Draft Regional Hdr

PLEASE DO NOT CIRCULATE

Project Document for the Regional Report UNDP- LAC 2008/2009 “Human Development and the inter-generational transmission of inequality” Luis Felipe López Calva (RBLAC-UNDP) and Isidro Soloaga (El Colegio de México) 1 This version, June 10 2008

Objectives The Human Development Regional Report for LAC (hereafter the “Report”) has the following objectives: (a) to document current trends in inequality in LAC countries; (b) to incorporate inequality issues into HDI measures; (c) to estimate the determinants of the inter-generational transmission of inequality; (d) to identify the most important binding constraints in education, health and labor markets participation for breaking with the intergenerational transmission of inequality, and (e) to link LAC countries´ extreme inequality to main political economy outcomes. This Report intends to build upon previous studies and, after documenting recent developments of inequality in LAC countries, make progress toward our understanding of two main issues: a) the mechanisms by which inequality is inter-generationaly transmitted and equality of opportunity in enhanced, b) the mapping between LAC countries´ extreme inequality and the region’s economic policy design and economic policy outcomes. 1

We thanks inputs provided by Marcelo Delajara and David Mayer, comments received from participants at the Research Conference on the Intra- and Inter-generational Transmission of Inequality (El Colegio de Mexico, February 28 and 29, 2008) as well as discussions with Francisco Ferreyra and Jaime Saavedra at the beginning of this project. Florian Wendelspiess and Monica Villanueva provided helpfull assisstance along the preparation of this project document.

1

In particular, the report will focus on the less researched area of identifying the most important binding constraints in education, health and labor markets for breaking with the intergenerational transmission of inequality and for enhancing the equality of opportunity.

Main Research Questions The Report will deepen our knowledge and understanding of the established fact of a high and persistent level of inequality in LAC countries. To achieve these goals, the Report will first document recent developments of inequality in LAC countries, will develop an inequality-sensitive HDI, and will address, within a Human Development Approach the following questions: 1)

What are the main determinants for the inter-generational transmission of inequality?

2)

Which are the main determinants for the enhancement of the equality of opportunity in LAC countries?

3)

Can we identify the main binding constraints that impede the poor to accumulate human capital (e.g., education and health) and use this knowledge to derive policy implications?

4)

Is there a mapping between LAC countries´ extreme inequality and the region’s economic policy design and economic policy outcomes?

Methodological Approach 1) Current trends in inequality in LAC countries. This section of the Report will document current inequality trends in education, health and income in LAC countries. It will include the following sections: 1.1 Income inequality since the 1990s 1.2 Labor and non labor income inequality 1.3 Regional inequality and LAC in world perspective 1.4 Inequality in education 1.5 Inequality in health 1.6 Inequality in infrastructure and services 1.7 Group-based inequalities: gender 1.8 Group-based inequalities: ethnicity 1.9 Group-based inequalities: race 2) Current inequality dynamics in LAC countries.

2

This section of the Report will show the importance of human capital in the transmission of inequality, including the analysis of income, occupation, and education mobility, as well as of chronic poverty and polarization. 2.1 The issues. The actual mobility observed in a society is the result of both the arrival of opportunity (stochastic technical change, alternations in policy, etc.) and the relative incentives to seize those opportunities (Genicot and Ray, 2008). This section will review different concepts of mobility and will provide research on the study its determinants. In particular, on people’s ability to seize a new opportunity when it presents itself (e.g., through migration, schooling investment, occupational change, or technology adoption) and on determining in which way individual's incentives to seize an opportunity depend on her relative position in the distribution. 2.2 Intra-generational mobility. Fields et al. (2007). 2.3 Inter-generational mobility. Behrman et al. (2001), IADB (2007) Solon (2004) 2.4 Chronic poverty. 2.5 Polarization Cruces & Gasparini (2008) 2.6 An inequality-sensitive Human Development Index (Foster, López Calva & Szekely, 2003) 3) Which are the main determinants for the enhancement of the equality of opportunity in LAC countries? 3.1 The issue. Although the level of economic development of a country is measured in general by its per capita income, several authors have expressed concerns that this is an imperfect proxy for human capacities or the standard of living. For instance, Amartya Sen´s approach for the evaluation of individual welfare and social states, assesses people’s welfare in terms of their functioning’s and capabilities, which are defined as an individual’s actual and potential activities and states of being, respectively. Other authors have challenge also the notion of income per capita from the point of view that economic development should be measure by the extent to which a society equalizes opportunities. (World Bank, 2008), (Paes de Barro and Carvalho, 2007) (Roemer—2006, World Development Report—2006, ). The Report will build on the findings from the above reviewed literature and extend the analysis in order to have a greater understanding of the determinants of the conditionings. Furthermore, the Report will structure those determinants in a framework

3

that takes into account the binding restrictions for breaking with the intergenerational transmission of inequality. To the best of our understanding these issues have not yet been addressed within this conceptual framework in applied work. It will thus be part of the Report to develop such a framework to do the empirical work on these matters. 3.2 Inequality of life chances. This part of the Report will concentrate the analysis on the concept of inequality of life chances and measure it via two indices of social immobility derived from the Theil and Gini indices of inequality and via the concept of a social immobility curve. The section will also look at equality of opportunity. A distinction will be made between various kinds of equality of opportunity, the analysis being always based on the comparison of the distribution of (children's) incomes for various educational or income levels of the fathers (various types or circumstances). The main tool of analysis will be the concepts of first and second order stochastic dominance. For the overall comparison of regions or countries, use will be made of the concept of sequential stochastic dominance. The paper will use the Latinobarómetro survey which provides information for several LAC countries on the educational levels of both "father" and "children", on the occupation of the "child" and on the income class (on a scale from 1 to 10) to which the "child" thinks he belongs and on that to which he thinks his father belonged. This survey provides also information on the ownership of durable goods that will proxy the economic wealth of the families. 3.3 Disentangling the impact of abilities and family background on inequality. This section will develop a structural model of income inequality that explicitly considers the direct and indirect mechanisms by which parents’ human capital and individual’s abilities affect both schooling decisions and labor market outcomes. The model will be useful to analyze the relative role of inherited individual’s characteristics (circumstances) and endogenous decision (opportunities) in explaining income inequality, using rich data from Chile. The structural model of income inequality will explicitly consider the direct and indirect mechanisms by which parents’ human capital and individual’s abilities affect both schooling decisions and labor market outcomes, and consequently, income distribution. The empirical implementation will assess to what extent the high inequality in labor market outcomes in Chile is due to (i) unequal opportunities that individual inherit

4

from their parents (parents’ human capital and family background), (ii) early developmental conditions (abilities), and (iii) heterogeneity in schooling investments and their return. Thus, the empirical model explicitly will consider both schooling decisions and labor market outcomes as endogenously determined, as well as unobserved heterogeneity. These elements represent a significant departure from the previous literature, since schooling decisions and labor market outcomes are likely affected by the same unobserved components. Furthermore, following recent econometric advances, this section will use a methodology that allows linking these unobserved components to cognitive and noncognitive abilities (Heckman, et al. (2006); Urzua (2008)). The distribution of these latent abilities will be extracted by estimating educational production functions that link test scores to observable characteristics and to unobservable cognitive and non-cognitive factors. Importantly, since identifying the main binding constraints is one of the main focus of this Report, this part will incorporate explicitly supply-side factors into the analysis to explain individuals’ schooling decisions. This will be done by incorporating proxy variables of primary and secondary schooling accessibility and college proximity into the schooling decision model. The controls will be: (i) the number of primary schools per children at the local level (municipalities) before 1999; (ii) the number of secondary schools per children at the local level before 1999; and (iii) the number of post-secondary institutions per person at the local level before 1999. This will capture different schooling costs (rural vs. urban). 3.4 Contrasting the equality of opportunity and the human development approaches. Since the Human Development approach for the breaking of the intergenerational transmission of inequality differs from that of the Equality of Opportunity approach, this section of the Report will discuss and contrast these two approaches. 4. - Identifying the binding constraints and the policy recommendations for breaking with the intergenerational transmission of inequality. 4a. The issues. A recent paper by Devarajan and Kambur2 sketched a framework for thinking about the policy challenge of scaling up small scale interventions that address poverty reduction successfully. Central in their approach is the identification of the roots of poverty: poor people are poor because markets fail them, and governments fail them. For 2

Devarajan and Kambur, 2005.

5

instance, market failures in the capital markets prevent poor people from getting adequate education. Often, when governments step in to “correct” the problem it introduces its own problems, such as capture of the public resources by the non-poor, or the inability to monitor service providers, which leave the poor still poor. Binding constraints for breaking the intergenerational transmission of inequality have been also found in the use of public services--including health, water and sanitation, and credit markets. This section 4 of the Report will develop a framework for the analysis of these and other issues and will also identify plausible applications that would elicit policy recommendations. 4b. A conceptual framework. In the context of the literature on economic growth, Hausmann et al., 2005 (HRV hereafter) indicated that the performance of policy reforms that are potentially growth-promoting is heavily dependent on circumstances: policies that work wonders in some places may have weak, unintended, or negative effects in others. In this context, good policy recommendations would come only after we develop a better understanding of how the binding constraints on economic activity differ from setting to setting. Crucial in turning general principles into operational policies is the knowledge of local specificities. Moreover, since there always are administrative and political limitations, policy recommendations should include a sense of priorities3. The methodology proposed by HRV is to go sequentially in locating the distortion(s) with the largest potential impact on economic growth. In this way, after a few steps, the binding constraints are identified-see figure 1 in HRV for a decision tree that conceptualizes this methodology, in a way that enables the design of remedies that are as closely targeted as possible. In this chapter we will first translate HRV´s methodology—designed to economic growth diagnostics-- to the analysis of the determinants of poor people’s education, health and labor income outcomes. For instance, what keeps schooling at relatively low levels in poor people (compared to better-off people)? Is it because poor families demand less education? Or is it because there are no school facilities available for these families? If it is 3

In words of the authors: “What we propose to do in this paper is to develop a framework for growth diagnostics—that is, a strategy for figuring out the policy priorities. The strategy is aimeed at identifying the most binding constraints on economic activity, and hence the set of policies that, once targeted on these constraints at any point in time, is likely to provide the biggest bang for the reform buck” (Hausmann et al, 2005).

6

the first case (low demand by poor families), is it because returns to schoolings are inadequate? Or rather the cause is credit constraints that make poor families unable to make the investment? And so on. Clearly, policy recommendations will be totally different for the case of a demand-side binding constraint than in one where the binding constraint is on the supply-side. For instance, very little in terms of schooling will be gained by building more schools, if low schoolings levels in poor people is due to perceived low returns to human capital or to credit constraints. This Chapter will lay out a conceptual framework that will link the observed problem that favors the intergenerational transmission of inequality (low schooling, high child mortality rates, etc.) to plausible “binding constraints”. In turn, these “binding constraints” will be mapped to the nature of the evidence needed to make the assessments, as well as the empirical implementation of econometric models to elicit information. Candidates for the latter are the concentration index approach as well as education and health production functions4. 4c . Concentration Index approach: health and cognitive abilities 5. A way to study the binding constraints that affect the accumulation of health capital is to study inequalities in health using some type of decomposition analysis (Mayer Foulkes and Larrea, 2005), Wagstaff and van Doorslaer (2000b) (WV) 4.c.1 Binding constraints on the demand side. Our main dependent variable is an index for Health Service Use. An additional dependent variable that can be considered for its own interest and for comparison is child height for age z score. 4.c.2 Binding constraints on the supply side. Amongst the variables we can consider as determinants of supply are the means of the indices mentioned above, by geographic locations for which the survey is representative, as well as average local literacy and assistance to primary, secondary and higher schooling. 4d..Education and health production functions approach. Another way to elicit which are the main binding constraints that affect the accumulation of health and schooling will be to estimate education and health reduce-form households’ demand and production functions. For the case of cognitive achievement, Todd and Wolpin (2003) develop a general 4

See Tood and Wolpin (2003) for a recent discussion for the case of education production functions and Grossman (2004) and Schultz (2004) for the case of health production functions. 5 Inputs this section were provided by David Mayer Foulkes.

7

conceptual framework for estimating the relationship between achievement outcomes and family and schooling inputs and consider how to implement that framework under different kinds of data limitations6. Although less formally, Schultz (2004) discusses similar issues for the case of health economics. Glewwe et al. (2005), develop a common approach to estimate education and health production functions in a series of LAC countries. 4.d.1 Education production functions. Glewwe and Kramer’s (2005). This Report will identify suitable data sets in LAC countries where education production functions could be estimated. We expect to commission at least 5 country-case studies. 4.d.2 Infant health production functions. Grossman (2004) and Schultz (2004). 4.d.2.1 Health production and household behavior. The health production function (HPF) is a conceptual framework which allows us to estimate demand and supply factors affecting infant health outcomes. The basic assumption is that health and infant health in particular is produced at the level of each individual and that there is a production technology by which health inputs are combined to produce health outputs in the context of the household. 4.d.2.2 Estimating the determinants of health outcomes. Rosenzweig and Schultz (1983) Empirical attempts to estimate the determinants of child health rely mainly in three models: Structural models. In structural specification effort is made to distinguish between direct (or immediate) and indirect (or underlying) determinants of health within the HPF. Reduced-form models In these models the biological-HPF is of no interest, and health outcomes are directly linked to the vectors of underlying determinants Xh and Xc, child characteristics Xi, and the vectors of prices and incomes. 4.d.2.3 Issues of specification and data availability. Granted this advantage, estimating health production functions is not without serious problems of i) specification (Rosenzweig and Schultz, 1983), ii) availability and quality of instrumental variables –the policy variables themselves- (Schultz, 2004), and iii) the qualitative nature of some of the health input variables (Delajara, Bertranou, and Amiune, 2007). For the problems (i) and (iii) there are known solutions, fortunately. It is problem (ii) the most difficult to solve in the context of developing countries. Although many countries now have regular household surveys which provide information on important instruments measured at the household level - such as parental education and employment status, family income, household infrastructure, and access to health insurance, data on instruments 6

See Glewwe, et al (2004) for a discussion of these issues.

8

measured at the community level are still unavailable in many countries; sub-national statistics on relevant prices, socio-economic conditions, and health service provision are still scarce. 4.d.2.4 Infant health and intergenerational mobility in LAC. There are very few examples of studies that aim to estimate infant health production in LAC. We seek to estimate the production function of infant health outcomes that play an important role in the subsequent acquisition of human capital and thus determine the long-run educational and social achievements of children; prominent among these early child health outcomes are birthweight and height-for-age. 4.d.2.5 LAC data. We have identified seven Latin American countries for which we could estimate infant health production functions using comparable household data. These are last rounds of the DHS for Bolivia (2003), Colombia (2005), Dominican Republic (2002), Haiti (2000), Honduras (2005), Nicaragua (2001), and Peru (2000). Thus, comparable measures of health output, health inputs, and household level instrumental variables are available for all these countries. Regarding aggregate level instruments, we have a limited set of comparable data for all these countries, except Haiti, from national health statistics; and an expanded set of comparable aggregate data for Bolivia, Nicaragua, and Peru from the same source. 4.d.3 Education quantity vs. education quantity. A well-known stylized fact is that while Latin America has made great progress in terms of improving educational quantity (measured for instance in terms of enrollment rates), progress in reaching adequate quality (measured for instance by standardized international tests) seems to be distinctly lacking (Vegas and Petrow, 2007). In turn,

Urquiola and Calderon (2006) rely on arguably

comparable household survey data to provide comparative evidence on: i) how effective different countries’ educational systems are at getting children into school, and ii) how successful they are in terms of turning this contact there into years of schooling. This section will complement previous research by incorporating education quality into the analysis. Using data from testing programs for countries in the region, the paper will make an assessment on how educational quality maps into quantity-related performance. 5) What political economy issues are explained by LACs’ extreme inequality levels?

9

A recent World Bank study inquired into the historical roots of inequality in LAC7. The main conclusion from this study is that the causes and determinants of high and persistent inequality in LAC have their origin in the early period of colonization and the subsequent evolution of interactions among economic, political and socio-cultural factors. Moreover, the study stresses that institutions in LAC countries were shaped largely by the factor endowment the Europeans found in Central and South America, rather than the nature of the colonial powers themselves. The study also considers some of the political and socio-cultural mechanisms through which states in the region were controlled by elites, failing to meet the needs of the poor or other specific groups, finding two main types of explanations. First, the macro institutional factors, which are the ways in which subordinate groups have been incorporated into local and national polities, with a predominance of vertical forms of political connection through personal, clientelistic, and corporatist structures (in contrast to the more horizontal alliances found in Europe, for instance). Second, socio-cultural processes tend to result in practices, dispositions, and attitudes on the part of both the dominant and subordinate groups that generate unequal relationships that complement inequalities in economic and political structures. The Report will seek to review these issues from a political science and a sociological point of view. We see that causality goes both ways between inequality levels and a given policy outcome: inequality determines the political economy, the political economy determines inequality. Aiming at improving our understanding on how important are political economy issues and in the way they operate to foster inequality in LAC, we plan to address this question mostly in one direction: given current inequality levels in LAC, What economic policy and economic outcome shall we expect?.

7

De Ferranti, et al., 2003.

10

Outline Human Development and the inter-generational transmission of inequality 1.-Introduction Luis Felipe López Calva & Isidro Soloaga This section will address general issues on inequality, including: a) Different approaches to the topic (UNDP, WB, IADB, Political Economy, etc.). b) The need to focus new research on the inter-generational transmission of inequality. c) The need to link inequality issues to the functioning of the government, markets, and civil society. d) The usefulness of identifying main binding constraints for breaking the intergenerational transmission of inequality to derive policy implications. 2.-Dimensions of Inequality James Foster, Luis Felipe López Calva & Isidro Soloaga This section will address with some detail the different approaches for the study of inequality. 2.1 Non Human-Development Approach Isidro Soloaga 2.1.1 Income & Consumption Inequality 2.1.2 Income dynamics 2.1.2 Inequality of Opportunities 2.1.3 Inequality in the political arena 2.1.4 _____ 2.2 Human-Development Approach Luis Felipe López Calva 2.3 Inequality-Sensitive Human Development Index James Foster 3.-Recent developments on Inequality in LAC countries Guillermo Cruces(GC), James Foster (JF), and Leonardo Gasparini (LG) This Chapter will summarize information for Latin America on many dimensions of inequality. It will be highlighted here some of the main features of inequality in the region in order to help readers understand the level, structure, and trends related to this phenomenon. 3.1 Income inequality since the 1990s GC & LG 3.2 Labor and non labor income inequality GC & LG 3.3 Regional inequality and LAC in world perspective GC & LG 3.4 Inequality in education GC & LG 3.5 Inequality in health GC & LG 3.6 Inequality in infrastructure and services GC & LG 3.7 Group-based inequalities: gender GC & LG 3.8 Group-based inequalities: ethnicity GC & LG 3.9 Group-based inequalities: race GC & LG 3.10 Inequality Sensitive HDI. JF

11

4.3 Chronic poverty This sub-chapter will review current concepts of chronic poverty and will review the evidence of the importance of chronic poverty in LAC countries. Chronic poverty—or persistent impoverishment-- is not only a symptom of past deprivation, it is also the cause of future destitution of persistency of inequality. For this Report, the distinguishing feature of chronic poverty is extended duration in absolute poverty. Therefore, chronically poor people always, or usually, live below a poverty line, which is normally defined in terms of a money indicator (e.g. consumption, income, etc.), but could also be defined in terms of wider or subjective aspects of deprivation. This is different from the transitorily poor, who move in and out of poverty, or only occasionally fall below. Chronic poverty can be also defined as that poverty experienced by individuals and households for extended periods of time or throughout their entire lives8. Sections of this sub-chapter include: 4.3.1 The concept of chronic poverty and its relationship with inequality issues. 4.3.2 Available estimates for LAC and non-LAC developing countries. 5.-Income, education and occupational dynamics Dante Contreras, Gary Fields, Garance Genicot, Hugo Ñopo, Debraj Ray, Florencia Torche,. An important concept related to inequality is that of mobility (e.g., income, education, or occupation mobility). This Chapter will firstly summarize definitions and measures of income dynamics, and will review the available evidence for LAC countries on intra-generational income mobility. Secondly, following a strand of the sociologic literature that studies education and occupational intergenerational dynamics, the Chapter will include explicitly the sociological view of intergenerational mobility, in particular that referred to education levels and occupations, and will relate this approach to that of economics. Finally, this Chapter will also consider that aspect of mobility determined by individuals´ ability to seize new opportunities when they present themselves (e.g., through migration, schooling investment, occupational change, or technology adoption) and the determinants of this ability (e.g., relative position in the income distribution). 5.1.- Intra-generational Mobility 5.1.1.- Mobility (in general) Gary Fields 5.1.2.- Group Mobility 5.1.2.1.- Exogenous (gender, ethnicity) 5.1.2.2.- Endogenous (location, sector of occupation) 5.2 Pseudo panel Hugo Ñopo 5.3 Intergeneracional Florencia Torche & Dante Contreras 5.3.1 Review of sociological approaches 5.3.2 Occupational 5.3.3 Socioeconomic status 5.3.3 Sociology’s and Economics point of view on the transmission of inequality: a comparative analysis. 5.3.4 Empirical evidence for LAC countries. 5.3.5 An application to the case of Chile Dante Contreras 5.4 New opportunities Debraj Ray and Garance Genicot 8

See for instance, The Chronic Poverty Report, Chronic Poverty Research Center, 2007.

12

5.4.1 Accessibility to new opportunities and its determinants 6.-Inequality of opportunities approach Jeremie Gignoux Joseph Gaush, Julio Guzman, Jacques Silber & Sergio Urzúa 6.1 1 Educational expansion and equality of earnings opportunities. Jeremie Gignoux 6.1.2 Application to the case of Brazil, 6.1.3 Application to the case of Colombia 6.1.4 Application to the case of Ecuador. 6.1.5 Application to the case of Panamá, 6.2 Inequality of life chances. Applications to several LAC countries Jacques Silber-Joseph Gaush 6.3 Inequality of opportunity controlling for ability. An application to the case of Chile Julio Guzman & Sergio Urzúa 7.-Concentration Index approach: health and cognitives abilities David Mayer Foulkes 7.1 1 Health Concentration Index 7.1.2 Application to the case of Bolivia 7.1.3 Application to the case of Brazil 7.1.4 Application to the case of Guatemala 7.1.5 Application to the case of Peru 7.2.1 Cognitives abilities Concentration Index 7.2.1 Application to the case of Mexico 8.-Identifying the most important binding constraints in poor households´ health, education and labor markets participation for breaking with the intergenerational transmission of inequality: 8.1 A framework for the analysis of binding constraints Ravi Kanbur 8.1 Education production functions (review articles by Miguel Urquiola and by Isidro Soloaga) Note: These ongoing review articles are taking stock of the wealth of work done on education production functions aiming at proposing a common approach to be applied in several LAC countries. It is expected that at least five papers are going to be commissioned to elicit policy implications using this approach. 8.2 Health (review article by Marcelo Delajara) Note: This ongoing review article is taking stock of the wealth of work done on health production functions and aims at proposing a common approach to be applied in several LAC countries. A pilot study for child health for the case of Bolivia is currently ongoing. This pilot study seeks to assess the feasibility of applying this health production function approach to data available in LAC countries. If deemed useful for the purpose of this Report, it is expected that at leas five papers are going to be commissioned to elicit policy implications using this approach. 8.3 Labor Market participation 9. Political economy issues derived from the relatively high inequality levels of LAC countries. 9.1 Review of what is already known

13

9.2 Is there a particular type of economic policies that could be attributable to high inequality levels?. 10. Review of findings Marcelo Delajara, Luis Felipe López Calva, David Mayer Foulkes & Isidro Soloaga 10.1-Findings: binding constraints in education 10.2-Findings: binding constraints in health 10.3-Findings: binding constraints in labor market participation policies. 10.4-Findings: the importance of sex and ethnicity as cross-cutting issues. 10.5-Findings: binding constraints in the political economy 11.- Discussion and policy recommendations. Luis Felipe López Calva & Isidro Soloaga

14

References Anand, Sudhir and Amartya K. Sen. (1993) "Human Development Index: Methodology and Measurement." Human Development Report Office. Occasional papers 12. UNDP, New York. http://www.undp.org/hdro/papers/ocpapers/oc12.pdf Behrman , J.R., A. Gaviria, and M. Szekely. 2001. Intergenerational Mobility in Latin America. Research department Working Paper #452 Inter-American Development Bank CEPAL (Comision Económica para América Latina y el Caribe). 2006. Panorama Social de América Latina 2006. Santiago:CEPAL: De Ferranti, D, G. E. Perry, G.H.G. Ferreira, and M. Walton (2003). Inequality in Latin America and the Caribbean. Breaking with History?. World Bank. Washington DC. Delajara, M. (1999). “The Measurement of Intergenerational Inequality. A Sourcesof-Growth Approach”; mimeo, Universidad Nacional de Tucumán, Argentina. Presented at the LACEA annual meeting, Santiago de Chile, 1999. Devarajan, S and R. Kanpur (2005) "A Framework for Scaling Up Poverty Reduction, With Illustrations from South Asia", mimeo. Available at http://www.arts.cornell.edu/poverty/kanbur/DevarajanKanburAug05.pdf Eckstein, Zvi and K. I. Wolpin (1999) “Why youth drop out of high school: the impact of preferences, opportunities and abilities”. Econometrica. Vol 67, No 6 (November), pp-1295-1339. Esteban J, and Debraj Ray (1994) “On the Measurement of Polarization. Econometrica, Vol. 62, No. 4, (Jul.), pp. 819-851 Fields, G.S., R..Duval Hernández, S. Freije Rodríguez, and M. L. Sánchez Puerta, 2007. “Intragenerational Income Mobility in Latin America”. Economía. Spring 2007. Forthcoming. Foster, James E., Luis F. López-Calva and Miguel Székely (2003) “ Measuring the Distribution of Human Development: Methodology and an Application to Mexico” and other background papers to the 2002 Report http://www.undp.org.mx/desarrollohumano/estudiosfr.html García Gómez, P. and A. López Nicolás (2006). “The evolution of inequity in the access to health care in Spain:1987-2001”, Working Paper, Universitat Pompeu Fabra, Barcelona. Gaviria, A. (2006) “Movilidad Social y preferencias por redistribución en América Latina”. Documento CEDE 2006-03. Gaviria, A. (2007) “Social mobility and preferences for redistribution in Latin America”. Economia-Journal of the Latin American and the Caribbean Economic Association. Vol.8, Number 1. pp.55-96. Genicot , G and D. Ray.(2008). “Mobility: outline”. Mimeo. NYU Glewwe, Paul, Michael Kremer, Sylvie Moulin and Eric Zitzewitz (2004) “Retrospective vs. Prospective Analyses of School Inputs: The Case of Flip Charts in Kenya,” Journal of Development Economics, 74: 251-268

15

Glewwe, Paul (2005) “The Impact of Child Health and Nutrition on Education in Developing Countries: Theory, Econometric Issues and Recent Empirical Evidence.” Food and Nutritional Bulletin. Grossman, M., (2004) “The demand for health, 30 years later: a very personal retrospective and prospective reflection” Journal of Health Economics 23, pp. 629–636 Guzman, J. and S. Urzúa (2008). "Disentangling the Role of Nature and Nurture when Explaining Inequality". Manuscript, Northwestern University. Hausmann, R., Dani Rodrik and Andrés Velasco (2005) “Growth Diagnostics” Mimeo. Harvard University. Haveman, R and Barbara Wolfe (1995). “The Determinants of Children's Attainments: A Review of Methods and Findings.” Journal of Economic Literature, Vol. 33, No. 4. (Dec), pp. 1829-1878. Heckman, J., J. Stixrud, and S. Urzúa (2006). “The Effects of Cognitive and Noncognitive Abilities on Labor Market Outcomes and Social Behavior”. Journal of Labor Economics, 24(3). Hicks, D.A. (1997) “The Inequality Adjusted Human Development Index: A constructive proposal” World Development, 25, (8), 1283-1298. Inter American Development Bank, 2007. Outsiders? The Changing Patterns of Exclusion in Latin America and the Caribbean. IADB. Washington DC. Mayer Foulkes, D. and C. Larrea (2005) “Racial and ethnic disparities in health in Bolivia, Brazil, Guatemala, and Peru”, CIDE Working Paper, Mexico City. Mayer Foulkes, D. and C. Larrea (2005), “Racial and Ethnic Health Inequalities: Bolivia, Brazil, Guatemala, and Peru”, CIDE Working Paper 333, Mexico, DF. Paes de Barro, R. and M de Carvalho. “Measuring Inequality of Opportunity in Latin America and Caribbean “ (2007). Paper presented at LACEA Meetings in Bogota, Colombia. Roemer, J. (2006) “Economic Development as Opportunity Equalization.” Cowles Foundation Discussion Paper No. 1583 Sapelli, C and A. Torche (2004). “Deserción Escolar y Trabajo Juvenil: ¿Dos Caras de una Decisión?.” Cuadernos de Economía, Vol. 41, N 123, pp. 173-198. Schultz, T. Paul (2004). “Health economics and appliactions in developing countries”. Journal of Health Economics 23, pp.637-641 Solon, Gary (2004) “A Model of Intergenerational Mobility Variation over Time and Place,” in Miles Corak (ed.), Generational Income Mobility in North America and Europe, pp. 38-47, Cambridge: Cambridge University Press, 2004. Todd, Petra E. and Kenneth I.Wolpin (2003). “On the specification and estimation of the production function for cognitive achievement” The Economic Journal 113 (February), F3-F33. UNDP (United Nations Development Programme). 2007. Making Globalization Work for All. Annual Repor 2007. United Nations. NY. UNDP (United Nations Development Programme). 2005. Human Development Report 2005. International cooperation at a crossroads: Aid, trade and security in an unequal world. UNDP. New York. Urquiola, M and V. Calderón (2006). “Apples and oranges: Educational enrollment and attainment across countries in Latin America and the Caribbean“. International Journal of Educational Development 26, pp. 572–590.

16

Van Doorsaler E, Andrew J. (2002). “Inequalities in self-reported health: validation of a new approach to measurement”, mimeo. Van Doorslaer E, Wagstaff A, van der Burg H, et al. (2000). Equity in the delivery of health care in Europe and the US. Journal of Health Economics, Vol. 19, pp. 553583. Vegas, E. and J. Petrow (2007) Raising student learning in Latin America: The challenge for the 21st century. Washington, D.C.: The World Bank. World Bank (2008). Regional Study. Measuring Inequality of Opportunity in Latin America and the Caribbean. Latin American and the Caribbean Region. World Bank (200?). upcoming cross-country review of Conditional Cash Transfer Programs Wagstaff A, van Doorsaler E. Paci P. (1989). Equity in the finance and delivery of healthcare: some tentative cross country comparisons. Oxford Review of Economic Policy, Vol. 5(1), pp. 89-112. Wagstaff A, van Doorslaer E. (2000a), “Equity in health care finance and delivery,” in A.J. Culyer and J. P. Newhouse (Eds.) in Handbook of Health Economics, Vol 1B, Elsevier Science. Wagstaff A, van Doorslaer E. (2000b). Measuring and testing for inequity in the delivery of health care. Journal of Human Resources, Vol. 35(4), pp. 716-733. Wagstaff, A., E. Van Doorslaer, and N. Watanabe (2003). On decomposing the causes of health sector inequalities, with an application to malnutrition inequalities in Vietnam. Journal of Econometrics, 112(1): 219-27.

17

18