The Inter-American

Development Bank

and Poverty Reduction:

An Overview

Nora Lustig

and

Ruthanne Deutsch

Washington, D.C.

March 1998--No. POV-101

Nora Lustig is Senior Advisor and Chief, and Ruthanne Deutsch is an Economist in the Poverty and Inequality Advisory Unit (SDS/POV) of the Sustainable Development Department.

The authors are grateful to Marcia Aireira (SDS/POV) who prepared the discussion of PTI performance and also contributed to the discussion of IDB lending highlights. Inputs were also provided by Luis Fierro and Martin Chrisney (SDS/SDS) based on their work for the document A Path Out of Poverty, currently being prepared for the Summit of the Americas. Additional inputs were provided by Andrew Morrison (SDS/ WID), Hege Gulli (SDS/MIC), and Héctor Eduardo Rojas (SDS/SOC). Thanks are also due to Peter Hakim of the Inter-American Dialogue for his insightful comments. The excellent research assistance of César Bouillon (SDS/POV) and the collaboration of José Antonio Mejía (SDS/POV) on the poverty and inequality estimates are greatly appreciated. Janet Herrlinger (SDS/POV) and Norma García (SDS/POV) assisted in the preparation of this document.

The views expressed herein are those of the authors and do not necessarily represent the official position of the Inter-American Development Bank.

Preface

This document provides an overview of how lending and non-lending activities of the Inter-American Development Bank can contribute to reducing poverty in Latin America and the Caribbean, and also offers suggestions for future actions that the Bank can take in this area. The first half of the document provides the empirical and analytical backdrop from which the IDB's performance in contributing to poverty reduction can be assessed. An initial discussion of poverty trends in the 1990s establishes that despite moderate improvements, poverty remains a persistent problem in the region, and that income inequality has failed to decline. Given that economic growth alone is not enough, a summary is presented of the array of directed policy actions that can help to reduce poverty. The second half of the document appraises the many ways in which the IDB can foster poverty reduction in Latin America and the Caribbean. The document concludes by looking forward to areas that merit increased attention and can improve the Bank's ability to succeed in reducing poverty in the region.

Contents

Introduction 1

Part I: Empirical and Analytical Backdrop of Poverty and

Economic Growth in Latin America and the Caribbean 2

Poverty Trends in the 1990s 2

Economic Growth: Essential but Not Enough 4

Recent Patterns in Employment Generation and Wage

Disparities: Not Exactly What We Want 5

Policy Actions for Poverty Reduction 7

Part II: The Inter-American Development Bank and

Poverty Reduction 10

Bank Tools and Performance in Poverty Reduction 10

The Overarching Challenge: Promoting Stable and

Sustained Economic Growth 11

Lending Highlights of Projects With Benefits to the Poor 12

Non-lending Actions 22

Areas for Future Action 25

References 30

Data Appendix 33

Annex 1: Poverty Studies Used in the Paper 44

Annex 2: Poverty Lines 47

Annex 3: Extreme Poverty Lines 48

Introduction

Part II of the document appraises the many ways in which the IDB can contribute to reducing poverty in Latin America and the Caribbean. First, the Bank's mandate, with its overarching goal of poverty reduction, is reviewed. Subsequently, a brief description of the instruments which the IDB can use to contribute to the fight against poverty is provided. A summary discussion of how the Bank works to promote stable and sustained growth in the region--the backbone of any anti-poverty strategy-- is then followed by an overview of recently approved projects, or projects in the pipeline, representing some of the Bank's best efforts in financing actions that can foster poverty reduction. In addition to this discussion of lending highlights, the document describes the many non-lending activities which also contribute to the Bank's efforts in this area. The document concludes by looking forwards to areas that merit increased attention and can improve the Bank's ability to contribute to poverty reduction in Latin America and the Caribbean.

Part I

Empirical and Analytical Backdrop of Poverty and Economic Growth

in Latin America and the Caribbean

POVERTY TRENDS IN THE 1990s

It can be inferred that approximately 150 million people in Latin America and the Caribbean, or one out of every three individuals, are currently living on incomes that are less than US$2 a day,⁽¹⁾ although there are several countries for which recent data does not exist.⁽²⁾ At least twelve Latin American countries now have reliable poverty studies with information as recent as 1994 or subsequent years.⁽³⁾ Of those, nine have information for both urban and rural areas, while three (Argentina, Bolivia and Paraguay) have information for metropolitan or urban areas only, and some have no data at all.

Between the late 1980s and the mid-1990s, the incidence of poverty has fallen in the majority of countries.⁽⁴⁾ The exceptions are Venezuela and the Greater Buenos Aires area of Argentina, where the incidence of poverty rose; and, Mexico, where it remained unchanged. To what extent are the observed trends in poverty due to changes in per capita income or to changes in the distribution of income? Almost universally, the declines in poverty were accompanied by positive growth rates. In more than half of the episodes, the trends in the distribution of income were neutral (that is, there was no change) or countered the positive impact of growth on poverty reduction (that is, income became more concentrated). The high, and in some countries growing, inequality characteristic of the region has tempered the benefits from economic growth accruing to the poor. Likewise, rising inequality has exacerbated the impact of low or declining growth rates on poverty rates, for example, in Greater Buenos Aires and Venezuela.

It is important to stress that any exercise which just compares aggregate figures for two points in time hides important facts. For those countries where frequent surveys exist, poverty estimates often change sharply from one year to the next, closely mimicking macroeconomic performance. Adverse shocks, fiscal retrenchment and high inflation result in rising poverty. Years of bonanza and of rapid decline in inflation are accompanied by reductions in poverty.

In the Greater Buenos Aires area, for instance, poverty increased sharply in 1989-90 (the years of hyperinflation) fell in the subsequent years, and as a result of the crisis which followed the Mexican devaluation, the headcount ratio rose from 13 percent to 20.2 percent between 1994 and 1996. In Brazil, poverty rose between 1989 and 1992 (the years of hyperinflation) remained the same between 1992 and 1994, and fell in 1995 as inflation fell sharply. In Costa Rica, poverty rose by five percentage points from 1990 to 1991 when average growth slowed, and fell by 12 percentage points between 1991 and 1994, when growth was vigorous. In Venezuela, one of the most volatile economies in the region, poverty decreased by 10 percentage points between 1989 and 1991, rose by almost 20 percentage points between 1991 and 1994, fell in 1995 and rose again in 1996. The implications of these observed patterns are that poverty rates can be quite volatile, and that adverse shocks can have devastating effects on the poor and the near poor. Also, more often than not, the upturns do not return the poor to previous income levels. As discussed further below, not only growth but steady and sustained growth is the key to poverty reduction.

The second fact which remains unobserved when using aggregate figures is that trends may not be homogenous among regions and among age, occupation, ethnic and gender groups, and, also, between the moderate and extreme poor. To give some examples, in Colombia, while urban poverty fell sharply between 1978 and 1995, rural poverty increased; in Chile, moderate poverty declined between 1992 and 1994, but the real average monthly per capita income of the bottom ten percent of the population declined from 13.8 pesos to 13.5 pesos; in Mexico, while country-wide extreme poverty fell slightly between 1989 and 1994, it rose by significantly in the South and Southeast. Acknowledging and understanding the causes of these differences is important for policy design.

Who are the poor? Where do they live? What do they do? Are certain regions or groups of society more prone to being poor? Poverty profiles are essential for effective policy action. Despite important differences among countries in terms of per capita income and socioeconomic characteristics, common patterns emerge. One universal trait is that the poor tend to live in households whose head has little or no education, larger households and households with higher dependency ratios. There is no systematic pattern between poverty and female-headed households: in a few countries a higher proportion of the households living in poverty are headed by women, but this is not the case in the majority of countries. The proportion of indigenous groups and blacks is higher among the poor, and poor households are more likely to live in rural areas. The poor tend to live in households whose head is more likely to work in the informal sector and in primary activities or services, than the non-poor. Although the majority of the poor live in households whose head is working, the proportion of heads who are unemployed is higher among the poor than the non-poor (see data appendix Tables 3 and 4).

The likelihood of being poor is thus higher for the less educated, those living in rural areas, and those working in primary, service, and construction activities, and in the informal sector. The same is true for the unemployed. Indigenous groups and blacks are more likely to be poor than whites. Also, working women tend to have a higher probability of being poor than working men even after controlling for education and other variables. No systematic pattern is observed with respect to the gender of the head; in only one out of the 10 countries in which this was analyzed (Costa Rica), female headed households had a significantly higher incidence of poverty. In terms of age, children and young workers are more likely to be poor. As for the age of the head of the household, in several countries of the region, the incidence of poverty was higher for those aged 60 or older. Because so many of the poor in the region are peasants or self-employed, the likelihood of being poor among non wage earners or independent workers is higher than for wage earners in some of the countries (for example in Mexico and Nicaragua).

Where are the poor most concentrated? While it is true that in several countries more than fifty percent of the poor live in rural areas, the share of the poor living in urban areas is far from trivial. In countries such as Brazil and Chile the urban poor constitute the largest portion of the poor (see data appendix Table 4). Households living in extreme poverty, however, appear to be more concentrated in rural areas but, again, this is not the case for all countries.

ECONOMIC GROWTH:

ESSENTIAL BUT NOT ENOUGH

We now know that economic growth is essential for poverty reduction. Empirical estimates find that a one percent annual rate of growth in consumption per person results in a 1.5 to 4 percent decline in poverty.⁽⁵⁾ In the 1990s, growth resumed in the majority of the countries in the region and the prospects--although in the short-run less bright than before the Asian financial crisis--are fair. The greatest accomplishment is that for the first time in many decades, the countries in the region are enjoying considerably lower inflation rates, and that hyperinflation seems to have disappeared.

Although there are a number of policies which are conducive to growth, there are elements to the process of economic progress which remain elusive. Not all countries can be expected to grow at rates which would ensure the elimination of extreme poverty within a reasonable period of time: let's say, in fifteen or twenty years. Furthermore, growth alone may not be enough. If income distribution remains unchanged, the pace of poverty reduction might be quite slow even with sustained growth. For example, at yearly growth rates of three percent per capita, it could take close to 50 years to over a century, depending on the country, to completely eradicate poverty as measured by the proportion of individuals living below US$2 per day. The time span would be longer if we used country specific poverty lines, if growth rates were lower, or if income distribution worsened. At recent growth rates and country-specific poverty lines, there are cases for which it could take four hundred years or more (!) to eradicate extreme poverty.

Growth does not guarantee that everybody benefits. In Chile, for example, despite the unquestionable success in economic performance and overall reduction in poverty, the mean household income (in 1994 pesos) per equivalent adult for the bottom decile of the population declined between 1992 and 1994 from 13,800 to 13,500 pesos. In Mexico, although extreme poverty fell slightly between 1989 and 1994, the incidence of poverty in the South and Southeastern regions rose. In Colombia, urban poverty fell but rural poverty rose between 1978 and 1995.

Furthermore, growth does not guarantee that poverty will fall at all. Most of the Latin American countries analyzed in this study present an inverse relationship between the incidence of poverty and economic growth. But there are a few exceptions to this. In Argentina and Mexico income growth was accompanied by almost no decrease in poverty. Per capita GDP in Argentina grew strongly between 1991 and 1994, but poverty remained nearly constant. Mexico's per capita GDP grew slightly between 1992 and 1994, but poverty remained almost unchanged. In Venezuela, per capita GDP growth was accompanied by increases in poverty between 1991 and 1992. Finally, there has been a steady decline in urban poverty in Paraguay (Asuncion), even though per capita GDP fell slightly in 1991, 1992 and 1995.

Finally, growth can be unstable. Countries are likely to face occasional adverse shocks due to policy mistakes, volatility in the capital markets, sharp turns in commodity prices, natural disasters, and so on. As we saw above, volatility in growth rates translates into sharp turns in poverty rates. In fact, the sharpest increases in poverty in the region have been the result of adverse shocks caused by economic, political or natural factors. So, not just growth, but steady growth is important for poverty reduction.

RECENT PATTERNS

IN EMPLOYMENT GENERATION

AND WAGE DISPARITIES:

NOT EXACTLY WHAT WE WANT

In order to speed up the process of poverty reduction, growth should be intensive in the region's abundant factor: unskilled labor. For any given growth rate, such a pattern would yield lower poverty rates. However, for reasons which are not yet fully understood, recent growth episodes in a number of countries have been accompanied by high unemployment rates, and the wage disparity between the skilled and the unskilled has been on the rise.

Depending on the country, we can observe relatively high levels of open unemployment and/or an expansion of underemployment and the informal sector. In 1997, open unemployment in Latin America⁽⁶⁾ was estimated at 7.5 percent; in 1991, the unemployment rate was 5.8 percent. In 1997, in 12 out of 22 countries for which data on open unemployment is regularly reported, the rate was close or above 10 percent (Argentina, Barbados, Colombia, Ecuador, Jamaica⁽⁷⁾, Nicaragua, Panama, Peru, Dominican Republic, Trinidad and Tobago, Uruguay, and Venezuela). In 10 out of the 22 countries, open unemployment was higher in 1997 than in 1991, and in 11 of the 22, open unemployment has increased since 1994 (see data appendix Table 5). Informal employment (own-account workers, unpaid family workers and workers employed in microenterprises and in domestic service) rose substantially during the 1990s in the region. According to the International Labor Organization (ILO), informal employment rose from 51.6 percent of the total working population in 1990 to 56 percent in 1995.⁽⁸⁾

Open unemployment is often disproportionately present among the poor. For example, in the Greater Buenos Aires area of Argentina in 1996, the proportion of unemployed workers was 29.8 percent for the poor compared to 4.5 percent for the non-poor. In Chile, in 1994 the proportion of male unemployed workers in the lowest decile of the population was 21.1 percent, versus an average of 4.8 percent for the remaining nine deciles. For Chilean female workers the average unemployment rate of the bottom ten percent of the population was 36.8 percent, while the average for the rest was 11.1 percent. In Costa Rica, in 1996, the percentage of unemployed among the poor was 3.6 percent, versus 1.9 percent for the total. These unemployed workers are unlikely to benefit from formal mechanisms of social protection (severance payments and unemployment insurance, for example) and are likely to be more prone to engaging in illegal activities, particularly because unemployment is more common among the young. Regardless of whether the poor are more likely to be unemployed or vice-versa, both phenomena are clearly associated.

Although average real wages have been rising in a number of countries (see data appendix Table 6), not all wage earners enjoyed similar increases. During the last decade the wage gap between the skilled and the unskilled increased sharply in a number of countries, and in several the real wages of the unskilled actually fell. According to ECLAC⁽⁹⁾ the difference between the incomes of professional and technical personnel and workers in low productivity sectors rose between 40 and 60 percent between 1990 and 1994. During this period, real labor income for highly skilled workers employed in modern sectors of the economy rose substantially in the majority of the countries (except in Bolivia, Brazil, and Honduras in which all incomes fell). For unskilled workers employed in low productivity sectors, real wage increases were moderate or even negative. More specifically, the gap between the income of the skilled and the unskilled widened in eight out of ten Latin American countries⁽¹⁰⁾ between the end of the 1980s-beginning of the 1990s and 1994. For example, in Mexico, the ratio of average income of high skilled workers to unskilled workers (professional and technical workers to workers in low productivity sectors) increased from 138 percent in 1989 to 203 percent in 1994; in Chile, it grew from 169 percent in 1990, to 205 percent in 1994; and in Brazil, it rose from 226 percent in 1987 to 326 percent in 1993.⁽¹¹⁾

These trends come as a somewhat unpleasant surprise. At least among economists, the expectations were that the elimination of trade barriers and structural reform in general would stimulate labor-intensive growth, that is, a growth process with rapidly rising employment opportunities particularly for the less-skilled. Some of the factors behind the observed trends are transitory or could be corrected by eliminating some of the laws which regulate labor markets. However, other causes behind these trends could be more durable, imbedded in the characteristics of the technology and labor demand which comes with economic modernization.

Recent research at the Inter-American Development Bank looks at the demand, supply and institutional factors affecting labor market performance. The conclusions are that although labor participation rates have not changed much, there has been a change in the composition of the supply of labor with a larger proportion of female workers entering the labor force.⁽¹²⁾ The latter are more prone to unemployment because they work more frequently in the informal or unprotected sector. There is also evidence that demographics and schooling trends have increased income inequality in most countries.⁽¹³⁾

Overall, the pattern of demand for labor has not been biased in favor of the less skilled. That demand for labor in the region may be skill-biased is a puzzle. Economic theory would have predicted that trade liberalization would result in an increase in demand for goods which are produced with unskilled labor. Although no full answer exists, available studies point to four different explanations--not necessarily mutually exclusive--for this apparent paradox. The first explanation is that trade liberalization, financial deregulation and the resumption of capital flows cheapened the price of capital goods.⁽¹⁴⁾ Since skilled labor is complementary with capital, its demand rose. Second, although the region's abundant factor may be unskilled labor when compared with industrialized countries, this may not be the case when compared with countries such as China.⁽¹⁵⁾ Third, because of the characteristics of industrialization based on import substitution, the less skilled labor-intensive sectors were relatively more protected, and hence those were the hardest hit by the removal of barriers.⁽¹⁶⁾ Finally, in some countries, the appreciation of the domestic currency has led to a rise in the relative output of non-tradeables, often more intensive in skilled labor.

Research also indicates that the institutional characteristics of labor market legislation in the region may be another factor explaining the lower than expected employment generation.⁽¹⁷⁾ Higher rates of protection for formal sector workers, in terms of severance arrangements, unemployment benefits, etc., than are common in some OECD countries lead to rigidities in the labor market and high fixed costs of contracting new workers in the formal sector. In sum, the trend towards greater wage disparity is generated by supply, demand, and institutional factors.

These observed labor markets trends may be one of the factors explaining the persistence of high inequality in Latin America observed in the 1990s (see data appendix Table 1b). With the exception of Colombia and Costa Rica, the decade of the 1980s was a period of rising inequality for the nine countries with data at the national level.⁽¹⁸⁾ Unfortunately, not all the countries have information for the mid-1990s. The trend for the majority of the countries where income inequality indicators exist has been for inequality to stay the same or rise slightly between 1989 and circa 1995. This happened in Brazil, Colombia, Costa Rica, Mexico and Venezuela. Inequality declined only in Chile, Uruguay (urban), and Bolivia (urban).⁽¹⁹⁾

POLICY ACTIONS

FOR POVERTY REDUCTION

The inadequacies of current patterns of economic growth suggest that additional directed policies are called for in order to reduce poverty in Latin American and the Caribbean. Growth may be too slow or too volatile in the majority of countries to result in significant declines in poverty. Given the existing distribution of income in other countries, even steady growth rates will result in only modest poverty reduction. Throughout the region, market forces are not producing the unskilled-labor intensive growth pattern desirable for poverty reduction. All these factors have important policy implications.

The brief overview presented above leads to important implications in terms of actions to reduce poverty. Besides the obvious one that policy efforts should concentrate on promoting steady growth, the implications are seven, as follows:

Building Human Capital and

Upgrading the Skills of the Poor

Because recent observed growth patterns in the region do not appear to be intensive in unskilled labor, a tendency that could continue well into the future, policy initiatives should both eliminate any biases against the use of labor (particularly biases against unskilled labor which may be imbedded in the countries' legal, regulatory and tax frameworks), and reduce the relative supply of unskilled labor. The latter could be accomplished through two main mechanisms: reduced population growth rates (particularly among the poorer groups of society since, in general, the latter continue to show much higher population growth rates than the non-poor), and upgrading the skills of labor, particularly new entrants into the workforce.

Changing the Distribution of Assets

Since poverty is linked to the lack of assets, policies in this area are of great relevance.⁽²⁰⁾ Changing the distribution of assets could be accomplished through, for example, land reform, programs which distribute shares among the population and tax reform, including inheritance taxes. And, over time, more education can prove to be the critical asset for the poor.

Enhancing the Poor´s Access to

Market Opportunities

Enhancing the access of the poor to market opportunities can be obtained through a large variety of actions such as correcting the market failure present in, for instance, the credit market; reducing discriminatory practices in, for example, the labor and credit markets and in the judiciary system; improving the poor's bargaining power; improving access to poor regions through the provision of physical infrastructure; enhancing the poor's ability to use their assets as collateral through, for example, land titling; and, increasing the poor's access to information and technology. Programs that counter crime and violence can also open market opportunities for the poor by fostering an environment where they can commute to and from work without fear.

Improving the Quality of Life of the Poor

Neighborhood improvement programs in urban areas and investments in water and sanitation and environmental clean-up efforts can disproportionately benefit the poor, who are more likely to lack access to physical infrastructure and services and who most often suffer the effects of environmental degradation. Such policies offer triple benefits. First, the quality of life is directly improved through the provision of clean water and sanitation services, or improvements in housing, transportation, and the like, with resulting improvements in health status and economic opportunities. Secondly, the value of dwellings increases, and if land titling is provided, so does the collateral of poor households. Finally, investment costs for infrastructure provided can represent direct transfers to poor households, as cost recovery is normally only considered for operating costs of water and sanitation services.

Direct Transfers to the Poor

As mentioned in the second section, even with sustained growth, it could take a long, long time to eliminate extreme poverty. Although policy actions such as improving the skills of the poor and changing the distribution of assets could speed up the process, it would still take time to reduce extreme poverty significantly. In the meantime, tackling extreme poverty and striving to avoid its intergenerational transmission will require direct transfers to the poor, particularly the poor who live in backward areas or belong to underprivileged groups such as the indigenous population. Interestingly, in most countries the effort should not be unsurmountable: some very simple calculations indicate that if it were possible to target between less than 0.5 and 2 percent of GDP to the people living in extreme poverty, they would no longer qualify as extremely poor. The best direct transfers are those which "kill two birds with one stone." For example, transfers in the form of scholarships for poor families to entice them to keep their children in school or as compensations for visits to health posts could increase current income and at the same time help build the human capital of the poor.

Helping the Poor Cope with Adverse Shocks

As shown above, adverse shocks can lead to very sharp increases in poverty at the country level or for the groups that are hit. Countries, regions, sectors, villages, and individuals are likely to face adverse shocks. These shocks can be systemic or idiosyncratic. Systemic shocks can be of economic, natural or political origin. They can also be the product of reforms designed to correct market imperfections. Recent examples abound: the peso crisis in Mexico, the fall in international coffee prices in the early nineties, El Niño, episodes of rapid trade liberalization and privatization, and the impact of the Asian currency crises on growth in Latin America and the Caribbean. Likewise, the poor are likely to be particularly hurt by idiosyncratic shocks such as illness, physical and mental disabilities, unemployment and so on.

The poor are ill-equipped to deal with shocks, and informal insurance arrangements have serious limitations. Notably, most countries lack the institutional mechanisms and instruments to mitigate the impact of adverse shocks on the poor and the responses are frequently improvised. Developing such mechanisms would do a lot for poverty reduction. The data indicate that the open unemployment problem is disproportionately present among the poor, so workfare programs may be one obvious solution, but not the only one.

Providing Social Protection

for the Unemployable

There is always going to be a group of people in society that is unemployable due to permanent sickness, old age, physical and mental disabilities, and so on. The poor are particularly vulnerable to these phenomena and society must provide adequate safety nets for them. In Latin America and the Caribbean, the demographic transition indicates that the "aging" of poverty is likely. Policies aimed at protecting people from poverty in their old age are fundamental and it is not clear whether current systems of social security (be they existing or reformed programs) or social assistance are ready to meet this challenge.

The seven policy spheres discussed above apply to actions which are most often taken at the national or sub-national levels. In addition, there are a set of initiatives particularly salient for institutions such as the IDB which can be taken in multilateral forums or by industrial nations bilaterally and that affect the evolution of poverty in the developing world. International organizations and industrialized countries can make a contribution to poverty reduction in the developing world through at least three main channels: First, richer nations can make capital available for the capital-poor countries through grants and long-term loans at concessional rates;⁽²¹⁾ reducing current official debt levels such as the recently launched Highly-Indebted Poor Countries initiative; providing financial safety nets in the face of adverse shocks; and, direct bilateral aid. Second, opening markets in industrial countries for agricultural products and promoting free trade or extending benefits of preferential agreements can help boost developing countries' exports, improve access to modern technology, and encourage private capital inflows. Imposing non-onerous terms in transfers of intellectual property rights can help poorer countries gain easier access to modern technology. Third, multilateral institutions can assist countries in the design of sound policies, and through their lending program and policy dialogue influence the policies and allocation of resources by individual countries to better target the poor. The second half of this document addresses the spectrum of options available to the Inter-American Development Bank through which it can contribute to poverty reduction in Latin America and the Caribbean.

Before that, one important comment is in order. Policies that help the poor need to address aspects other than their economic well-being. There are dimensions to poverty besides lack of income whose effects have as great an impact, if not more so, on the lives of the poor and which can cripple the ability of individuals, households, or entire segments of the population to fully participate in the economy. Indigenous groups are disproportionately represented among the poor, particularly those in extreme poverty. Urban poor living in marginalized slums are more often victims of violent crime than their wealthier neighbors. Poverty can contribute to vicious circles within the household, as low-income and lack of economic opportunities provokes acts of domestic violence, which in turn result in heavy economic and social costs for household members who bear the effects of violent acts.⁽²²⁾ Policies aimed at overcoming marginality, social exclusion, and violence are an important branch of any anti-poverty strategy.

Part II

The Inter-American Development Bank

and Poverty Reduction

Since its Fifth General Increase of Resources (IDB-5) in 1979, the Bank has had a target of lending 50 percent of its resources to benefit lower income groups. That commitment was renewed in subsequent replenishments and strengthened in IDB-7. In 1994, under IDB-8, the Board of Governors reaffirmed its commitment to the needs of low-income groups and the goal of reducing poverty.

BANK TOOLS AND PERFORMANCE

IN POVERTY REDUCTION

To support the goal of poverty reduction, the IDB works with its borrowers to consolidate economic reform, while improving the efficiency and equity of social spending, providing social safety nets that protect the poor during economic downturns, and offering emergency support against devastating natural disasters. The policies and programs of the Bank reflect both the lessons learned from past lending and the results of an ongoing dialogue with stakeholders both in the public and nongovernmental sectors.

The basic tools available to the Bank are lending, advisory services and technical assistance, and the development of strategies and policies and conducting research.

Lending. The Bank has built up a portfolio of project and sector reform loans. These have had a sizable impact on the region. For example, through its education loans the Bank has provided training for nearly 20 percent of the region's primary and secondary teachers.⁽²³⁾ Support for social investment funds has totaled over $1.3 billion, helping to ameliorate some of the social and economic costs of structural adjustment policies in recent years. Debt reduction programs are being considered to lower the debt burden of the poorest countries, such as Bolivia and Guyana.

Advisory Services and Technical Assistance. As a complement to its loans, the Bank works to create the right conditions for social lending with its technical support to ministries, government agencies and nongovernmental organizations; training in social policy; and stakeholder dialogues to develop national social agendas.

Strategy, Policy and Research. The Bank is engaged in an ongoing process of policy research and strategy definition in order to best respond to the changing conditions in the region, benefit from lessons learned and incorporate new practices. Strategy formulation for critical sectors and instruments, such as a reappraisal of social investment funds, and research on selected topics are part of a continuous effort to keep Bank lending and advisory services up to date.

THE OVERARCHING CHALLENGE:

PROMOTING STABLE AND SUSTAINED

ECONOMIC GROWTH

During 1997, the Bank officially presented its strategy for poverty reduction in the region.⁽²⁴⁾ The strategy is based on the fundamental premise that the only way to permanently reduce poverty is to create the conditions where the poor can earn their way out of it. Stable and sustained economic growth is a prerequisite for this and the Bank is contributing to this goal in a number of ways.

For the very poor countries, the Bank is filling the vacuum of capital flows left by the private sector. For the richer countries in the region, the magnitudes of lending may not be at center stage given the relative magnitudes of private capital flows. In both the poorer and richer countries, through lending operations, research activities, training and knowledge dissemination, the IDB is contributing to effort to put in place the policies and institutional environment that are conducive to stable and sustained growth.

Among the main ingredients required for stable and sustained growth are: sound macroeconomic policies, healthy and stable financial systems, adequate infrastructure, a well-trained labor force, open markets, good governance, and social and political peace. The Bank portfolio includes a number of projects which contribute to enhancing the countries' macroeconomic stability: for example, the financial sector loans, the pension reform loans, and loans to reform budgetary institutions. These operations improve the prospects for fiscal discipline and promote the development of domestic financial systems.

Moreover, the Bank's research on sources of economic and financial vulnerability has helped focus operational efforts to reduce the weaknesses inherent in budgetary processes and financial systems. This research has led to innovative lending operations such as the "Program for the Budgetary Process and Strengthening Policy Analysis" in Venezuela approved in 1996. This operation was directed at the interface of macroeconomic policy and development. It attempts to tackle two key problems: the coordination of macroeconomic policy between the Executive, the Central Bank, and Congress, and enhancing fiscal policy analysis, without requiring profound changes in the legal structure surrounding the budgetary process. The operation finances the creation of two offices: the Office of Macroeconomic Policy in the Treasury and the Office of the Budget in Congress. The project was accompanied by the creation of a committee to coordinate policy within the Executive, and reform of congressional procedures such that any change in a law or new law, being considered by Congress, has to be accompanied by a fiscal cost estimate prepared by the Congressional Budget Office before leaving the committee stage.

Even if countries do everything right--or almost right--they are likely to face occasional adverse shocks due to volatility in the capital markets, sharp turns in commodity prices, natural disasters, and so on. The IDB can and has been an important player in international financial rescue operations designed to cope with the effects of adverse shocks of economic or other origin. Recent examples are the emergency loans to Argentina and Mexico in 1995 in the aftermath of the Mexican devaluation, and the loans to Ecuador and Peru approved in 1997 to cope with the impact of El Niño. The IDB is also helping the countries prepare themselves for adverse shocks by monitoring and analyzing the implications of external events, and by advising the countries on how to create their own macroeconomic safety nets and ensure the soundness of their financial systems. One of the conduits for the exchange and dissemination of knowledge is the Latin American Network of Central Banks and Ministries of Finance sponsored by the Office of the Chief Economist which meets in Washington twice a year.

Open markets and economic integration are supported by a series of initiatives undertaken by the Integration, Trade and Hemispheric Issues Division to give technical assistance to the Central American Common Market, CARICOM, MERCOSUR, and the Andean Community. These initiatives are directed to improving the institutions, expertise, information systems and infrastructure which will facilitate the integration of the countries into the regional and world economy.

Good governance is promoted through loans which concentrate on institutional reforms and the reforms of the judiciary system and, for example, the training provided to mid-level managers in the social sectors in the Inter-American Institute for Social Development. The IDB has also supported peace processes and reconciliation in the region, through projects such as the Guatemala Community Development for Peace Program, approved in 1996, the Small Projects and Technical Cooperation Facility for Marginal Groups in southeastern Mexico, approved in 1997, and widespread support to social investment funds in post-conflict areas.

LENDING HIGHLIGHTS OF PROJECTS

WITH BENEFITS FOR THE POOR

As described above, the IDB-8 mandate culminated many years of effort to raise the amount of social lending by the Bank. As a result, the share of social sector lending (largely for health, education, urban development, water and sanitation projects) has burgeoned to over 40 percent of annual lending, up from an average of less than 20 percent in the period 1961-93. More than one-half of the IDB's social sector lending (in real terms) has taken place since 1990 (see data appendix Graph 1). In addition, the Bank's efforts have led to improvements in poverty targeting within projects. During the 1995-97 period, more than one-third of Bank loans were poverty targeted, amounting to $6.7 billion in lending, one-third the value of all lending⁽²⁵⁾ (see data appendix Tables 7 and 8).

Moving beyond the aggregate targets, the Bank has taken steps to increase the poverty focus of its lending program. The broadest measures include the C and D Action Plan, targeting of FSO resources to the poorest countries, and interest rate subsidies for poorer countries under the Intermediate Financing Facility. At the project level, the Bank is tackling poverty reduction with social investment funds and microenterprise lending to reach the poorest within country, and new targeted programs to the neediest groups in society, such as women, children and ethnic groups. The discussion that follows highlights Inter-American Development Bank lending activities which aim to directly benefit the poor. The projects are grouped into the seven policy areas which are likely to have the most leverage in poverty reduction, as presented above.

Building Human Capital

and Upgrading the Skills of the Poor

One of the most important ways of upgrading the skills of the poor is by investing in their human capital. This concept includes a wide array of projects, such as extending the delivery of good quality social services, improving the social and urban infrastructure of the poor, and training programs to direct transfers tied to nutrition, health and educational objectives.

Since the approval of the 8th Replenishment, the Bank has sharpened its focus on support to social equity and poverty reduction efforts in LAC with a series of simultaneous country- and sector-wide actions. Region-wide, the Bank has supported actions aimed at improving the reach and quality of services while, at the same time, promoting cost-effectiveness in their delivery, either through sector reform programs or through decentralization and municipal strengthening projects.

At the sectoral level, the Colombia Support to Health Sector Reform approved in 1994 and the Paraguay Health Care Reform recently approved in 1997, are aimed at promoting access to health care services as well as their efficient provision. However, while the Colombian program's emphasis is mostly on preventive and primary health care and on rural and poor areas in general, the program in Paraguay is focused on maternal and child health. Both programs, however, stress the need to focus improving quality and access to health care in a cost-effective manner, with the Paraguay program being part of the ongoing effort to decentralize the delivery of services in the country.

In Education, emphasis on quality has been a major drive of reform programs. As of mid-1997, the Bank had 20 projects under execution in the areas of primary and secondary education. A study of the current portfolio of basic education projects⁽²⁶⁾ highlights the Bank's movement away from top-down methods of program preparation and implementation and towards more participatory approaches to service delivery. Throughout the 12 years of investment history covered, the programmatic emphasis of Bank loans have evolved from construction and infrastructure to an explicit concern for sustainability and accountability. This evolution can be observed along various dimensions: concerns for improving educational coverage largely have been replaced by concerns for quality improvement; responsibilities (e.g., for maintenance) once falling to central bureaucracies increasingly fall to schools and communities; consultant input for the preparation of books and other materials is gradually being replaced with that from teachers and the community at large; and, through the implementation of mutually reinforcing activities such as training, assessment, and incentives, the teacher-learning link is being made explicit. This, in turn, makes public systems more accountable. The magnitude of such changes in the evolution of project design in the past twelve years, significant in and of itself, appears even greater when current investments are compared with those of past portfolios.

Trends evident since the approval of the first loan to primary education (1976) point to the gradual increase in lending to primary education so that it now outpaces higher education and vocational training. Investments in primary education have increased at an annual rate of almost three times that of overall Bank investments in the education sector: 14.3 percent versus 5 percent, respectively for 1977-1995. Whereas primary education accounted for 28.7 percent of Bank resources in the education sector for the period 1976-1980, shares for higher education and vocational/technical training were 42.1 percent and 29.2 percent, respectively. Comparable figures for a decade later effectively illustrate the depth and speed of change in the Bank's education portfolio: between 1986 and 1990, primary/basic education accounted for 45.3 percent of Bank resources invested in education; with the shares for higher education and vocational/technical training reaching 23.3 percent and 26.3 percent, respectively.

These trends have continued in recent years and are evidenced in the portfolio of loans currently under execution. As a result, Bank investments in education are overwhelmingly concentrated in primary education. Over two-thirds (67.6 percent) of such investments support primary education; a mere 3 percent goes to higher education, and 14.7 percent supports vocational/technical training.

Total current Bank investments in books and educational materials exceed US$450 million. In the aggregate, education projects under execution finance the production and distribution of almost 35 million books for study at the pre-primary, primary, and secondary levels. In the aggregate, the 20 loans under execution in support of basic education invest more than US$828 million in physical infrastructure. Such investments include the construction, rehabilitation, repair, and/or expansion of more than 30,000 classrooms and over 5,000 schools. Investments in infrastructure consume a larger share of project resources than any other investment category.

If implemented as stipulated in the loan documents, training activities included in Bank projects currently under execution will benefit approximately 17.5 percent of the region's teachers. The volume thus invested is significant, totaling almost US$335 million, and the level of preparation of such activities is high. The amount that the Bank has invested in teacher training for the region is truly outstanding. Nearly one-fifth of teachers in Latin America and the Caribbean have received or are to receive training financed, at least in part, by the IDB.

Another range of interventions which contributes to the building of human capital are early childhood care and development programs, currently receiving a resurgence of interest and public funding the world over. The combined benefits of custodial care, which offers the ever growing percentage of women participants in the labor force alternatives for taking care of their children while working, and integrated early childhood development programs, which research has shown to yield positive longitudinal effects on the subsequent social and economic development of participating children, have led to an increased attention on the part of the international organizations in funding such programs in the developing country context. As do other social programs, such as those providing family subsidies for school attendance, child care and development programs offer double benefits, wherein not only is the family's income increased, but the children themselves are provided critical extra support and attention which improves their school-readiness and affords them a better chance of breaking the cycle of poverty. Recent research carried out by the Poverty and Inequality Advisory Unit (using survey data from a sample of favelas in Rio de Janeiro which receive support from the Rio de Janeiro Slum Upgrading Program) suggests that having access to child care services can contribute to an increase in mothers' monthly earnings by as much as 25 percent.⁽²⁷⁾

The Inter-American Development Bank has joined the ranks of agencies funding early childhood development programs, with more than US$1.9 billion of its current portfolio of loans in execution supporting these programs--either free-standing day care programs, or pre-primary education. Independent operations in support of early childhood programs are beginning to appear, but for the most part the IDB's support to these has been contained as sub-components or activities within projects in education, urban development and housing, integrated rural development, and social investment funds. The current pipeline continues this trend, with an increasing importance for independent early child care programs grounded in national strategies or policies in this area which reflect the growing status of these programs within both the Bank and the borrowing countries.

In addition to these approaches to upgrading the skills of labor through long-term investments in human capital, the IDB is also involved in training projects which have more immediate effects. The IDB has been contributing to poverty reduction in the region through an array of projects designed to upgrade skills of new entrants into the labor force, thus reducing the relative share of unskilled labor over time. Examples of such efforts are the Program to Support the Employability and Productivity of Youth in Argentina, approved in 1997 and the 1992 Chile Jóven (Young Chile) program.

The Chile Jóven project is nearing completion, and represents a bellwether project in vocational training within which a variety of forms of training have been mounted to meet a diversity of needs of both private sector employers and disadvantaged youth. It offers, for example, on-the-job training in enterprises; alternating work and training; courses for the self-employed; programs for the technically and socially unskilled youth and special women's programs. The program merits special attention for the lessons and benefits it promises to yield. The Bank has already used elements of this program in the design of projects in Argentina and Uruguay.

In Argentina, the Programa de Apoyo a la Productividad y Empleabilidad de Jóvenes, a US$370 million loan approved in 1997, aims at supporting the Argentine government in its efforts to increase the productivity and employability of poor youngsters through a series of subprograms designed to provide training and orientation for work, to reduce dropout rates through the distribution of scholarships for youngsters at risk, and to bring overall improvements in the quality of education provided in PSE (Plan Social Educativo) schools (this project is also discussed below in the section covering direct transfers). Direct benefits will result in the form of higher wages and higher productivity of young workers, while training costs and the impact of social marginality on youngsters are expected to be significantly reduced.

The emergence of Social Investment Funds during the past ten years in most Latin American countries constitutes an important development in the field of social policy and poverty reduction. The Funds have successfully delivered small social projects to poor and underserved areas, improving the welfare of the poor. Mainly, they have invested in schools and health posts thereby contributing to the human capital formation of the poor. The Bank has been the principal external backer of SIFs in the region. Its contributions represent about half of all the external funding that the SIFs have received. The Bank has financed SIFs in 16 countries, largely through concessional loans, for a total of US$1.3 billion. With the exception of the emergency SIF loan to Bolivia, all of these loans were made in the 1990s. Learning from other experiences, the loan supporting the Jamaica Social Investment Fund, approved in 1997, is indicative of a new generation of Bank support in which the Social Investment Fund plays a central role in implementing an overarching poverty reduction strategy for the country, with enhanced flexibility for the Fund's activities menu based on careful analysis of the local causes and consequences of poverty.

The Bank is also fostering the development of human resources through the subset of projects targeted to the poor through the Human Resources window of the Multilateral Investment Fund (MIF). In its four years of operations, the MIF has approved 25 projects for a total of US$47 million through this window. On average, roughly 60 percent of MIF projects go to C and D countries.

Finally, the Bank is currently preparing a US$2.4 million regional technical cooperation operation on labor markets and poverty to strengthen the technical capacity of the ministries of labor to monitor and evaluate the evolution of the labor market and its impact on poverty. The program consists of two main components. The first, to be carried out by the IDB, is a regional in nature and involves the development of methodologies to evaluate the labor market's impact on poverty and efficiency, as well as the effectiveness of employment generation activities. It also involves the development of systems to collect and process data, as well as the realization of workshops. The second component will be implemented by the various ministries of labor, and will finance the implementation of the systems developed under the first component, evaluations of the impact of the labor market evolution and employment generation activities on poverty, and publication of the results of the pilot studies and methodologies.

Changing the Distribution of Assets

Recognizing that lack of access to land leads to the perpetuation of inequality in LAC, the Bank actively supports initiatives in urban and rural areas to increase the poors' access to land and title to their dwellings. Along these lines, the objectives of Colombia's Programa de Titulación y Modernización de Registro approved in 1997 are to consolidate an open and transparent land market in urban and rural areas through the legalization of land ownership by issuing deeds and officially recording them and modernizing the deed-recording and cadastre systems. This program also stresses the protection of environmentally fragile areas and has built-in safeguards to protect the collective rights of ethnic minorities against the potential effects of land titling in rural areas. Registered title will be awarded to approximately 100,000 parcels of uncultivated land in 200 rural municipalities and 150,000 lots of urban property in 50 municipalities.

The Independence for Rural Settlements project in Brazil, which is currently under preparation, represents an innovative attempt to systematize and accelerate the process of land reform in a country where unequal patterns of land distribution were exacerbated by the distribution of subsidies that drove land prices above the present level of agricultural returns, making it inaccessible to small farmers. This is a pilot program targeting 12,500 families (approximately 75 communities) living in land reform settlements under the jurisdiction of the National Institute of Land Reform in seven states. By its end, a decentralized model will have been developed and tested to more quickly emancipate land reform communities from government support.

The program's major objective is to make land reform settlements independent from government support and self-sufficient, thus allowing for the accelerated incorporation of new landless families to the land reform process. To reach this goal, the program addresses the three major impediments that have prevented a speedier process of self-sufficiency in land reform communities; namely, lack of infrastructure in settlements; lack of social and technical assistance services to settlers; absence of a reasonable set of procedures to guide the process of transforming settlements into economically sustainable units.

Another way of giving the poor a more equitable access to assets is to support actions to correct market failures, like those to enhance the poor's access to credit. The Bank is a pioneer and leader in the field of microenterprise development beginning with projects in the late 1970s, when the sector was made a priority under IDB-7, and steadily increasing since that time. Most recently, the Bank launched an ambitious program to lend $500 million over five years as part of its Microenterprise Development Strategy--MICRO 2001.

In the past two decades, the results of IDB lending have been impressive. The Bank has funded various microfinance institutions many of which have grown substantially, and serve thousands of microentrepreneurs. These include Bancosol and Caja de los Andes in Bolivia, and Financiera Calpia in El Salvador, among others. The microenterprise sector is a significant area of IDB investment; 471 microenterprise operations totaling US$452 million were approved betweeen 1990 and 1996. To date, it is estimated that IDB-sponsored microenterprise programs have reached approximately 600,000 microentrepreneurs and supported 1.8 million employment opportunities. This success has been founded on institutional support with the creation of the Microenterprise Unit, the innovative use of global loans for small and microenterprises, and the establishment of the Multilateral Investment Fund.

Lending experience in this sector demonstrates that microenterprise operations are most successful when an enabling policy and regulatory environment exist. The Programa Global de Crédito para Microempresarios (PR-0097), approved in 1997, continues the efforts initiated with the 1992 microenterprise global credit program. It supports the development of production and business capacity for microentrepreneurs in Paraguay through actions to facilitate their access to credit at market rates from financial intermediaries in the formal financial system, to establish microenterprise lending as a regular and self-sustaining financial activity, to eliminate regulatory barriers that might prevent microentrepreneurs from having access to the formal financial system, to change the way in which banks relate to microentrepreneurs, and to bring into the program other eligible market operators like cooperative banks, for instance.

The Bank gives particular attention to paving the way for successful credit and investment programs through its policy dialogue and reform programs. For example, support for financial sector reform and better bank supervision enhance financial intermediation and hence improve the access of microenterprises to financial services. Aiding the supervisory authorities to devise the appropriate regulatory framework for microlending institutions also helps to expand the sources of funds.

Bank projects in Jamaica and Trinidad and Tobago are aimed at strengthening the credit union industry and reforming the legal, regulatory and supervisory structure governing credit union operations. These programs help to create and strengthen both formal and non-formal institutions that provide needed business development and financial services on a sustainable basis. Attention also is paid to strengthening institutional capacities to provide services to low-income and disadvantaged microentrepreneurs efficiently and effectively. The Bank serves as a catalyst for private investment by linking NGOs and commercial investors and promoting the formalization of microenterprises.

Various tools are employed in microcredit operations. Since many microfinance institutions have a high proportion of non-poor clients, and reach the moderate poor more than the very poor, the Bank has relied on its Small Projects Program as to reach the productive activities of the poorest sectors of society. Small Projects help mobilize the actions of grassroots organizations such as cooperatives and non-profit foundations and associations. Currently, the Bank is reexamining this program to make it more effective.⁽²⁸⁾

Global loans have also been an important way to transfer resources to financial institutions such as NGOs, credit unions, and commercial banks that lend to microenterprises. Additional support is given by technical cooperation both within these operations and separately, that help to build microfinance and business development institutions (e.g. improvements of management information systems) and promote policy reform. Equity investments are also essential with the Multilateral Investment Fund providing capital investments to develop intermediaries that provide services to microentrepreneurs.

Modernizing the judicial system and improving the poor's access to it is yet another way to address equity issues and reduce poverty in the region. Access to a well-functioning judicial system will help the poor protect their assets and their rights. In the Dominican Republic, the Program to Modernize the Real Property Adjudication and Registration System approved at the end of 1997 proposes to implement a more efficient and transparent property adjudication and registration system within the country's judicial system, through components addressing legal and institutional consolidation in the areas of land act reform and administrative organization of the judicial system, the modernization of technological and information systems, the improvement of infrastructure, plant and equipment to protect record security and the strengthening of human resources.

In Peru, the Program to Improve Access to the Justice System, also approved in 1997, was designed to assist efforts to ensure better access to the justice system and improve its quality through investments in better built and equipped buildings and improved technological and information systems resources, training of human resources and technical assistance for the design of streamlined procedures for dealing with the public, as well as investments to improve protection of women's and children's rights through the use of lay magistrates trained by the program and the drawing of national policy recommendations based on this experience.

Enhancing the Poor's Access to

Market Opportunities

The Bank has also provided support to projects that aid the development of physical infrastructure in and around poor areas, particularly infrastructure enhancing the poor's physical access to services and to the market. Such is the case of Haiti's Program of Secondary and Tertiary Road Rehabilitation (HA-0075), approved in 1997, a good example of how the Bank can contribute to improving the access of rural populations to certain services. This project, aimed at the rehabilitation and maintenance of selected road networks in Haiti (with support to institutional development to ensure the sustainability of the investments) is expected to benefit a total of 750,000 rural residents in a country where the poor represent 85 percent of the rural population. The sustained improvement of the Haitian road infrastructure, to which this project contributes, is expected to have a significant development impact, particularly for the rural population, with support to initiatives to build local capacity and motivation for tertiary road management as well as actions aiming at strengthening small enterprises and community groups participating in road works.

More recently, in 1998, the IDB approved the Programa de Seguridad Ciudadana in Uruguay with the global objective of preventing violence and reducing the risk and perception of insecurity in the country. The program includes three components that address institutional strengthening issues and invest in the development of a culture of crime prevention among youths and local communities as a way of creating a solid crime prevention network involving both State and civil society in an effort to increase calm and reduce the economic and social costs associated with urban and domestic violence. A similar project was just approved for Colombia--the Peaceful Coexistence and Citizen Security Program. The project was specifically designed to reduce levels of crime and violence and supports a series of initiatives at the municipal level to promote peaceful coexistence and the prevention and control of urban violence, as well as help the national government develop policies and programs to promote an environment of peaceful coexistence and tolerance among citizens.

In addition, as the decentralized and cost-effective provision of services have increasingly become a central tenet in the Bank's lending program in recent years so has the recognition of the need to increase the stock of social capital in the communities reached by Bank-financed projects and programs. Through a series of actions, the Bank has been actively supporting initiatives to increase the involvement of beneficiaries not only in the definition of the investments they desire and will be paying for, but also in monitoring their implementation. As a result, the number of projects and programs specifically designed to provide support to community strengthening were coupled with the provision of community participation and community strengthening components in many other programs in recent years.

The Dominican Republic Community Action Development Program, for instance, approved in 1994, aims at a greater rationalization of social services delivery, particularly to low-income groups, while fostering participation in project management by NGOs, cooperatives, local private enterprises and community organizations. More recently, the Program of Support for Civil Society Organization Initiative and Management in Venezuela, approved in 1997, represents a pioneering experience in creating and developing partnership arrangements between the public and private sectors and civil society to fund projects and activities to be developed and executed by civil society organizations.

The Bank has also been active in improving access to job market opportunities. The Regional Program for Strengthening Technical and Professional Training for Low-Income Women intends to develop new training methods and encourage the participation of women in technical jobs. The project will research new job opportunities and provide on-the-job training through private-sector employers. The thrust is to strengthen the quality of training and options offered to women in traditional female occupations, higher paying male occupations and new occupations with proven demand in the labor markets. This program, which is currently being executed in Argentina, Bolivia, Costa Rica, and Ecuador should provide a demonstration effect in the region. Following an evaluation of the program, its results will be disseminated to regional training institutions.

Programs such as the US$15 million loan under preparation under preparation in Argentina for Institutional Strengthening for the National Women's Bureau are also exemplary with respect to the objective of increasing economic opportunities for women. The program seeks to create and strengthen provincial and municipal entities for women, whose functions would include training in formulating and implementing specific equal opportunity policies for men and women, carrying out prevention activities, and guaranteeing more efficient use of social policies and resources.

Increasing the poor's access to information and technology is another way of promoting social equity. This can be achieved directly through investments aimed at widening and expanding the poor's access to information and technology. Alternatively, it can take the form of mechanisms designed within programs which take the participation of beneficiaries in the design of actions one step beyond, providing them with information on a program which enables them to play a more active role in monitoring its execution and evaluating its impact.

Traditional solutions to improving the quality of rural education, such as hiring more teachers or constructing more schools, are often precluded by budget constraints. With IDB support, education authorities are embracing innovative solutions, such as introducing information technologies to education. New multimedia technologies hold the promise of improving the quality of teaching and student performance in rural areas. At the same time, methods of distance learning based on these technologies offer a cost-effective method for reaching rural and isolated communities.

Two recent examples are a project to support distance education in Mexico and a program to support education technologies in El Salvador. These projects finance the introduction of various technologies, including distance education based on television and interactive radio, and computers to teach specific subjects, such as mathematics and natural sciences. Perhaps as important as the financing of equipment, the IDB has insisted on a comprehensive approach to issues of implementation and evaluation. To maximize the educational value of these new technologies, considerable resources have been devoted to adapting curricula; providing adequate materials and infrastructure; training teachers in the use of technology and the application of innovative teaching methods; and involving the community in the purchase and maintenance of the equipment. To test new methods and techniques, many of the activities are planned at the pilot level. Both the pilot and full-scale efforts include proper evaluation designs aimed at measuring student performance.

A similar project, but within a different educational context, is the Barbados Education Sector Enhancement Program, in the pipeline for approval in 1998. Having achieved universal basic education, Barbados is now facing the challenge of continued improvement in educational quality as a means to strengthen its capacity to keep pace with economic and technological change in a cost-effective way. As a means of responding to this challenges, the Barbadian system of education proposes revising the curriculum to emphasize new "higher order" learning skills, improvement in basic-skills instruction and strengthening the capacity of the teaching corps to target instruction according to students' needs. As the backbone of this shift, investments in information technology will provide the means of addressing the needs of a well-educated and skilled labor force to cope with the challenges of an ever-changing market.

On the other hand, programs like the Ceará Social Reform (BR-0177) approved in 1996 and, more recently, the Program of Comprehensive Services for Children under Six (BO-0130) address the need for specific mechanisms to make information on the project's actions widely available to beneficiaries, enhancing the active role of communities not only in the design of actions but also in monitoring their implementation and evaluating their impact.

Improving the Quality of Life of the Poor

The concern with more equitable access to assets and improvements in the quality of life of the poor is also present in projects designed for the urban poor. Programs like the Rio de Janeiro Slum Upgrading Project, approved in 1995, have sharpened even more the focus on improving the living conditions in urban areas through a concentration of investments in the poorest, deteriorated slum areas of big cities. The program, known locally as the Favela-Bairro program, represents an ambitious attempt to provide infrastructure, social services and land titling to slum residents while fostering, through a series of architectural solutions, the visual and physical integration of slums to neighboring areas, a movement expected to contribute to the virtual elimination of some of the social barriers traditionally setting slums apart from the rest of the city. This program, now two years into execution and reaching 90 favelas, will soon be followed by a second phase.

Current Bank housing policy stresses the need for sector reform to increase the capacity of the housing sector to satisfy the needs of the poor. The most recent generation of housing projects emphasizes reforms facilitating the development of efficient private mortgage finance for middle- and low-middle-income families--the latter with direct up-front subsidies from the government --to prevent the crowding out of housing solutions targeted to the poor. Low-income housing solutions in these projects include a vast array of actions to facilitate poor households' access to land and promote secure tenure while supporting progressive housing programs. Examples include recent projects in Guatemala and Ecuador, where housing sector program loans are expected to significantly contribute to improving the poor's access to the housing market through components addressing the need for legal and institutional reforms and the creation of credit mechanisms and market-oriented incentives.

Both programs combine, to a greater or lesser extent, components aimed at the development of mechanisms to increase the poor's access to the housing market, be it through direct targeted subsidies or through accessible credit mechanisms for the purchase of a house or renovation of the informal one some families already occupy. Also included is reform of land market legislation to simplify land titling procedures and the regularize properties in informal settlements. Expected benefits of programs like these are the increase in the equity and cost-effectiveness of public investments in the housing sector, with the design of more transparent demand-based, targeted subsidies which, in turn, provide incentives for a more active role of the private sector in the housing sector through additional investments.

Fighting pollution in urban areas disproportionately benefits poor households. A 1995 sanitation loan for the Valley of Mexico supports improvements in basic sanitation in Mexico City through a series of actions in wastewater treatment and disposal, the rehabilitation of the draining system in the metropolitan region, and management and control of industrial waste disposal. A similar operation was approved for Nicaragua in 1996 for the control and regulation of water and land resources in Managua, with the objective of controlling industrial pollution and improving wastewater and solid waste disposal methods.

The Uruguay Municipal Development Program approved in 1997 responds to the need to increase the administrative and fiscal efficiency of the country's departmental governments so they can better promote local development. The program finances investments in 18 departments of the country's interior in the areas of neighborhood improvement, urban streets and roads, drainage, sanitation and city services. It also finances strengthening of the technical, financial and city works and services management capabilities of departmental governments. Generally speaking, through infrastructure and urban development programs, the IDB contributes to enhancing the health and nutrition levels of the poor, two key ingredients for improving skills. An upcoming project in Mexico, Modernization of the Water and Sanitation Sector in Rural Areas, will expand coverage of these essential services to poor communities in isolated areas of rural Mexico.

Direct Transfers to the Poor

The IDB has had limited experience in this area. The Honduras Family Assistance Program, approved in June of 1995, provides US$20 million to support the provision of cash transfers to poor families in municipalities where extreme poverty is concentrated. The program seeks to raise the purchasing power of these families, as well as improve their education, health, and nutritional status. Cash transfers are provided to families who keep their children in school as well as to families where the mothers utilize public health facilities for maternal and child health services. A mid-term evaluation of this program will be carried out in 1998, and should yield useful lessons which can be applied in the design of new operations.

Another IDB project which provides targeted cash scholarships to the families of secondary school students is Argentina's Support for Youth Productivity and Employability (mentioned above). One subcomponent of this program offers scholarships to poor households to fight against high dropout rates at the secondary level. The scholarships are targeted according to geographic poverty criteria, then according to educational performance indicators, and finally, through needs assessments at the household level. The programs in Argentina and Honduras both provide the double benefits of immediate income smoothing as well as fostering long-term investment in human capital.

Helping the Poor Cope

with Adverse Shocks

More often than not, the poor's share of social costs in sector reform environments is disproportionately high. The poor also tend to be harder hit than other groups by the effects of adverse economic shocks as well as natural disasters. For this reason, the Bank has often supported projects designed to provide social protection to the poor in such situations as well as projects with embedded mechanisms to cushion the poor from the adverse impact of necessary reforms. This was the case of the PROSSE Program in Mexico (Programa de Servicios Esenciales--ME-0187) and of the Sector Reform Program in Argentina (AR-0189) which were approved in 1995 in the wake of the economic crisis to cushion the poor from its effects. The Mexican program provided basic services and protected essential social services for the poor with actions in specific areas like health and education. Support to government actions included labor force training, temporary employment and nutrition programs. The loan to Argentina protected targeted social programs in maternal and child health, nutrition, employment, and other social services from budgetary cuts. Efficiency and effectiveness criteria were used to select those programs with the greatest likelihood of reaching the poor.

More recently, in 1997, the IDB approved two emergency loans, to Peru and Ecuador to support actions designed to alleviate the anticipated effects of the El Niño cyclical weather pattern which was expected to hit these two countries particularly hard between September of 1997 and March of 1998. Both programs provide financial assistance at three different stages: (i) for preparedness activities, before the emergency, which have been carried out since July; (ii) during the emergency, through the provision of immediate, short-term relief to the affected populations; and (iii) after the emergency, for the rehabilitation and rebuilding of infrastructure and for economic recovery.

Providing Social Protection

for the Unemployable

The Bank has supported efforts to provide social protection for the unemployable by supporting the reform of social security systems and through programs which provide direct transfers to individuals unable to participate in the labor market.

The Venezuela Social Security Reform Loan, (VE-0100) a US$395 million operation to be approved this year, is one such effort. The program's specific objectives are to provide the regulatory framework that allows for full implementation of the proposed social security reform, while continuing to support development of the capital market and adapting pension plans that cover particular occupational categories. The proposed reform also includes provisions for minimum pensions for the working poor who are unable to cover the full costs of their annuities through contributions alone.

Programs such as Argentina's Support to Vulnerable Groups, approved in 1997, demonstrate the way in which the Bank is supporting direct transfers to poor and vulnerable groups that are unlikely to participate in the labor market. Working with municipal governments and civil society organizations, the program finances social services, food assistance, and provision of housing for vulnerable groups including female heads of households, unemployed youth, the handicapped, and the elderly. A subcomponent of the program is specifically targeted to indigenous populations. Also in this vein are the many technical cooperation operations which support programs for children and youths in especially difficult circumstances, particularly street children in urban areas, a population group which has borne a disproportionate share of the social costs of poverty and whose employment prospects are meager, at least in the short-term. As of June 1996 the Bank has supported eight free-standing operations for children and youths in extremely difficult circumstances, totaling more than US$70 million. The number of projects in this area has been growing exponentially.⁽²⁹⁾

NON-LENDING ACTIONS

Improving the Quality of Information

on Poverty and Inequality

Recognizing the importance of having adequate information on poverty for the design, implementation, and monitoring of policies, the Bank implemented the Program for Improving Surveys of Living Conditions in Latin America (ISLC) in 1997. Financed with RTC funds, the ISLC program was established in recognition of the urgent need of having an information system with reliable data to measure poverty, inequality, and general social indicators in all the countries in the region. The IDB is the lead agency involved and, together with the World Bank, ECLAC and the country governments, is providing financing, technical leadership, and overall coordination for a program to strengthen the institutional capacity in each country to implement and analyze a high quality multipurpose household survey system.

The program's objective is to adequately generate more reliable information (in terms of its scope and coverage, accuracy, and most importantly, relevance for policy making) on the living conditions of the region's population. It is hoped that the improved surveys and know-how fostered by the ISLC program will aid in the design, follow-up, and evaluation of the programs, projects and policies to reduce poverty and inequality. As of today the ISLC Program is being implemented in El Salvador, Paraguay and Peru. The plan for Argentina will be signed in the near future. The program expects to add two countries per year during the next three years.

In addition, the Bank made a concerted effort in 1997 to improve the quality and availability of information on poverty and inequality, as well as other social indicators. The Poverty and Inequality Advisory Unit, in partnership with the Office of the Chief Economist, has initiated work on revising the poverty and inequality estimates for countries in the region, with final results expected in the first quarter of 1998. This work has been greatly facilitated by the database on household surveys established as part of the ISLC program, an interactive household surveys data bank. The data bank is located in a server and administered by the Poverty and Inequality Unit. It is currently undergoing pilot testing to guarantee user-friendliness, and should be widely available within the new year. At present it contains household surveys for 17 countries in the region.

Research on Poverty Reduction

In the past, the Bank made four special poverty assessments to help design specific actions to aid the poor in countries with a serious temporary or structural poverty problem. Poverty assessments were done for Argentina, Barbados, Central America and Paraguay. In Argentina, the analysis focused on the unemployment problem; in Central America it concentrated on the structural determinants of poverty in El Salvador, Guatemala, Honduras, and Nicaragua; and in Paraguay it focused on the problem of rural poverty. In addition, the Social Agenda Policy Group has analyzed social policy and the causes of poverty in Bolivia, Chile, Costa Rica, Peru, Trinidad and Tobago, and Venezuela. These studies have been or are in the process of being published. In addition to these studies and the Poverty Strategy Paper, the Bank has also completed two new strategy papers, one on reducing rural poverty and another on the use of Social Investment Funds as an instrument for combating poverty, which should be formally approved during 1998.⁽³⁰⁾ As inputs to the strategy paper, the Bank commissioned individual studies for the countries where SIFs are in operation.

In conjunction with the UNDP and ECLAC, the Bank undertook a research project on the links between the macroeconomy and poverty in several countries of the region. The results will soon be published in a collected volume.

In addition, the recently created Poverty and Inequality Advisory Unit is undertaking research in the following areas:

Adequate Social Protection for the Poor

The Unit has initiated a multi-year effort to develop recommendations for the implementation of an "early warning" system and income (consumption) smoothing mechanisms for the poor in the face of adverse shocks at the macro, sectoral, regional, household, and individual level. The project will last at least two years and will address several issues. First, it will examine the extent of social protection received by the poor in the existing and reformed social security systems, unemployment programs, labor market regulations, employment programs, etc., in specific countries. The objective of this first phase is to provide recommendations on how social protection systems, labor market regulations, and other public programs could be improved to enhance social protection for the poor. In addition to an inventory of existing programs, SDS/POV is also financing theoretical work on the design of optimum unemployment insurance programs for the poor, and exploring how social insurance systems might be able to act as income-smoothing safety nets in responding to adverse shocks.

The Unit will also seek to provide staff in the regional departments of the Bank and policymakers in the region with guidelines on how to anticipate the effects of adverse shocks on the poor and for the implementation of adequate social protection measures. The project will explore how social protection mechanisms for the poor can be incorporated in the lending program for individual countries. As a specific example, the unit is currently financing research in Ecuador regarding how best to mitigate the effects of El Niño and strategies for developing better preventive measures for future natural disasters.

Explaining Changes in Income Distribution:

A Comparative Study of East Asia

and Latin America

Poverty reduction is a complex dynamic process which results from the combined effects of four phenomena: changes in personal earnings arising from productivity gains and/or shifts from a job or sector of activity to another; changes in occupational status; changes in the demographic composition of families; and, the replacement of older less educated by younger, better educated and more productive cohorts.

Understanding how this complex process works and how it depends on the structure and pace of economic growth is essential for designing policies aimed at reducing poverty. Yet, relatively little is known to date on this subject. Only recently has it become possible to make more or less systematic comparisons of aggregate poverty and inequality measures over time in a reasonably large number of developing countries. Economic phenomena governing the dynamics of poverty are of a microeconomic nature. Understanding them requires more than following the evolution of some summary poverty or inequality measures.

This project, to be undertaken in partnership with the World Bank, is designed to fill that gap. Based on the available household surveys for a number of countries in both Asia and Latin America, the project will undertake microsimulation exercises to obtain quantitative estimates for the contributions to observed changes in household income of four sets of effects. These are: the price effect, caused by changes in the hourly compensation of employed workers or in the prices of the output of the self-employed, and the shifts in the sector or occupational activity; the income effect, caused by changes in tax and public spending policies; the demographic effect, caused by changes in behavior related to marriage, fertility, etc.; and, the labor effect, caused by the individual decisions in terms of participating in the labor force, occupational choices, work "effort," etc.

The selection of countries was based on two factors: a diversity of development experiences and the availability of at least two comparable national household surveys that are more than ten years apart. It is proposed that the study be undertaken in: Brazil, Colombia, Mexico and Venezuela in Latin America, and in Indonesia, Malaysia, Taiwan and Thailand in Asia.

Allocation of Resources to the Poor

The Unit is working with John Roemer of the University of California, Davis, in developing a model for calculating the equitable distribution of poverty-reduction funds among countries by an international agency, and by national governments. The model will be tested empirically working with data on distribution of the IDB's loans and country information on poverty (as a measure of need) and per capita expenditures on social programs (as a measure of effort).

Design of Evaluation Schemes for

Anti-Poverty Projects

One of the priorities of the Poverty and Inequality Advisory Unit is to promote the incorporation of evaluation strategies to measure the impact of poverty reduction programs within Bank lending operations. In November, the Unit held a meeting with a group of leading experts in the field of program design and evaluation schemes, together with the representatives of PROGRESA (the new anti-poverty program in Mexico). The Unit plans to continue its work in the area of evaluation on several fronts. First, it will continue to offer its services as facilitator to convene short meetings on the topic, bringing to the Bank outside experts who can comment on specific projects or themes. Second, one pilot project will be selected by each of the Bank's three regions which would design and implement a rigorous evaluation scheme from inception, which could then be tracked over time and later disseminated as a best practice. Finally, the unit plans to commission a series of technical notes in the areas of monitoring and evaluation which would serve as guidelines for the operational departments of the Bank. Topics to be addressed include a discussion of the methodological issues involved in designing adequate evaluations of social and economic programs, and a checklist on elements to be included in the design of a successful evaluation strategy.

Best Practices in Targeted Social Programs

The Poverty and Inequality Advisory Unit is carrying out a series of applied studies of best practices in targeted social programs. Currently in the works are studies of PROGRESA, Mexico's new anti-poverty program, and an overview of the national beneficiary targeting system in Chile (CASEN) and how it is used with an array of social programs. The Unit is also finalizing research on decentralization and the delivery of targeted social programs in Argentina.

In addition, the Office of the Chief Economist and its network of research centers in the region have engaged in a number of poverty-related studies whose results are regularly reported in the Bank's Economic and Social Progress Report.

AREAS FOR FUTURE ACTION

The previous overview suggests some areas that should be more emphasized in future Bank lending and non-lending activities. This list is not meant to be comprehensive , but rather indicative of the type of measures and activities which are likely to have an important leverage in contributing to poverty reduction in Latin America and the Caribbean.

Closing the Education Gap of the Poor:

Supply and Demand Side Measures

The region's educational systems are in urgent need of reform to extend the quantity, but above all, improve the quality of education. However, a wholesale reform of the educational system is likely to take several years. In the meantime, governments and nongovernmental organizations could make an all-out effort to increase the quantity and quality of education that reaches the poor through other instruments which can complement the reform effort.

In Latin America there are many communities which do not have the minimal educational infrastructure and for which the "building-schools" solution still applies. To close the infrastructure gap in education, the region could make a more vigorous use of the Social Investment Funds. However, in order to target the resources to the poorest areas, the selection of projects will have to be preceded by a selection of geographical areas or 'barrios' where the infrastructure is lacking. Furthermore, in order to make the supply of education effective, Funds in some countries may choose to move into areas beyond financing the construction or refurbishing of schools and include among their tasks guaranteeing that the necessary complements to school buildings exist: e.g., textbooks, desks, blackboards, paper, pencils, etc. Even more, the Funds could assist the communities in improving the quality of the teachers by, for example, funding training courses or providing merit-based bonuses. Most importantly, if the Funds take on an expanded menu of activities, they need to ensure that the investments in their various forms are not duplicating existing functions and are sustainable by appropriate coordination with the line ministries and by training community leaders.

Working on the supply-side of education is not enough. For poor households sending children to school may be too expensive either because of the costs of transportation, school supplies, etc., or because of the opportunity cost of the children's labor. Given the imperfections in credit markets, poor households are liquidity-constrained, unable to borrow and without the resources necessary to keep their children in school. A recent study suggests that financial depth explains much of the differences in secondary schooling across countries.⁽³¹⁾ Targeted programs which give a transfer in cash or in kind to households in exchange for keeping their children in school become a necessary component of any policy initiative whose objective is to close the educational gap.⁽³²⁾

While the Bank's efforts on the supply side have been quite systemic, the demand side has not received as much attention. A comprehensive approach which promotes the overall reform of education systems is most likely to address both demand side and supply side issues in a systematic way which goes to the heart of the constraints involved in providing quality education services for the poor. An innovative element being contemplated for future Bank support to educational reform is the possibility of a line-of-credit loan to support reform-oriented activities in education. Possible activities eligible for funding under this very flexible lending instrument would include improvement in educational statistics and management tools, better processes for student evaluation, a rethinking of teacher training, and pilot programs for improving learning. All of these activities could have impacts on both improving the quality of the supply of education, as well as increasing demand for educational services.

The Need for Mechanisms to Mitigate the

Impact of Adverse Shocks on the Poor

There is evidence that income variability can have a more devastating impact on the poor than on the rich. For one thing, poorer people face higher risks to infant life from a fall in consumption. Also, bad things tend to happen simultaneously for the poor. Seasons with bad weather (or changes in water temperatures in the case of fishermen) tend to bring low employment and low wage rates at the same time. Often, the incidence of disease rises too, affecting the ability of household members to work. Although much less researched, the urban poor's plight might not be substantially different when a macroeconomic shock, a natural disaster or the transient costs of economic reforms cause rising unemployment and falling real wages.

The risk-prone poor tend to set up informal arrangements to defend themselves from fluctuations in consumption. However, community-based risk sharing arrangements can have serious limitations. In particular, they break down when adverse shocks affect the community, region, or country as a whole. This may justify public actions to partly insure or subsidize poor people's production, employment and price risks. Several mechanisms have been tried with mixed results. Relief work schemes have proven less prone to negative consequences.

The Bank could play a leading role in helping countries put in place the safety nets required to protect the poor from adverse shocks. In recognition of the fact that the poor are often the most severely affected by adverse economic or climactic shocks, with the largest losses in the fight against poverty often occurring as a result, the Bank has already undertaken a series of initiatives to include components specifically designed to protect poor and vulnerable groups as part of emergency support loans for Mexico and Argentina in 1995, and for Peru and Ecuador in 1997. Nevertheless, problems in responding to crises recur because the institutional mechanisms to protect the poor from the brunt of the shocks are not in place beforehand. Therefore, responses are often rushed and are frequently improvised. Often, programs that were designed for other purposes are used to respond to shocks, and may not necessarily be targeted to those most in need.

Social Protection for the Unemployable

Poor and Direct Transfer Programs

Demographic trends in the region will lead to an "aging" of poverty. The Bank could help countries set up financially viable institutions to keep this from becoming a major problem in the future. There are three dimensions to be addressed in resolving this problem. First is reform of existing pension systems which, for the most part, cover formal sector workers, including the working poor.

Second, the Bank needs to address the question of social protection for informal sector workers who are currently excluded from existing social security systems but, given the opportunity, would have the means to contribute. The design of mechanisms to encourage their contributions to social insurance schemes and financially sound payment plans remains an unmet, yet priority, need. Third, there is the issue of providing social protection for those who either are unable to generate a minimum level of subsistence through their contributions because they are too poor, or are unable to work due to physical or mental disabilities. Contributory pension systems will not provide the solution for these people. The challenge is to devise the most effective ways of transferring income while at the same time minimizing institutional inefficiencies and the crowding out of private transfers (which can account for a significant share of the income of the poor).

There is ample scope for knowledge-building in this area, including country-level reviews of existing transfer programs and a better understanding of the relative effectiveness of different transfer mechanisms. In its policy and operational dialogue with borrowing countries, the Bank can advocate for a review of the existing old age disability programs in the region, an analysis of future needs, and the design of more streamlined mechanisms to protect this vulnerable segment of the population.

In addition to the need to provide income maintenance for the elderly poor, a need will continue for direct transfers which serve to maintain the income of households with a high proportion of non-economically active members. The points raised above remain equally salient: incentive effects, appropriate targeting, and adequate delivery mechanisms must all be considered in order to design transfer programs which provide a minimum income for the non-working poor in the most cost-effective way.

Building an Information Base on Poverty

and Inequality

The Inter-American Development Bank is in a privileged position to build a high-quality information base on poverty and inequality for the countries of the region. The cornerstone of such an information base are the improved surveys funded under the ISLC/MECOVI program. However, so far this program is only operating in four countries. Some of the poorest countries in the region such as Guatemala, Guyana and Haiti have no surveys and their current institutional capability to produce ongoing surveys to measure the evolution of poverty and its determinants is probably weak. Others, such as Bolivia, have surveys which exclude the rural areas. Furthermore, throughout the region, non-sampling errors make estimates of poverty and inequality and their trends highly questionable, as revealed by the large discrepancies found in poverty estimates that use the same survey and the same poverty line. A lot more effort on the part of countries and the Bank is necessary in this area to generate reliable information on what is deemed one of the central concerns of both.

In addition, the Bank is also in an ideal position to be a central archiver of information and analysis on poverty-related issues. A step in that direction has been taken with the creation of the databank of household surveys as part of the ISLC/MECOVI program. However, there is an abundance of other information, available from country analytical work, project feasibility studies, and ongoing evaluation activities which, when taken together, could yield a very comprehensive picture of poverty and inequality in the region, and the Bank's role in reducing poverty and improving income distribution.

Improving Impact Evaluation Systems

Project preparation should not take place in a vacuum. Ideally, each new program and policy reflects past experiences, both failures and successes, and embodies the latest advances in the relevant field. Although Bank-financed anti-poverty programs may face added burdens during implementation, due to an extra layer of bureaucratic procedures, they presumably can compensate for these costs by benefiting from outside experience and expertise, the "lessons learned," from similar projects in other countries. Bank-funded projects can offer further value-added to the borrowing counterpart agencies by financing management information systems and monitoring and evaluation components. These information systems can go beyond the provision of findings for the specific activities financed by the Bank, and can also serve to evaluate other programs and projects, current and future, of the counterpart agencies. During the design phase of new operations, (or in other forums, such as INDES, the Research Network, the Social Policy Dialogue, etc.), the Bank can serve as a catalytic force for its clients and a channel for new ideas, stimulating policy debate and discussion. The Bank can help its counterpart institutions learn better from their own and each other's experiences, through funding for horizontal exchanges across projects, regional technical cooperations, and in the financing of preliminary studies and pilot phases.

In order to learn from its experiences, the Bank needs to know what has worked in the past, and what has not, at reducing poverty. Unfortunately, the answer to the question of whether, and to what extent, IDB projects are actually reducing poverty remains elusive. Poverty reduction effects are certainly intended, hoped for, and stipulated as project objectives, but more often than not the actual impact on the poor is not measured. With the introduction of the new project performance reporting system in recent years, this trend is changing, as all loans in execution are required to report regularly on a series of performance indicators. Nonetheless, a great deal of work remains to be done to improve the know-how within the IDB for determining to what degree the poor are affected by Bank operations, and specifically, to what extent projects can contribute to raising the poor out of poverty.

While many projects now have some kind of benchmarks and performance indicators, cases are rare where operations actually attempt to measure poverty reduction impact. Some projects which have given more attention to the measurement of poverty impact use surveys to establish base line data for the "before project" conditions, and follow up with "after project" surveys to measure the socioeconomic impact. This approach, while an improvement on non-direct measures of poverty, is still far from ideal in that it does not control for the many factors which influence the selection of beneficiaries of the program at hand. The most technically valid evaluations would include some use of a control group, or matched comparisons, in order to get an unbiased view of the effect of the program, which could separate from the many intervening factors which determine the selection of beneficiaries.

It is neither cost-effective, nor pragmatic in terms of implementation capacity, for every Bank project to include a full-fledged control-based impact evaluation scheme. However, it is important that some projects (in key areas where the Bank is investing heavily) go the distance to carry out a rigorous evaluation. In countries where the political will, and informational, financial, and implementation capacity exist, more rigorous evaluations can be carried out. Dissemination of these results can inform policy dialogue and project design in other countries, and contribute to the building of an institutional knowledge base regarding which interventions are most cost-effective in reducing poverty.

References

Birdsall, N., and J. L. Londoño. 1998. No Tradeoff: Efficient Growth Via More Equal Human Capital Accumulation in Latin America. Washington D.C.: Inter-American Development Bank. Mimeo.

______. 1997. Asset Inequality Does Matter: Lessons from Latin America. American Economic Review 87 (2).

______, D. Ross, and R. Sabot. 1995. Inequality and Growth Reconsidered. World Bank Economic Review 9 (3).

Cortázar, R. 1997. Chile: The Evolution and Reform of the Labor Market. In Labor Markets in Latin America, Combining Social Protection with Market Flexibility, e's. Sebastian E. and N. C. Lustig. Washington D.C.: The Brookings Institution.

______, N. Lustig, and R. Sabot. 1998. Economic Policy and Labor Market Dynamics in Latin America. In Efficiency and Equity Enhancing Reforms in Latin America, e's. N. Birdsall, R. Sabot, and C. Graham. Washington D.C.: Inter-American Development Bank. Mimeo.

Deutsch, R. 1998. Does Child Care Pay?: Labor Market Participation and Earnings Effects of Access to Child Care in the Favelas of Rio de Janeiro. Washington D.C.: Inter-American Development Bank, Poverty and Inequality Advisory Unit. Mimeo.

______, and A. Verdisco. 1997. Learning Lessons from the IDB's Portfolio of Primary and Secondary Education Loans in Execution. Washington D.C.: Inter-American Development Bank, Social Programs Division. Mimeo.

Duryea, S., and M. Székely. 1998. Labor Markets in Latin America: A Supply-Side Story. Paper Prepared for the IDB/IIC Annual Meeting, The Employment Problem in Latin America: Roots and Remedies. Cartagena de Indias, Colombia.

Economic Commission for Latin America and the Caribbean. 1997a. Preliminary Overview of the Economy of Latin America and the Caribbean. Santiago de Chile: Economic Commission for Latin America and the Caribbean, United Nations.

Economic Commission for Latin America and the Caribbean. 1997b. The Equity Gap, Latin America and the Caribbean and the Social Summit. Santiago de Chile: Economic Commission for Latin America and the Caribbean, United Nations.

Flug, Karnit, A. Spilimbergo, and E. Wachtenheim. 1998. Investment in Education: Do Economic Volatility and Credit Constraints Matter? Journal of Development Economics. Mimeo.

Gulli, W., and Pollack. 1998. Possibilities and Limitations of Poverty Reduction Through Microfinance: Conceptual Framework and Preliminary Findings. Washington D.C.: Microenterprise Unit, Inter-American Development Bank.

Hanson, G. H., and A. Harrison. 1995. Trade, Technology, and Wage Inequality. NBER Working Paper No. 5110. Cambridge, Massachusetts: National Bureau of Economic Research.

Lipton, M., and M. Ravallion. 1995. Poverty and Policy. In Handbook of Development Economics, Volume III, eds. J. Behram and T. N. Srinivasan.

Londoño, J. L., A. Spilimbergo, and M. Székely. 1997. Income Distribution, Factor Endowments and Trade Openness. Working Paper Series 356, Washington, D.C.: Office of the Chief Economist, Inter-American Development Bank.

Lora, E., and F. Barrera. 1997. A Decade of Structural Reforms in Latin America: Growth, Productivity and Investments are Not What They Used to Be. In Latin America After a Decade of Reforms: What Comes Next, eds. R. Hausmann and E. Lora. Washington D.C.: Inter-American Development Bank.

______, and M. L. Henao. 1997. Colombia: The Evolution and Reform of the Labor Market. In Labor Markets in Latin America, Combining Social Protection with Market Flexibility, eds. S. Edwards and N. Lustig. Washington D.C.: The Brookings Institution.

Lustig, N., and D. Mc Leod. 1997. Minimum Wages and Poverty in Developing Countries: Some Empirical Evidence. In Labor Markets in Latin America, Combining Social Protection with Market Flexibility, eds. S. Edwards and N. Lustig. Washington D.C.: The Brookings Institution.

Márquez Mosconi, G., and C. Alvarez. 1996. Poverty and the Labor Market in Venezuela: 1982-1995. Working Paper SOC96-101. Washington D.C.: Inter-American Development Bank.

______, and C. Pagés. 1998. Ties that Bind: Employment Protection and Labor Market Outcomes in Latin America. Washington D.C.: Inter-American Development Bank. Mimeo.

Morán, and Castro. 1997. Street-children and the Inter-American Development Bank: Lessons From Brazil. Discussion Paper, SDS/SOC. Washington D.C.: Inter-American Development Bank.

Morrison, A., and M. B. Orlando. 1998. The economic impact of domestic violence in Chile and Nicaragua: Labor force participation, earnings, service utilization and children's educational performance. Washington D.C.: Women in Development Unit, Inter-American Development Bank. Mimeo

Pessino, C. 1997. Argentina: The Labor Market During the Economic Transition. In Labor Markets in Latin America, Combining Social Protection with Market Flexibility, eds. S. Edwards and N. Lustig. Washington D.C.: The Brookings Institution.

Pissarides, C. A. 1997. Learning by Trading and the Returns to Human Capital in Developing Countries. The World Bank Economic Review 11 (1): 17-32.

Robbins, D. 1996. HOS Hits Facts: Facts Win, Evidence on Trade and Wages in the Developing World. Cambridge, Massachusetts: Harvard Institute for International Development. Mimeo.

Sustainable Development Department (SDS)/Environment Division (ENV). 1998. Strategy for Reduction of Rural Poverty. A Bank Strategy Paper, subject to Board approval. Washington D.C.: Inter-American Development Bank. Mimeo.

______, / Poverty and Inequality Advisory Unit (POV). 1998. The Use of Social Investment Funds as an Instrument for Combating Poverty. A Bank Strategy Paper, subject to Board approval. Washington D.C.: Inter-American Development Bank. Mimeo.

United Nations. 1997. Trade and Development Report, 1997. New York:United Nations.

Wood, A. 1997. Opennes and Wage Inequality in Developing Countries: The Latin American Challenge to East Asian Wisdom. The World Bank Economic Review 11 (1): 33-57.

World Bank. 1997. Poverty and Income Distribution in a High-Growth Economy, Chile: 1987-1995. Report No. 16377-CH, Washington D.C.: The World Bank.

1. Measured at 1985 purchasing parity prices.

2. Some have no data at all.

3. These countries are: Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Ecuador, El Salvador, Mexico, Paraguay, Peru, and Venezuela. Summary bibliographic information on these studies is provided in Annex 1.

4. Poverty here refers to moderate poverty and is measured by the proportion of individuals living below a country-specific poverty line, also known as the headcount ratio. The moderate poverty lines are in Annex 2.

5. See studies cited in Lipton and Ravallion (1995), p. 2603. Interestingly, the lowest elasticity for a sample of eight countries from different regions was found for Brazil (-1.5). These studies assume a distribution neutral increase in per capita consumption. Poverty in these exercises is measured by the poverty gap.

6. Figures are from ECLAC (1997a). They do not include the Caribbean countries.

7. Figure for 1996.

8. ECLAC (1997b).

9. ECLAC (1997b).

10. According to ECLAC (1997b), the gap widens in Bolivia, Brazil, Chile, Colombia, Mexico, Paraguay, Uruguay and Venezuela, stays almost constant in Costa Rica, and narrows in Honduras.

11. ECLAC (1997b), Table II-5.

12. Duryea and Székely (1998).

13. Duryea and Székely op.cit.

14. Lora and Barrera (1997), Hanson and Harrison (1995), Robbins (1996), Pissarides (1997), and Wood (1997).

15. See Londoño, Székely and Spilimbergo (1997) and Wood (1997).

16. Hanson and Harrison (1995).