. Banerjee, Gertler, and Ghatak (2002) show that Operation Barga explained around 28 percent of the subsequent growth of agricultural production. Shaban’s (1987) study of eight Indian villages and the study of Laffont and Mantoussi (1995) based on Tunisian data show that a shift from sharecropping to either fixed-rent tenancy or owner cultivation raises productivity significantly. Such increases in productivity following tenancy reforms can raise the incomes of the poor.

Raising the Productivity of the Poor

Redistributive land and tenancy reforms are only two reforms among many that could augment the resources commanded by the poor or raise the returns on the assets they own. Others include subsidizing the consumption of the poor (for example, providing a limited ration of essential items of consumption such as food) or their investment in human capital (education subsidies). In practice, subsidy policies work rather poorly: the subsidies are not often well targeted at the poor. Also, the cost of transferring a dollar to the poor through subsidy schemes (particularly poorly targeted ones) often exceeds a dollar by a substantial margin. Hijacking by the nonpoor of subsidies intended for the poor occurs as well. In India there is a public distribution system (PDS) through which fixed amounts per person of food grains and a few other essential commodities are sold at subsidized prices. The system, which is a legacy of urban rationing during the Second World War, remained mainly urban and untargeted for several years before it was extended to rural areas. More recently, targeting has been introduced by limiting the subsidies to the poor. Until the late 1960s, when the Green Revolution took hold, the supplies for the PDS were compulsorily procured from farmers and traders at below-market prices (the so-called procurement prices). As the Green Revolution took hold and market prices declined, procurement operations in effect became price support operations, under which the government stood ready to buy as much as was offered at procurement prices. It should cause no surprise that a farm lobby soon emerged to agitate for every rising procurement price.

The cost of food grains distributed through the PDS included the procurement price plus the cost of transportation and storage by the public sector agency, the Food Cooperation of India (FCI). Again it should cause no surprise that the FCI was grossly inefficient, so that over time the subsidy cost (the difference between the cost and the sale or issue price to the purchasers from the PDS) widened. The economic cost (Ministry of Finance, 2002, Table 5.9) to the FCI of a kilogram of rice was 11.74 rupees, and the price at which it was sold to poor purchasers was Rs 5.65. Nonpoor purchasers paid a higher, but still subsidized, price of Rs 8.30. Major subsidies of the central government alone, of which food subsidies form a large part, are budgeted at 1.6 percent of GDP for the year 2002–03 (Reserve Bank of India, 2002, Table 4.6). As purchases by the PDS did not grow as fast as the volume of grains procured by the FCI, the stocks held by the FCI grew, reaching 60 million metric tons in 2001. This is not the place to discuss in detail the depressing political economy of the PDS in India. Almost all politicians are united in their opposition to dismantling the PDS (ostensibly because it is pro-poor). The farm lobby is interested in ensuring that the PDS stays, so that it provides price support at rising procurement prices. The employees of the FCI (along with their political backers) oppose its privatization or dismantling. Rao and Radhakrishna (1997) analyzed India’s PDS from a national and an international perspective. They found that in 1986–87, PDS and other consumer subsidy programs accounted for only about 3 percent of the expenditure per capita of the poor, and their impact on poverty and the nutritional status of the poor was minimal. Abolition of the PDS would have had a negligible impact on the rural poor, who are the majority of the poor. The central government alone spent more than 4 rupees to transfer 1 rupee to the poor.

Raising the Productivity of the Poor

Raising the productivity or returns to the assets of the poor would alleviate their poverty. As noted earlier, the chief asset of the poor is their labor. Thus a sustained increase in real returns to labor in wage and self-employment would contribute significantly to poverty alleviation. Clearly, if there is an outward shift in the demand for wage labor and for goods and services produced by the self-employed, real returns to labor will increase. Needless to say, such an outward shift is most likely to occur in a rapidly growing economy. Returns to the abundant factor, which in most poor countries is unskilled labor, would rise with trade liberalization. Increasing the human-capital endowments of the poor would also raise the productivity of their labor. This can be done by providing incentives for the poor to invest in human capital on their own, and through public expenditure on, and improving the access of the poor to, public education and health care programs.

Human Capital Accumulation

Public policy interventions could influence directly the investment in human capital by the poor by subsidizing the out-of-pocket costs to the poor of using publicly provided educational and health-care services. As the enormous literature on human capital has emphasized, the fertility decisions of households have an impact on human-capital accumulation decisions through the so-called quality-quantity trade-off. This means that having fewer children would enable a couple to spend more on the education, nutrition, and health of each child. Policy initiatives that combined the provisions of family planning and prenatal health-care services to households, and improved the quality of and access to schooling, successfully brought down total fertility rates and raised the school attainment of children in Bangladesh, particularly poor children. Other programs that provide incentives for parents to keep their children in school (particularly female children), such as midday meal schemes in some states of India and the Progresa program in Mexico, have been successful. Differences in total fertility rates across countries and subnational units within nations have been shown to be lower in those countries and units in which the social status of women is higher, and their educational attainments and labor-force participation rates are higher as well. Given the well-known bias against women and female children within households and in society, any social or state intervention that reduces this bias will have a positive impact on the well-being not only of women but also of female children. Public expenditure on health and sanitation, by reducing the incidence of communicable diseases and raising life expectancy, contributes directly to the productivity of the poor and indirectly to making investment in human capital more attractive. The tragic consequences of the AIDS epidemic on productivity and life expectancy have been dramatic. The poor bear a disproportionate share of the losses.

It is now well recognized that even if governments invest in schools and health-care service centers so that they are well spread, particularly in rural areas where the poor live, such investments may not increase school enrollment or improve the health of the rural poor for many reasons, the foremost among them being the nonnegligible opportunity cost to the poor of accessing the services, even if they are provided free of charge. Although high opportunity costs reduce the demand for these services, serious supply problems often exist as well. Governments often do not anticipate and provide for the operational costs of running schools and health clinics once they are built. This results in fewer teachers being hired than would be needed to provide good-quality services, and less resources being available for purchasing essential school supplies and medicines.

Governance problems frequently compound the resource constraints: these include ensuring that teachers and health care personnel in fact show up at schools. In systems where hiring and firing decisions are made centrally, local authorities and means of services (parents and such) have no say in monitoring the performance of staff. Absenteeism and poor performance of staff can be serious problems. For example, in the state of West Bengal in India, ruled since 1977 by a left-wing coalition headed by the Communist Party of India (Marxist), the government has been powerless to tackle absenteeism by teachers, who are shielded by a powerful union affiliated with the political parties in the ruling coalition. As a result, the educational attainments of children in West Bengal are worse than in some other Indian states that are comparable in terms of real product and expenditure on social sectors per capita. On the other hand, the education-guarantee project in the state of Madhya Pradesh, in which responsibility for teacher hiring and performance monitoring was transferred to local authorities and parent representatives, has made a perceptible impact on enrollments, dropout rates, and other indicators of performance. Failures of the public school system can, and often do, drive parents (particularly poor parents) to send their children to uncertified and poorly monitored private schools, including religious schools in which the time devoted to nonreligious subjects such as arithmetic, science, history and civics, language, and literature is woefully inadequate.

Thus far I have illustrated some possible public-policy interventions that, by design, have poverty alleviation as their objective. These include various subsidies and other mechanisms targeted at the poor. But the fact (if it is a fact in a given context) that such interventions did improve the condition of the poor is in itself only a necessary, and not a sufficient, condition to demonstrate that the return on social and private resources spent (in terms of social and private benefits from such improvements) is comparable to the return from alternative uses of such resources. The example I cited of the Indian government spending over 4 rupees to transfer 1 rupee to the poor through the PDS is unlikely to meet this test. A mechanism for just giving away a rupee to the poor would surely cost less. It is fair to say that it is the exception rather than the rule for a rigorous social-cost-benefit test to be applied either before a program is implemented or afterward. In fact, it is common for governments to refuse to undertake a baseline survey and to build in a mechanism for a proper ex post social-cost-benefit evaluation of an intervention that affects large numbers of people. Interestingly, the Progresa program in Mexico is an exception: areas in which Progresa was to be implemented were deliberately chosen randomly. This enabled a scientifically sound evaluation of its effects. It is understood that few politicians (or even international agencies) would relish the prospect of their pet intervention programs being subjected to a rigorous scientific evaluation. This is perhaps the most likely reason that Progresa is an exception.

Let me now turn to those processes and policies that do not have poverty alleviation as their objective but nonetheless have a major impact on the extent of poverty and its trend over time. Indeed, a large body of evidence shows that these processes and policies have a far greater impact on poverty alleviation than those with a narrow focus on poverty alleviation. I will discuss the four most important of these, all of which are interrelated: openness to foreign trade, capital flows, and technology, as well as domestic market integration; public spending on social and physical infrastructure; macroeconomic stability; and aggregate growth.

Openness

In the standard Heckscher-Ohlin-Samuelson model of international trade, opening an economy to trade or reducing trade barriers raises the return to its most abundant factor and reduces that of its least abundant factor.⁴ Since many of the countries in which most of the world’s poor live, such as China and the countries of South Asia, are abundant in unskilled labor, opening them to trade would raise the return to unskilled labor, a factor that is owned mostly by the poor. Among the scarce factors is capital—lowering the return to capital through trade liberalization reduces the cost of capital. Poor people who do not have capital of their own, but borrow what they use, gain. Thus trade liberalization is a pro-poor policy. Although scarce factors lose from trade liberalization, the gains to abundant factors more than offset the losses to scarce factors, so there are gains to the economy as a whole. In principle, the losers can be compensated since there is a net gain.

This traditional argument about static factor price effects and gains from trade assumes that resources move smoothly and without cost from import-competing to exporting activities. Obviously, if resources cannot or do not move, exporting industries will not expand as import-competing industries contract because of increased competition from imports after trade liberalization, thus creating unemployment. This somewhat extreme but elementary argument against trade liberalization has been raised by Stiglitz (2002, p. 59), who says, “It is easy to destroy jobs, and this is often the immediate impact of trade liberalization, as the inefficient industries [those created under the protectionist walls] close down under pressure from international competition.” Since he assumes that no new, more efficient jobs would be created, he concludes, “moving resources from low-productivity uses [in inefficient industries] to zero productivity [to unemployment] does not enrich any country.” Trite but true! But the lesson is surely not that factors should be kept employed in less-productive uses forever, but rather that, while firmly and credibly committing to removing trade barriers at the end of a reasonably short time, policymakers should remove impediments to labor mobility.

There are also dynamic gains from trade liberalization. Trade is a vehicle through which technical knowledge is exchanged among trading partners. Empirical studies (Coe, Helpman, and Hoffmaister, 1997) suggest that total factor productivity (TFP) in poor countries, which do not have domestic research and development capacities, is higher the greater is their trade with industrialized countries, which account for the bulk of research and development in the world. (This is particularly the case with imports of equipment embodying technical knowledge.) Lastly, openness could raise aggregate growth, at least in the short and medium run, through several channels, including higher investment if the gains from liberalization are not entirely consumed, and greater TFP. Several empirical studies report a strong association between openness and growth (Sachs and Warner, 1995; Dollar and Kraay 2000; and Frankel and Romer, 1999). There are also critics of their findings (Rodriguez and Rodrik, 1999). However, those who find strong empirical support for the positive association between openness and growth, as well as their critics, use the same and, in my view, faulty methodology, namely, the blunderbuss of cross-country growth regressions. Studies in the 1970s and 1980s (Little, Scitovsky, and Scott, 1970; Bhagwati, 1978; Krueger, 1978; and Balassa, 1971), which adopted a more nuanced and robust methodology of studying individual country policies in a comparative framework, provide sounder empirical evidence for the association of growth with openness.

Foreign direct investment (FDI), when it is not merely a response to high tariff barriers (tariff-jumping investment) and tax inducements but is attracted by low labor costs in poor countries (after adjusting for productivity differences) and the prospect of using them as an export platform, can contribute to poverty alleviation. The removal of barriers and controls on financial capital flows in countries with fragile domestic financial sectors has proved to be deleterious to the poor, as the Asian and other recent financial crises demonstrated. Stiglitz (2002) is eloquent on the dangers of liberalizing financial capital inflows. But once again the relevant lesson is not that capital controls are intrinsically and always beneficial. It is that financial liberalization, like trade liberalization, is beneficial if the domestic financial sector is strong and undistorted. Therefore, policymakers, while credibly and firmly committing to liberalize capital flows at a specific future date, should use the time until then to reform the domestic financial sector and create an appropriate prudential regulatory framework.

Domestic Market Integration

In some developing countries, particularly large ones such as India, there are restrictions, including taxes, on internal trade. In India, there are interstate sales taxes on some commodities and restrictions on movement of food grains on private account even between districts within states. The Essential Commodities Act (once again a vestige of controls imposed during the Second World War) gives the government broad power to intervene in internal trade in such commodities. Price controls are also quite frequently imposed. Panterritorial pricing, which requires that the same price be charged throughout the country, is another restriction that has been imposed in some African countries and in India. Clearly, such a requirement thwarts the forces of regional and local comparative advantage from operating and in effect is a means for cross-subsidization of producers in regions that are more distant from the markets—and less efficient—by producers in closer and more efficient regions. Producer price controls are likely to affect poor and small producers adversely compared with larger producers. Attempts to stabilize prices received by commodity producers and to exploit market power in world markets through marketing boards failed miserably. Their inefficiency and adverse distributional effects in Africa and elsewhere are well known. More generally, the less the integration of domestic markets, the greater is the inefficiency of domestic resource allocation. In all probability, the poor bear a disproportionate share of the cost of such inefficiency.

I should also mention production controls, such as the “small-scale industry reservation” in India, which reserved certain labor-intensive products for production exclusively by small-scale producers. Although prima facie this might appear to confer an advantage on poorer producers, in fact it has operated as a tax on efficiency and prevented India becoming internationally competitive in such products. Other forms of production and export controls are prevalent in developing countries and almost always hurt the poor.

Infrastructure Spending

I have already discussed the possible pro-poor impact of public spending on social sectors such as education and health. In many poor countries, economic infrastructure projects (irrigation and flood control works, power, transport and communications, and ports) are largely publicly owned and operated, often inefficiently and at a high cost. Although privatization (accompanied or preceded by the creation of regulatory agencies) is the appropriate remedial policy in many areas, there are situations in which it is not realistic. In such situations, public spending on infrastructure could contribute to poverty reduction and growth. In another study, Fan, Hazell, and Thoret (1999) analyze the more or less steady decline in rural poverty in India since the late 1970s. They use state-level data to test an econometric model that allows the estimation of the marginal contribution of different categories of public infrastructure expenditure on poverty reduction. They find that the marginal benefit in terms of poverty reduction was the highest for expenditure on roads—such expenditure also contributed significantly to productivity growth. Expenditure on agricultural research and extension yielded the largest marginal benefit in terms of productivity growth, while also yielding significant benefits in terms of poverty reduction. Expenditure on education had the largest impact on poverty reduction, largely because of the increases in nonfarm employment and rural wages it induced. By contrast, investment in irrigation had only a small impact on poverty but had the third-largest impact on productivity growth. Lastly, public spending on rural and community development, including the integrated rural development program, reduced poverty, but to a smaller extent than did expenditure on roads, agricultural research and development, and education. The authors’ explanation for this difference is that, although spending on rural development is effective in reducing poverty in the short run, since it has little impact on agricultural productivity, it contributes little to sustained and long-term poverty reduction. This confirms that although employment guarantees, food-for-work, and other such programs certainly alleviate poverty and avoid problems of leakage of benefits to the nonpoor because of their inherent self-selection feature (that is, only the poor avail themselves of them), unless the program activities in which the poor are employed yield productivity gains, their impact on poverty reduction in the long run will be minimal.

Macroeconomic Stability

An unstable macroeconomy, especially an inflationary environment, is particularly damaging to the poor. In some countries of Latin America, wage labor is the dominant form of employment, and wages are indexed to inflation. But in other parts of the world where self-employment and casual work are dominant, there is no indexation of earnings. In such contexts, inflation is the cruelest tax—it hurts the poor most because their earnings are not indexed to inflation and they have no opportunities to invest in assets that provide hedges against inflation. This is not to say that indexing (particularly backward indexing) is the appropriate policy to mitigate the ravages of inflation. It is not appropriate because, among other things, it can perpetuate inflation. The right policy is to avoid inflation and to have a stable macroeconomic environment. There are other channels through which macroeconomic instability hurts the poor, two of them being lower aggregate growth and greater risk of financial collapse.

Growth

The instrumental roles of rapid aggregate growth and, secondarily, of a more equitable distribution of the fruits of growth in eradicating poverty were recognized long ago by policymakers in developing countries. For example, in 1940, before Indian independence, a national planning committee chaired by the future prime minister, Jawaharlal Nehru, completed its work on elaborating a development strategy for an independent India. The overarching objective of the strategy

… was to insure an adequate standard of living for the masses; in other words, to get rid of the appalling poverty of the people … the irreducible, in terms of money, had been estimated by economists at figures varying from Rs 15 to Rs 25 per capita, per month (at prewar prices) … [To] insure an irreducible minimum standard for everybody, the national income had to be greatly increased, and in addition to this increased production there had to be a more equitable distribution of wealth. We calculated that a really progressive standard of living would necessitate the increase of the national wealth by 500 or 600 percent. That was, however, too big a jump for us, and we aimed at a 200 to 300 percent increase within ten years. (Nehru, 1946, pp. 402–03)

India embarked on planning for national development in 1950. As early as 1960, when two five-year plans had been implemented, a socialist member of parliament questioned whether the fruits of growth in the two plans were reaching the poor. In his response, Nehru said:

Again it is said that the national incomes over the First and Second Plans have gone up by 42 percent and income per capita by 20 percent. Now a legitimate query is made: where has this gone? To some extent, of course, you can see where it has gone. I sometimes do address a large gathering in the villages, and I can see that they are better fed and better clothed, they build brick houses and they are generally better off. Nevertheless, that does not apply to everybody in India. (Government of India, 1964, p. 1)

Nehru followed up his response with the announcement of a committee to inquire into the trends in income distribution and standards of living. Even as the committee was engaged in its task, the Perspective Planning Division of India’s Planning Commission produced a 15-year development plan for poverty eradication covering the period 1960–61 to 1975–76, with a growth target of 7 percent a year. What is more interesting is that the plan explicitly recognized that there were individuals and households in the economy who, for various idiosyncratic and locational reasons, were weakly connected to the income generation processes of the economy, and that their poverty could not be alleviated simply by more-rapid growth in incomes. It advocated a two-pronged strategy for poverty alleviation: for the large majority of the poor, who are well connected to the production and income-generation activities of the economy, rapid income growth would eradicate poverty. For the weakly connected poor, transfers would be necessary.

More-recent empirical evidence (World Bank, 2001) suggests a significant association between aggregate growth performance and improvements in human development indicators, including the incidence of poverty. However, any association between aggregate growth and reduction in national poverty does not imply either a one-way causal relation between growth and poverty or even that the association would hold for all countries and for all periods. One has to identify the possible mechanisms through which, on the one hand, aggregate growth could affect, positively or negatively, poverty at the national or sub-national level and, on the other hand, how levels and trends in poverty could influence growth, again in either direction. Clearly, there is no reason to presume that all such mechanisms need operate everywhere and at all times or, even if they do, that they operate with the same intensity. Further, there may be leads and lags involved in their operation—for example, it could take several years before an acceleration in growth results in poverty reduction. In addition, only sustained increases in growth, and not any temporary and reversible increases, could bring about a reduction in poverty. Moreover, growth, poverty, and inequality are endogenous outcomes of economic, social, and political processes within a nation as well as of trends in the world economic and political environment insofar as it affects domestic processes. These processes themselves could in part be endogenous and interact with each other. The speed, strengths, and nature of interaction could and would vary over time and across countries. For all these reasons it is not easy to dismiss conclusively and convincingly through empirical analysis such statements as, “Aggregate growth is not enough” or “No poverty reduction is possible without aggregate growth.” Both could be seen as true and false, in the sense of being seemingly valid for some countries or periods and not valid for other countries or periods.

Thus, no universal (valid for all countries) and eternal (valid for all periods) causal relation between growth and poverty reduction can be derived from economic theory, and hence one cannot hope to find it empirically. Yet it turns out that poverty declined substantially in the last two decades of the twentieth century. This is evident in the greater integration of the developing economies with the world economy (through the process of globalization) and more-rapid growth in many of them. This is particularly true since the 1980s of the world’s two largest developing countries, China and India. The so-called miracle economies of East Asia adopted an outward-oriented development strategy much earlier, in the mid-1960s; achieved rapid growth; and virtually eliminated poverty until the financial crisis of 1997 temporarily increased poverty. I turn briefly in the next section to their performance.

East Asia

Since the mid-1960s—when they adopted an outward-oriented development at a time almost every other developing country was inward-oriented—the East Asian economies (Hong Kong SAR, Korea, Singapore, and Taiwan Province of China) achieved spectacularly high rates of growth and succeeded in achieving an economic and social transformation that had taken the industrialized economies of Europe centuries to accomplish. Their Southeast Asian neighbors—Indonesia, Malaysia, and Thailand—also achieved high rates of growth and rapid reductions in poverty. Although the financial crisis of 1997–98 did reverse their attainments (particularly in employment generation and poverty reduction), the quality of life of the populace remains largely intact and, with the exception of Indonesia, growth has resumed, although not at precrisis rates.

The success of East Asia has been studied extensively by many scholars. Nobel laureate Robert Lucas (1993) uses the success of these countries as a background for developing a theoretical framework in which the two distinguishing characteristics of East Asia (compared with South Asia)—namely, outward orientation and initial human capital endowments (and subsequent accumulation)—play a crucial role in generating sustained and rapid growth. The World Bank (1993) published its study of East Asia in the same year. Quibria’s study (2001) is one of the most recent attempts to draw lessons from the East Asian miracle.

Quibria correctly points out that far too many, and often conflicting, lessons have been drawn by different scholars from ostensibly the same set of facts. Studies by the World Bank (1993), Leipziger and Thomas (1997), and Yusuf (2001) do not draw identical lessons, and even if some lessons are the same, their ranking in terms of importance differs. Quibria’s revisit to the East Asian miracle uses some new evidence and revises, if not reverses, some of the earlier lessons.

The World Bank (1993), for example, emphasized the importance of two “facts”—that initial endowments of human capital were favorable and that the fruits of growth were well distributed among all segments of the population—in contributing to East Asia’s success. Quibria’s reading of the evidence raises doubts about these “facts.” The reductions, during the 1970s and 1980s, in inequality in Hong Kong SAR and Malaysia were modest and were from high levels by international standards. In Thailand inequality increased, and it remained constant in Singapore and in Taiwan Province of China. Quibria finds that the claim that these economies had favorable initial human capital endowments or an egalitarian land distribution following land reforms has no solid basis. In his judgment the main factor in explaining their success in poverty reduction is not any radical improvement in income distribution, but rather rapid economic growth. This growth in turn was founded on high rates of investment, sustained by a congenial investment climate. The congeniality of the climate was due to the market orientation of these economies, supported by a policy package whose critical element was openness to trade and technology. Since openness, at its core, creates opportunities, it can succeed only if the opportunities are taken. For this to happen, a set of complementary, credible, and stable policies that ensured a sound and stable macroeconomic environment, a flexible labor market, and incentives that favored productive rather than rent-seeking activities is needed. The East Asian economies had these, and their authoritarian regimes and efficient bureaucracies, insulated from politics, facilitated their adoption.

China and India

Quibria contrasts East Asia with South Asia and finds the latter deficient in many respects. However, even South Asia, and particularly its largest economy, India, performed well in the last two decades. Of course, the star performer of the two decades was the People’s Republic of China. According to the World Bank (2002), both countries enjoyed historically unprecedented average rates of growth of GDP at around 10 percent and 6 percent a year, respectively, during 1980–2000. Fewer than 10 of over 200 countries covered by the study exceeded India’s growth rate, and none exceeded China’s. Poverty in both countries declined substantially during the two decades of rapid growth, albeit at different rates.

In India, annual poverty estimates based on household expenditure surveys are available from the 1950s on. These suggest that poverty as measured by the head-count ratio fluctuated around a level of 50 percent without any downward trend through 1977–78, when it was 50.5 percent and 40.5 percent in rural and urban areas, respectively. From then on there was a decline (Table 1 in Srinivasan, 2001b). Deaton’s (2001b) estimates cited earlier show that the poverty ratio fell to 25.3 percent in 1999–2000 from 39 percent in 1987–88 in rural areas, and to 9.5 percent from 22.8 percent in urban areas.

Unlike the Indian data, Chinese poverty data based on household surveys are relatively recent—official poverty lines and poverty head counts going back to 1978 were first announced in 1994, superseding some earlier ad hoc estimates. These official data show that rural poverty has been virtually eliminated, falling from 30.7 percent in 1979 to 9.5 percent in 1990 and to 4.6 percent in 1998 (Park and Wang, 2001). A World Bank estimate quoted by the same authors, on the other hand, put rural poverty in 1990 nearly four times as high, at 42.8 percent; and it then fell to 24.2 percent in 1997. Although the estimated levels of poverty differ substantially between official and World Bank estimates, the trend is similar—a halving of poverty between 1990 and 2000 (Srinivasan, 2002, Tables I and II). An analysis of the factors behind the decline in poverty and the differences between the Chinese and Indian experiences is useful in understanding the interactions among openness, growth, and poverty reduction and the role of development strategies in these.

Let me begin with a bit of economic history. Maddison’s (2003) historical analysis suggests that China and India had the same real income per capita in 1870. But by 1950, when the Communist regime took over, China’s income per capita had declined by 17 percent while India’s had increased by 16 percent. It took nearly two-and-a-half decades, that is, from 1950 to 1973, for China to recover the lost ground, with double India’s rate of growth of income per capita. It is reasonable to presume that China and India again were at roughly the same level of income per capita in 1980, two years after Deng Xiaoping abandoned the Maoist economic strategy that had led to the death of 30 million people or more and initiated systemic reforms. India’s liberalization began in the 1980s, but systemic reforms came only after the macro-economic crisis of 1991. However, although both economies experienced an acceleration in growth during 1980–2000 compared with the previous three decades, China’s average growth rate of income per capita, at nearly 9 percent a year, far exceeded India’s 4 percent a year, so that China’s income per capita was nearly 70 percent higher than India’s by 2000. The differences in the rate of decline in poverty (85 percent in China between 1978 and 1998 and 50 percent in India between 1977–78 and 1999–2000) seem consistent with faster growth in China’s income per capita (Srinivasan, 2002, Tables IIIA and IIIB).

It is most likely that the differences in the reform and growth processes of the two countries also contributed to the differences in growth rates and their impact on poverty outcomes. Chinese growth was faster and seems to have been more pro-poor. Not only did China continue to save and invest a far higher proportion of its GDP than India, but its integration with the world economy became far deeper as well. Its share of merchandise trade in GDP, albeit an imperfect proxy for global integration, rose to nearly 44 percent in 2000 from approximately 13 percent in 1980. In India the share fluctuated around an average of 12 percent until the opening of the 1990s and has since risen to 20 percent. Another aspect of global integration, namely, inflow of FDI, also showed similar differences: the ratio of FDI to GDP in China grew from virtually zero when the reforms began in 1978 to 4.3 percent in 2000. In India, even after a decade of reforms, the ratio is only 0.6 percent (Srinivasan, 2002).

The sequencing of reforms was also somewhat different in the two countries. China reformed its agriculture first by abolishing collectives, introducing the household responsibility system, and reducing mandatory deliveries of output to the state by farmers, thereby enabling farmers to produce for the market. India’s agriculture, while always in the private sector, was insulated from world markets and riddled with government interventions in the domestic market for agricultural inputs and outputs, whose net effect was adverse to agriculture. The Indian reform process is still to be extended to agriculture. Clearly, the fact that China reformed agriculture first, and achieved spectacular results for several years, not only provided credibility to its reform process but also increased the incomes of the poorer segments of the Chinese economy. In India, agriculture in particular and the rural economy in general have yet to be reformed systemically. Until that happens, not much acceleration in reducing the rate of rural poverty can be expected.

Although India and China reformed their external sector by reducing tariff and nontariff barriers, as noted earlier, China’s opening went much deeper. In part this was because, while opening the special and coastal economic zones for foreign investment, China in effect allowed foreign investors 100 percent ownership and the freedom to hire and fire workers and provided an excellent infrastructure. In all these respects India lagged behind. Further, India’s reservation (until very recently) of labor-intensive products such as garments, leather products, and others to small-scale industry prevented the full exploitation of opportunities from its opening. China increased its share of world exports of labor-intensive products, while India struggled to maintain its slow development and in fact lost ground in some products. India’s shallower integration and failure, owing to lack of domestic policy reform, to take advantage of the opportunities in the world market limited not only the growth-enhancing impact of trade reform but also, and more importantly, its poverty-reducing effect.

In at least two other respects, the Chinese and Indian reforms differed significantly. These differences favored China both with respect to growth and with respect to poverty alleviation. The first difference is in their approaches to the reform of state-owned enterprises. The share of investment in these enterprises continues to be high in both economies (about 30 percent in India and two-thirds in China in 2000), and their employment has not fallen significantly despite the fall in their share of total output. These similarities notwithstanding, there is no Indian counterpart to China’s dynamic township and village enterprises, which by all accounts were labor-intensive and provided employment opportunities to the poor. The second area in which India lagged and continues to lag behind China is in the availability of reliable and affordable infrastructure, particularly electric power. Whereas China has succeeded in attracting foreign investment into this vital sector, India has failed to do so.