Report No. 32884-MX Mexico Income Generation and Social Protection for the Poor (In Four Volumes) Volume I: Integrated Executive Summary August, 2005 Colombia and Mexico Country Management Unit Poverty Reduction and Economic Management Unit Latin America and the Caribbean Region Document of the World Bank ECLAC Economic Commission for LatinAmerica and the Caribbean EMBRAPA BrazilianAgricultural ResearchCorporation ENCASEH HouseholdEconomic Characteristics Survey ENCEL HouseholdAssessment Survey ENE NationalEmploymentSurvey ENET NationalQuarterlyEmploymentSurvey ENEU NationalUrbanEmploymentSurvey ENHRUM Mexico RuralHouseholdsNationalSurvey ENIGH NationalSurvey on Household Incomeand Expenditure FAIS Social InfrastructureFund FA0 Foodand Agriculture Organizationof the United Nations FDI ForeignDirect Investment FGT Foster, Greer and Thorbecke Poverty Indicators FIRA AgricultureTrust Funds FIRCO Trust Fundfor SharedRisk FISM Municipal SocialInfrastructureFund FONAEVI NationalFundfor HousingEconomic Support FONDEN NaturalDisasters Fund FUPROVI Foundationfor HousingPromotion GATT GeneralAgreement on Tariffs and Trade GDP GrossDomesticProduct GE Generalized Enthropy GGAVATT LivestockGroups for TechnologyValidation and Transference GP Gross Profit GVO Gross 'v'alue of Output IICA Inter-AmericanInstitutefor Cooperation on Agriculture IMR Infant Mortality Rate IMSS Mexican Social Security institute ... 111 INDAP Agricultural DevelopmentInstitute(Chile) INDESOL NationalSocial Development Institute INEGI NationalStatistics, Geography and InformaticsInstitute INFONAVIT NationalInstituteof the HousingDevelopment INIFAP NationalInstituteof Forestry, Agricultureand LivestockResearch INTA NationalInstituteof Agricultural Technology ISSSTE Social Security and Services for the State Workers Institute LAC Latin Americaand the Caribbean LAG Local Action Group M&E Monitoringand Evaluation MXP MexicanPesos NAFTA North AmericaFreeTrade Agreement NBFI Non-BankFinancialInstitutions NGOs Non-Governmental Organizations O&M Organizationand Management OECD Organizationfor EconomicCooperationand Development PAC. Work Training Program PAHNAL NationalSavingsCouncil PAPIR Programfor the Support of InvestmentProjects PEC Special Concurrent Program PEMEX MexicanOil Company PET Temporary Employment Program PITT Technology Researchand TransferenceProgram PROCAMPO Programfor Direct Supportto Agriculture PROCEDE Programfor Ejido RightsCertification PRODESCA Programfor Rural Capacity Development PROFEMOR Programfor RuralEnterprises and OrganizationStrengthening PROMAP Program for the Modernizationof Public Administration iv PRONAF NationalProgramof Family Agriculture PROSAVI Special Programfor Housing Credits and Subsidies PSP ProfessionalServicesProvider RD RuralDevelopment RDS Mexico's Sustainable DevelopmentNetwork RNF RuralNon-Farm ROA Roles of Agricultura SAEBE Economic Support System for Job Seekers SAGARPA Ministry of Agriculture, Livestock,RuralDevelopment, Fisheriesand Food SCT Ministry of Communicationsand Transportation SE Ministry of Economy SEDESOL Ministry of SocialDevelopment SEGOB Ministry of the Interior SEMARNAT Ministry of Environmentand NaturalResources SEP Ministry of Education SFP Ministry of Public Management SHCP Ministry of Finance SHF FederalMortgageSociety SIACON Agricultural Consultation System SICAT Work Training System SINDER NationalSystem of RuralExtension SRA Ministry of Agrarian Reform ST Ministry of Tourism STDs Sexually Transmitted Diseases STPS Ministry of Labor and Social Prevision TFP Total Factor Productivity UNAM Mexico's NationalAutonomous University USD UnitedStates Dollars V EU EuropeanUnion VA Value Added WB The World Bank WHO World HealthOrganization vi Preface The collaboration between the Government of Mexico (GOM) and the World Bank on poverty has four objectives: (i) assist the GOM in improving the effectiveness of its programs to in reducing poverty; (ii)to support the analysis and design (or redesign) of specific policy interventions; (iii)to build capacity, especially in evaluation techniques and processes; and (iv) through workshops, to share best practice in poverty reduction policies in a national and internationalcontext. The combination of demand-driven analytical work and capacity building is designed to strengthen the government's institutional capacity to implement policies that bring about effective poverty reduction results on the ground. The results of the first phase of the Programmatic Poverty Work of the World Bank in Mexico were published in 2004: Poverty in Mexico: an Assessment of Conditions, Trends, and Government Strategy (World Bank, 2004). This report summarizes and consolidates the findings of three World Bank studies on poverty issues in Mexico written as part of the second phase of this work: Urban Poverty, Rural Poverty, and Social Protection. The second phase of the Poverty Work was coordinated by Gladys L6pez-Acevedo and Jaime Saavedra. JosC Maria Caballero was the task manager of the rural poverty report; Marianne Fay and Anna Wellenstein of the urban poverty report; and Gillette Hall and Laura Rawlings of the social protection report. Marcela Rubio Siinchez, Jonathan Goldberg, and Sara Johansson provided valuable technical assistance. This summary as well as the different studies have greatly benefited from the comments received from SEDESOL, the ministries that form the Social Cabinet, Presidencia, peer reviewers of the programmatic poverty work (Rodrigo Garcia VerdG, and Margaret Grosh and Norbert Schady, World Bank), peer reviewers of the individual studies, and participants at the second phasereview meetings. Special thanks to Mike Walton and Andrew Mason for their insightfulcomments on this summary and inthe overall design of the 2"dPhaseof the Programmatic Poverty Work. This report exclusively represents the views of the World Bank team. Vice President: Pamela Cox Country Director: Isabel Guerrero Sector Director:Ernest0 May Sector Leader: David Rosenblatt Task Managers:Jaime Saavedra and Gladys L6pez-Acevedo vii Acknowledgements The Urban Poverty report was managed by Marianne Fay and Anna Wellenstein. Other team members include Caridad Araujo, Magdalena Bendini, Sara Johansson, Gabriel Montes, Mary Morrison, Caterina Ruggeri Laderchi, Mauricio Santamaria, and Jesus Torres Mendoza. The team i s grateful to the peer reviewers Emmanuel JimCnez and Christine Kessides, and to Gladys L6pez-Acevedo for their comments. The team benefited from inputs and suggestions from counterparts in the mexican government, notably Rodrigo Garcia Verdci of Banco de Mkxico; Claudia Acuiia, Craig Davis, Cristian Muiioz and JosC Luis Ramos y Fusther of Subsecretaria de Desarrollo Urbano y Ordenacidn del Territorio, SEDESOL, and Carlos Maldonado Valera, of the Subsecretaria de Prospectiva, Planeacidn y Evaluacidn, also of SEDESOL. We are grateful to Lic. Concepci6n Steta GAndara for documentation and insights on the expansion of Oportunidades in urban areas. We thank SEDESOL for sharing the results of the survey of poor barrios, and Jorge Santibaiiez and Alejandro Garcia of the Colegio de la Frontera Norte for their help in accessingthe data. The Urban Poverty report i s part of an overall analytical program on poverty in MCxico managedby Gladys Lopez-Acevedo, Michael Walton and Jaime Saavedra, that includedan initial Poverty Assessment (Poverty in Mkxico: An Assessment of Conditions, Trends and Government Strategy), and companion reports on rural poverty (A study of Rural Poverty in Mkxico) and social safety nets (Mhico: An Overview of Social Protection), in addition to the present report on urban poverty. The team thanks Marcela Rubio Sanchezfor her efforts inensuring consistency of data and data sources across the various reports. Vice President: Pamela Cox Country Director: IsabelGuerrero Sector Director: John Henry Stein Sector Leader: Anna Wellenstein Task Manager: Marianne Fay, Anna Wellenstein The Rural Poverty report was prepared by a team comprising JosC Mm'a Caballero (LCSER, Team Leader), Fernando Barceinas (consultant), Tania Carrasco (LCSHE), Jorge Franco (LCSES), Sara Johansson (consultant), Salom6n Nahmad (consultant), Fernando Rello (consultant), Mario Torres (consultant), Dorte Verner (LCSEO), and Antonio Y6nez-Naude (consultant) . The Rural Poverty report is part of the Second Phase of the "Mexico: Programmatic Poverty Work", which is being carried out by the World Bank in coilaboration with the Government of Mexico. The report benefited from comments by Benjamin Davis and Alain de Janvry (peer reviewers), Mauricio Bellon (CIMMYT), and Bank staff Mark Cackler (LCSER), Marianne Fay (LCSFP), Ricardo Hernandez (LCSEN), John Kellenberg (LCSES), Gladys L6pez- Acevedo (LCSPP), Miguel L6pez (LCClC), Pedro Olinto (LCSPP), Jaime Saavedra (LCSPP), Emmanuel Skoufias (LCSPP), Michael Walton (LCRCE), and Anna Wellenstein (LCSFP). Feedback was received from Government officials and poverty analysis experts at a brainstorming meeting held in Mexico City on June 10-11 2004, and from high-level staff of the Secretaria de Agricultura, Ganaderia, Desarrollo Rural, Pesca y Alimentacidn (SAGARF'A) duringa presentationof preliminary results in August 2004. Special thanks are due Antonio Ruiz ... VI11 Garcia (Subsecreatry SAGARPA), Armando Rios Piter (Subsecretary SRA), J o d Antonio Mendoza Zazueta (Director FIRCO), Ivan Cossio (FAO) and Edgardo Moscardi (IICA). Vice President: Pamela Cox Country Director: IsabelGuerrero Sector Director: John Redwood Sector Manager: Mark Cackler Sector Leader: EthelSennhauser Task Manager: Jos6 Maria Caballero The Social Protection Report team benefited from discussions and input from counterparts in Mexico, including Rafael Freyre (Gabinete Social), Jose Antonio Gonzalez (SHCP), Santiago Levy (IMSS), Evelyn Rodriguez (IMSS), Deborah Schlam (Hacienda), and Miguel Szekely, Gonzalo Hernandez, Erica Ramirez and Alvaro Melendez (Sedesol). Very helpful comments were also received from World Bank colleagues including peer reviewers Margaret Grosh and Edmundo Murrugarra, as well as Cristian Baeza, Ariel Fiszbein, Isabel Guerrero, Gladys Lbpez-Acevedo, Andy Mason, Jaime Saavedra, Guillermo Perry, Helena Ribe. Michael Walton, and Steve Webb. Many thanks to Gabriela Falconi for editing and formatting this report. Country Director: Isabel Guerrero Sector Director: Ana-Maria Arriagada Sector Manager: Helena Ribe Sector Leader: Mark Hagerstrom Task Team: Gillette Hall, Laura Rawlings (co-task managers), Jaime Saavedra, Gladys Lbpez-Acevedo, Marcela Rubio, Trine Lunde, Carlos Noriega(consultants) and Maria Guadalupe Toscano (research assistant). ix Table Of Contents Introduction 1 Heterogeneity 3 Long-Term h o m e Growth, productivity, and poverty reduction 10 Vulnerability 18 Bibliography 38 X INCOME GENERATIONAND SOCIAL PROTECTION FORTHE POOR Introduction The first phsse report, Poverty in Mexico: an Assessment of Conditions, Trends, and Government Strategy (World Bank, 2004), was a poverty diagnostic which found that poverty is still a challenge in Mexico. Looking at the last decade, the severe 1994-95 crisis reduced real wages and sharply increased both extreme and moderate poverty. Poverty has subsequently fallen, but national and rural poverty rates have only recovered in 2002 to pre-crisis levels. As for the urban poverty rates, Cortds (2005) finds significant evidence of the recovery of poverty. Mexico has made considerable progress in achieving poverty reduction since the end of the nineties, with a performance superior to the Latin American average (Figure 1). More recent trends are encouraging, with extreme poverty falling by almost seven points in the 2000-2004 period. This reduction can be explained by positive developments in rural areas, where poverty fell from 42.4 to 27.9 percent, while urban poverty rates fell to a much lesser extent during the same period (Table 1and Comitk Tbcnicopara la Medicidn de la Pobreza,CTMP, 2005). Figure1. Share of Population livingon less than 2 USDa day 30 25 20 15 10 5 0 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 +Mexico LAC Brazil Source: WB staff estimates based on household surveys. 1 Table 1. Share of PopulationinPoverty Change 2000- Change 2002- Change 2000- 2000 2002 2004 2002 2004 2004 Nationa1 FoodPoverty 24.2 20.3 17.6 -4.0 *** -2.7 -6.6 *** Capacities Poverty 32.0 27.4 25.0 -4.6 *** -2.4 -7.0 *** Assets Poverty 53.8 50.6 47.7 -3.2 ** -2.9 -6.1 *** FoodPoverty 42.4 34.8 27.9 -7.6 *** -6.8 * -14.5 *** Capacities Poverty 50.1 43.9 36.1 -6.2 ** -7.7 * -14.0 *** Assets Povertv 69.3 65.4 57.4 -3.9 -8.0 ** -11.9 *** Urban FoodPoverty 12.6 11.4 11.3 -1.1 -0.2 -1.3 Capacities Poverty 20.3 17.4 18.1 -3.0 ** 0.8 -2.2 Assets Poverty 43.8 41.5 41.7 -2.3 0.1 -2.2 Note: The food-based poverty line is an estimate of the income required to purchase a food basket satisfying minimum nutritional requirements. The capacities poverty line includes non-food income for spending on education and health services. The assets poverty line also considers expenditures in housing, clothing, and transport. * Significant at the 10% level;**Significant at the 5% level;*** Significantat the 1% level. Source: WB staff estimates based on ENIGH applying the methodology o f the Technical Committee for Poverty Measurement. As discussedlater in the text, factors that have contributed to the reductionof rural poverty since 2000 include sustained macroeconomic stability, increased transfers, and income diversification into non-agricultural activities. Oporrunidades, Procampo, and, to a lesser extent, remittances have contributed to this reduction. On the other hand, poverty rates in the urban sector have not improved as much as in rural areas. The main challenge in the urban sector i s to increaseaccessto productiveemployment opportunities for the poor. This document summarizes the findings of three reports: Urban Poverty in Mexico, Mexico: a Study of Rural Poverty, and Mexico: an Overview of Social Protection, and focuses on (i)the generation of income opportunities for the urban and rural poor and (ii)social protection for the poor. The main messages can be summarized as follows: 0 The poor are a very heterogeneous group and, among other dimensions of poverty, location matters in the design of appropriate poverty alleviation interventions. Importantly, long-term income-generating Opportunities and coping strategies differ significantly between urban and rural areas, among different regions, between small and larger cities, and even within neighborhoods. The urban poor are limited to low-quality jobs marked by low productivity and with limited social protection. Labor market trends are not encouraging for the urban poor who are in low-productivity sectors marked by little income security and with few prospects for income growth. The urban poor are increasingly confined to the informal sector and as such have limited access to social protection. More goodjobs are needed in Mexico and the poor need to be able to access them. As extensively argued in World Bank 2004, this requirea policies and reforms that facor productivity growth in the 2 economy as well as policies that help the poor to become more productive. Access to quality education i s central to increasing the productivity of the poor. However, there are other factors that influence the productivity of the poor such as health financing, childcare systems, improved labor market regulations, access to transport, active labor market programs (ALMPs), and extending the coverage and the deepening of finance systems to the urban and rural poor inorder to encourage savings for investment. To continue supporting the rural poor to move out of poverty, it is important to increase agricultural productivity, especially for small- and medium-sized farmers, and facilitate their diversification into rural non-farm activities (RNF) of higher agricultural value-added. The rural poor are trying to move away from agriculture as the key source of income and employment and towards RNF activities, particularly in services and construction. Inpart, this i s the result of the poor performance of agriculture for small-scale farmers, who have not been able to increase their productivity over time. Nor have poor farmers been able to reap the benefits of RNF activities. Appropriate government support in areas such as education, health, and finance could help the poor to move more easily towards RNF activities, and also increase the productivity of agriculture. However, efficiency gains in agricultural and rural support programs are essential, given the high level of resourcesalready spent inthese activities. e Since its inception in the 1940s, Mexico's social protection system has not been well- suited to respond to the risks that the poor face. With a few notable exceptions, most prominently and recently Oportunidades, Mexico's social protection system has historically been insufficient and regressive in its coverage. This reflects its design, which i s linked to the worker's status in the labor market (Le. formal or informal). The vast majority of the working poor have not been covered by social insurance. A new program, Seguro Popular, aims to increase healthcoverage for the poor, but still many are left with only informal short-term mechanisms for safeguarding their income, protecting themselves against old-age poverty, and managing health risks. Although these mechanisms in some cases can be effective in coping with income shocks, they are costly in the long run since they have a negative impact on human capital accumulation. A reform of the social security system is needed in order to make it less regressive and expand its coverage. In parallel, other measures such as facilitating self-insurance through asset building in urban and rural areas could also help the poor to cope with shocks. This points to the need to complement social security reforms with reforms in other areas including the land, financial and housing sectors. Heterogeneity A key conclusion of this report is that geographical location must be taken into account in order to design adequate poverty interventions. The poor are heterogeneous, and poor people may differ markedly in their socioeconomic characteristics, sources of incomes, spending patterns, and coping mechanisms. This heterogeneity occurs at many levels. There are important differences between urban and rural areas, for instance, m t also among regions, between different urban settings, and even within urban areas, pointing to "pockets" of poverty. Understanding these differences is key in designing an effective poverty reduction strategy. Geographical location is one key variable in explainin2 differences among the poor. Income generation opportunities and social protection needs ary depending upon the poor's location. \ 3 Heterogeneity between and within rural and urban areas The urban and rural poor differ in their economic characteristics, sources of income, and spending patterns.The rurayurbandichotomy' can be misleading, as geographical settings are really a continuum of settlements by population size. However, if we use this somewhat arbitrary dichotomy, there are important differences that are still informative for poverty analysis. In 2004, one out of three rural residents was living in extreme poverty, compared to one out of ten of the urban population. The highurbanization rate in Mexico implies that the majority (close to 63 percent) of the moderately poor live in urban settlements. Table 2 shows some of the key characteristics of the poor that differ between rural and urban settings. The rural poor depend mainly on self-subsistence agriculture, self-employment, and non-agricultural activities, and have typically not completed primary education. Conversely, the urban poor depend on access to salaried employment, on non-agricultural activities mainly as employees in manufacturingor services, and have not completed lower secondary education. IThe urban vs. ruralclassificationrefersto wttlements with population>of more thai: 15.000 people.ilsing CON4PO's (Consejo Nacional de Poblacidn)detinition. 4 Extreme Poor Moderate Poor Non-Poor All Urban Rural AI1 Urban Rural All Urban Rural Rural-urbancomposition 100 39.3 60.7 100 62.5 37.5 100 69.0 31.0 No Education- Primary 63.6 50.0 71,g 48.5 39.4 61.9 33.0 25.6 45.7 Incomplete Primary Complete 21.2 27.6 17.4 24.9 27.1 21.7 23.6 23.2 24.3 Lower Secondary Complete 12.6 17.6 9.7 21.4 26.1 14.5 22.2 23.4 20.1 UpperSecondary Complete 2.2 4.0 1.1 4.0 5.6 1.7 10.5 13.3 5.5 University Complete I0.3 0.9 0.0 I 1.2 1.9 0.1 I 10.8 14.4 4.5 Agricultural Laborer 27.0 13.4 35.1 14.2 5.8 28.0 4.9 1.7 11.8 Non-agriculturalLaborer 32.0 55.6 18.0 57.8 69.3 38.6 65.5 70.3 55.3 Employer 2.4 0.6 3.4 3.1 2.5 4.0 6.9 7.6 5.6 Self-Employed 38.0 29.7 43.0 24.5 21.8 29.0 21.9 19.8 26.6 Non-remuneratedWorkers 0.6 0.7 0.6 0.5 0.6 0.4 0.8 0.7 0.8 Profile by sector of activit: Agriculture 45.8 7.7 68.4 16.6 4.0 37.5 9.4 2.1 25.0 'Extractionand Utilities 0.1 0.1 0.1 0.6 0.6 0.5 1.4 1.6 1.o Manufacturing 9.6 15.3 6.2 19.5 22.1 15.2 19.6 19.3 20.1 Construction 12.1 20.5 7.2 15.0 15.1 14.8 8.2 8.0 8.7 Commerce 12.7 19.2 8.8 17.8 19.8 14.5 17.4 18.8 14.4 Transportation 4.2 6.8 2.7 6.3 7.8 3.8 8.6 7.5 10.8 Services 15.5 30.4' 6.7 24.2 30.5 13.6 35.5 42.7 20.1 Note: Extreme poverty is the share of the population under the food poverty line. Moderate poverty is the share of the populationbetween the food and assets poverty line. Non-poor are those with income over the assets poverty line. Source: WB staff estimates usingENIGH 2004. All panelsexcept the first one add to 100vertically. There are some notable differences between the rural and urban poor in both consumption patterns and income structures. The urban poor spend relatively more on housing, transport, and education. Conversely, the rural poor spend more on food and clothing (see Table 3). However, the expenditures in healthare similar for urban and rural areas. 5 Table 3. Expenditure patterns, by geographical location and poverty status, 2004 Urban Rural Extreme Moderate Non- Extreme Moderate Non- Poor Poor Poor Poor Poor Poor Food, Beverage and Tobacco 42% 39% 24% 52% 48% 32% Clothing 4% 5% 5% 6% 6% 5% Housing 18% 20% 23% 13% 14% 17% Health 57c 4% 5% 57c 5% 7% Transport and Communications 10% 13% 16% 7% 9% 16% Education and recreation 8% 9% 14% 7% 7% 9 9 Other 13% 10% 13% 10% 11% 14% Source: WB staff estimates using ENIGH 2001. More importantly, the urban poor depend almost exclusively on the labor market for income while transfers are significant among rural workers. A remarkable difference i s the surprisingly low share of urban income that i s derived from transfers (see Figure 2). In fact, this is the only dimension in which the situation of Mexico's urban poor i s substantially different from the rest of Latin America (Table 4). The high share of labor income in urban areas contrasts sharply with the high share of in-kindincome and transfers in rural Mexico.' Since the share of rural poor income that is derived from transfers is higher than the regional average, we cannot attribute the lower importance of transfers in urban income to low overall expenditure on social safety nets, but rather to the heightened importance of other sources of income. The expansion of Oportunidudes to urban areas, however, is starting to change this pattern, due to the progressivenessof this program's transfers. However, the actual amounts of iransfers (in absolute terms) received by the urban poor remain higher than those for the rural poor, although they ri`ccive considerably less (in absolute and relative terms) from Oportiinidades. Transfers received by urban households. whether poor or non-poor. tend to consist much more of pensions, scholarships, and gifts from other households. Transfers (as calculated abo>e: do nut include subsidies such as the electricity subsidy, which are highly regressibt and benefit mainly the uihn population. See Poverty in Mexico, World Bank, 2004. 6 Figure2. Sources of income, by location, 2002 IW"6 . ... . " . 90% 19% 24% 18% 31% 80% 5% 5% 3% 4% 0% 70% 5% 5% 5% 7% 2% 60% 14% 19% 1890 M% 16% 10% 30% 57% 57% 20% 39% 1090 0% Lower20% Middle20% UpperM)% Lower 20% Middle10% Upper 60% Rural Urban LILabor income BusinesslcooperativeBTransfers RemittancesOOthers ENonmonetarymcome Source: Urban Poverty in Mexico, World Bank, 2005. Table 4. Income Sources for Poorest Quintile Labor income Capital income Rents and profits Pensions Transfers Urban Mexico 91.8 0.8 3.9 3.5 LAC average 74.2 1.5 4.0 13.6 LAC median 82.5 1.1 3.2 12.8 Rural Mexico 81.4 0.5 0.8 17.2 LAC average 80.6 0.6 0.9 12.9 LAC median 87.2 0.6 0.6 10.2 Note: Table 3 is not strictly comparable to Figure 1because there is no allowance for imputed income, and labor income includes what is referred to as "business/cooperative" income inTable 3. Source: The Urban Poor in Latin America, World Bank, 2004. Factors including population size, location, and labor market characteristics are all key determinants of well-being and the types of policy interventions neededfor rural and urban areas. Keeping this, and the fact that semi-urban areas are more similar to rural areas than to urban ones, in mind, the urban-rural distinction is nonetheless informative for policy making, as the challenges facing the poor in the "average" rural vs. urban areas show many differences: e Income sources, forms of employment, and opportunities differ: In urban areas, the poor must generate cash for survival, either through self-employment or wage-paying jobs. Urban poor may face difficuities in securing decent-paying employment in the formal sector due to lack of skills. childcare, or transportation, or because of stigma associated with where they live. The rural poor are much more dependent on agriculture than the urban poor or the rural non-poor and have been less able to access high-return occupations in the RNF sector becauye of low human capital. 7 The urbanand rural poor face different risks and use different coping mechanisms: Because of different income sources, economic shocks affect the urban and rural populations differently; thus, their coping strategies vary from one setting to another and appropriate policy responses should differ. Urban poor incomes are more responsive to growth, implying the ability to take advantage of more jobs in good times, but, conversely, are more sensitive to macroeconomic shocks in terms of structural employment. While labor markets constitute the main source of income shocks for the urban poor, they also provide a key coping mechanism to such shocks, as households resort to sending additional household members to work intimes of crisis. The rural poor benefit from safety nets such as transfers, subsistence agriculture and other in-kind income, access to forest resources, and local community ties not available to the urban poor. 9 The urbanpoor generally have better access to services than the rural poor, but low quality and crowding-out reduce the effectiveness of these services: On average, the urban poor have better access to infrastructure, education, and health services than the rural poor. However, outcomes related to these services do not vary greatly: infectious diseases are equally common among poor urban and rural children, and school enrollment rates and test scores are low among poor rural and urban children. This suggests that while coverage has expanded in urban areas, quality i s not keeping up, limiting the impact of improved access. Another important issue is that the specific demands for infrastructure services, the cost of providing the services, the engineering, the organization and management systems, and the forms of community participation are usually different between rural and urban areas, makingcomparisons of access to running water or public sanitary systems somewhat inaccurate. Environmental hazards differ: Inurban areas, risks include air pollution, collection and disposal of domestic and hazardous waste, water scarcity and water quality, and occupation of fragilehisky areas for residential purposes. In rural areas, they include deforestation, soil degradation, oil contamination, fertilizer and pesticide contamination of soil and aquifers, and health hazards in their application. In both cases they are important determinants of well being. Marginality and violence: Urban marginality and violence are linked to family breakdown, drug use and trafficking, degraded neighborhoods, opportunities for specific types of robbery, close contact between the destitute and the well-off, and tribal youth cultures. Rural marginality i s related to income, employment, geographical constraints, and often ethnic characteristics. While rural violence exists, it i s typically linked to land conflicts and the fight of rural organizations for human or economic rights, making it different from the individual and mob criminality of the cities. Poverty incidences and the characteristics of the poor also vary with the size of the urbanagglomeration. Income levels and other indicators of well-being are, on average, lower in smaller urban settlements, showing the existence of a rural-urban continuum. It is evident that the difficulties faced by the poor in a settlement of 15,000 people may more closely resemble those faced by rural dwellers than those living, for example, in any of the large Mexican cities. Indeed, extreme poverty incidence in medium-sized urban areas (with a population between 15,000 and 99,000 people) is about three times higher than that of agglomerations with more than 100,000 inhabitants (see Figure 3).3 Semi-urban areas (2,500 to 15,000 inhabitants) can be seen as ?he ENIGH - the household urvey used to compute poverty levels and correlates - is not designed to be representative at a disaggregated stratum level, nor at regional level. As a resul:. there i s risk for measurement errors which must w kept in mind uhen interpreting these disaggregated statistics. 8 transition regions between the large urban settlements and the dispersed rural areas (less than 2,500 inhabitants). Figure3. Poverty trends by agglomeration size, 1992-2002 Extreme poverty incidence (left) and moderate poverty incidence (right) tm 90 BD m m Q 40 *-....-..---- a a m 10 10 0 . m O ?I 94 % 98 m PI 92 94 % 98 m 02 Source: UrbanPoverty in Mexico,World Bank, 2005. But even within urban areas, there is great heterogeneity in well-being, even down to the neighborhood level. Notably, the levels of well-being in different geographic areas that belong to the same city differ greatly; this holds true even for Mexico's largest cities. SEDESOL's own survey of poor barrios reveals inequality to be higher within these barrios than between them. And these types of inequalities seem to persist over time, as unequal cities may grow disjointedly, with different sub-areas improving at very different speeds. Regional Heterogeneity is also important Beyond the rurayurban distinction, Mexico is characterized by great diversity within and across regions and states in terms of socioeconomic outcomes, assets, and ethnicity. Regional characteristics have a significant impact on the incidence of poverty. In Mexico's poorest regions, urban areas are poorer than elsewhere - and not necessarily because they are smaller in size on a~erage.~There i s a clear geographic (both regional and urbadrural) effect on income poverty and other indicators of well-being - literacy rates, housing conditions, access to basic services. For example, as seen in Table 5, the incidence of extreme poverty in rural areas in the richer Northern regions has consistently been lower than the incidence of extreme poverty in the urban areas in the Gulf, South, Center-North, or Center region^.^ For many aspects of well-being, the national urban-rural gaps are in fact smaller than the regional differences between urban areas. Even for the two poorest regions, the Pacific and the South, the urban-rural literacy gaps are smaller within the regions than between urban areas in the Pacific states and those in the South. 'One possibleexception IS theSuuth Pacific, where there are fewer lirge urban agglomerationsthan elsewhere. Refers to inLome poverty and follous INEGI's classification into seven regions. See Mexico: A Study of Rurul PC-I '.&, World Bank. 2005. 9 Table 5. Extreme Poverty Trends, by Region Regional Rural Urban 1992 1996 2002 2004 1992 1996 2002 2004 1992 1996 2002 2004 North 9.4 22.0 6.4 7.7 13.3 30.9 13.8 13.2 8.2 19.7 4.8 6.5 Mexico* 9.9 25.8 8.6 9.3 26.7 49.9 15.9 11.9 6.9 20.1 7.1 8.3 Gulf 23.7 45.1 34.7 25.8 30.5 52.6 43.7 34.2 14.3 34.7 24.2 15.8 Pacific 12.6 26.7 13.7 11.8 18.5 32.3 21.8 27.0 8.5 23.0 9.4 6.4 South 41.1 60.0 39.9 32.8 45.6 66.7 47.9 40.3 30.9 45.7 24.4 22.8 Center-North 28.5 44.5 21.1 18.1 40.4 52.6 27.2 23.7 18.2 36.7 16.4 13.8 Center 44.7 49.5 30.1 24.1 53.0 57.9 41.6 29.4 34.5 37.0 15.4 17.2 * Mexico includes the Federal District and the State o f Mexico. Note: Rural localities are defined as those with less than 15,000 inhabitants. ENIGH is only designed to be representative at the national, urban and rural levels. As a result, there i s risk for large measurement errors which need to be kept in mind for these disaggregated statistics. Source: Mexico: A Study of Rural Poverty, World Bank, 2005. In sum, geographic location matters for the characteristics of the poor, the types of deprivation they face, and for designing appropriate policy responses. There are significant differences among the poor depending on where they live. Labor market characteristics, sources of income, ways in which they cope with shocks, and access to infrastructure vary with geographical location. This makes the rurahrban distinction useful for policy-driven analysis. The regional dimension i s also very important, and rural and urban poverty rates are in fact more closely correlated within regions than across regions (e.g. South vs. North). Moreover, inequality can be high within urban and rural areas, and even within specific neighborhoods. Understanding these patterns i s important to designing an effective poverty reduction strategy. Long-term income growth, productivity, and poverty reduction The poor suffer the consequences of low labor productivity and lack opportunities to move to higher productivity employment, which limits their potential for income growth. Inurban areas, real wages for the poor have declinedsince 1991, and eventhough pay levelshave recovered since 1996, the improvement was not sufficient enough by 2003 to regain the value lost since 1991; the share of self-employed who work without any own capital (informal, salaried workers without own capital, IWOC) has increased. In rural areas, lack of sufficient dynamism in the agricultural sector, particularly in small-scale farms, concentration of growth in the more commercial sector, and limited access to high-retum jobs in the RM; sector are key factors in explaining stagnant income growth for the rural poor. Slow productivity growth is a general problem for the Mexican economy and i s the main factor behind the slow growth in labor earnings. Slow productivity growth affects Mexico's ability to compete internationally, especially in the US, affecting boththe poor and the non-poor. Slow productivity growth has meant that fewer good jobs are available in the labor market. From the supply-side, the poor have less access to good jobs. Mexico's ability to compete in international markets, especially in the VS, has not improved. This is reflected in the country'Y poor performance in improving total factor productivity (TFP). Loayza, Fajnzylber, and Calderdn (2005) find that TFP growth in Mexico was among the lowest in Latin America, substantially below the 1.1 percent annual growth average for the seven largest Latin American countries during the 1971-2000 period. This was particularly true in the 1990s, when productivity grew at a low 0.4 percent per year Since 1998, unit labor costs have increased as productivity growth has lagged in comparison with the U.S. Productivity growth since the nineties has been particularly slow outside the industrial sector, i.e. in sectors where the poor tend to work. The Urban Poor working morefor lesspay 2 Employment is the main and frequently the only source of income for the urban poor. This is especially true in Mexico, where 92 percent of the income of the poorest urban quintile comes from labor, compared to an average of 74 percent for Latin American urban areas as a whole and 81 percent for Mexico's rural poor families. Although there has been improvement in wage levels since 1996, the evidence shows that the urban poor were still working more but for lower wages in 2003 as compared to 1991. Over the past fifteen years, labor market trends have not been encouraging. In general, work opportunities have shrunk in sectors with "higher-quality jobs" - such as the manufacturingsector - and the poor have resorted to low-productivity jobs in the informal or self-employment sectors. The participation rate of the poor has increased by 10 percentage points in the last decade. Higher employment rates, reflecting increasing participation rates, have eased extreme poverty, but many urban inhabitants still find themselves unable to escape moderate poverty. Despite the increases in participation, the poor remain less able to access the labor market than the non-poor - only 48 percent of the working-age poor have ajob, compared to 70 percent for the non-poor. Real wages for the urban poor are down 5 percent since 1991, despite the economic recovery after the Tequila Crisis.6Real wages have fallen mainly in the self-employment sector where the poor are very highly represented, and within each category wages have failen more for the poor than for other groups (Figure 4). Indeed, real wages for the extreme poor who are IWOC fell by 22 percent. The fact that the share of employment in IWOC increased so markedly in combination with falling wages suggests an increase in involuntary informality and a potential segmentation in the Mexican labor market .' '3ank staff calculation based on 2003 ENEU. The analysis on this section (for more details see Chapter 4 of Urban Poverty in Mexico, World Bank, 2005) is based on the ENEU, which only includes data about labor incomes coming from an individual's mainjob. 11 Figure4. UrbanAreas: The poor have seen the sharpest cuts inreal wages (Real wages in 2003 as percentage of real wages in 1991) 120 108 80 60 40 20 0 Informal without Informalwith Employer Public Employees Private Employees capital capital All flExtremePoor Source: Adapted from Urban Povero in Mexico, World Bank, 2005. Private employees stand for formal privateemployees The decline in real wages reflects the low quality of jobs available to the poor. The formal private sector has been unable to generate jobs at a high enough rate to absorb poor workers, which would allow them to increase their earnings and move out of poverty. In 2003, the poor were predominantly employed by smaller firms in the private non-tradable sector (formal and informal), especially commerce, personal services, and construction. For example, firms with less than 5 employees accounted for 65 percent of the very poor's employment, as opposed to 40 percent of total employment. Small firms are less productive' and less likely to comply with employment regulations or provide benefits or stable longer-termjobs. They also pay less than larger firms, even to comparable workers. In addition, wages in these firms have fallen: in 2003, for example, wages in firms with fewer than 10 workers paid below 1991 levels, whereas wages for workers in larger firms rose about 4 percent. The poor have tended to move towards low-productivity self-employment. As public sectorjobs shrank after the Tequila Crisis, the formal private sector was unable to create morejob opportunities. As shown in Figure 5, the share of the working poor who were informal without capital - typically low-productivity activities - grew from 14.5 percent to 19.6 percent between 1991 and 2003. Moreover, this share continued to grow after the Tequila Crisis, even as the economy recovered. The decline in public sector employment from 12.5 percent in 1991 to 7.2 percent in 2003 was mainly matched by increasing self-employment, not private formal salaried employment. The urban poor have tumed more to informal self-employment where real wages are still below 1991levels.' ''Tanand L6pez-Acevedo,2003, andFajnzylber,Maloney, and Montes, 2005. Chapter1of Urban Poverp in Mexico, World Bank,2005, shows the trends in the compositionof employment. Figure 5. The urban poor have turned to IWOC where real wages havedeclined 25% 1m 1M 20% 0 98 0% 15% 094 3 92 10"'o 090 088 5% 086 084 0% 082 1991 1995 1959 2003 +Share of Poor Worlung as IWOC +Red Wages of Poor IWOC Workers Source: Urban Poverty in Mexico, World Bank, 2005. The poor have also tended to move towards low-quality jobs in the non-tradable sector, outside the manufacturing sector. The share of the poor employed in manufacturing, where jobs are of higher average quality, also declined from 26 to 19 percent between 1991 and 2003. As a result, an increased share of poor workers now work in construction, commerce, and personal services - sectors characterized by lower-than-average wages, high informality, slow growth, and a tendency to decline sharply in recessions. Concurrent with the decline in manufacturing employment, the share of the working poor employed inexport industries declined from 39 percent to 30 percent between 1992and 2002. The poor are increasingly choosing work in the informal sector because of a lack of alternatives (World Bank, 2004). Push factors, rather than pull factors, explain the increase in the share of the poor in the informal sector. Rather than seeking informal sector work because of convenience (e.g. flexibility to work from home or combine with childcare or desire to work as an entrepreneur), the poor are restricted to the informal sector because of a lack of opportunities elsewhere. Until the early 1990s, unemployment and informality in Mexico were negatively correlated, suggesting that the majority of workers were voluntarily choosing the informal sector, rather than being pushed into it by a lack of formal opportunities. Maloney (1999) also found that during the first half of the 1990s the shift toward self-employment was mainly driven by profit maximization. The trend appears to have reversed (at least partially) after 1992, as informality began to follow a somewhat similar trend to unemployment, more compatible with a conventional pro-cyclical view in which a decline in labor demand results in a growing informal sector (Figure 6). It may be then that there are an increasing number of workers joining the informal sector because of deterioration in the demand for formal labor. 13 Figure6. Since 1991, informalityhasrisen with unemployment 8 1.3 7 d r f i I.2 1.1 ' 3 d 0.9 $ . 3 0.8 L 0.7 0.6 0.5 -(c Unemploymcnt -CInformaVFormaI Note: Informality includes bothsalariedinformalworkers and self-employed. Source: Poverty in Mexico, World Bank, 2004. Recent research suggests that workers with little education, labor market experience, and capi?al are unlikely to move voluntarily into the informal sector. Therefore, high rates of self- employment in poor households may also be considered involuntary or induced by labor market distortions (Fajnzylber, Maloney, and Montes, 2005). The fact that significant increases in IWOC were accompanied by a large fall in wages also supports the push-factor hypothesis. Why are the poor unable to compete?First, there is an important educational gap betweenthe poor and the non-poor. Although the educational level of the poor has increased, they continue to lag behindthe non-poor. Between 1991and 2003, the average years of schooling for urban poor household heads increasedfrom 6.6 to 8, but remained 30 percent lower compared to the non-poor. Despite progress on some fronts, this poorhon-poor education gap bas not substantially improved in the last decade, and 55 percent of the labor force still does not even have complete secondary education (ENEU 2003). Likewise, the quality of education has not improved markedly in recent years. But inferior socialcapital, information,andresourcesand a lack of transportation and childcare are also likely to hamper the competitiveness of the poor in the labor market. Education does not fully explain the lower income of the poor. An exercise conducted to evaluate the returns to education found that, conditional on having the same level of education and experience, poor households still receive lower wages, and the gap over time increasedfor all educational categories- in 2003, the wage gap between poor and non-poor workers with secondary education is close to 40 percent." The lower quality of education for the poor i s one possible explanation. The non-educational factors that explain this differential include: inferior social capital, information, and resources; the stigma often associatedwith slums and other poor areas; a lack of access to affordable childcare facilities; and institutional factors, such as labor market rigidities. Institutional factors such as stringent firing regulations, hiring modalities, promotions, and provisionsfor shutdownsand downsizingmay be distortingMexico's labor market. International evidence suggests that excessive or very rigid regulations, even if well- intended, curtail the creation of formal employment, as employers seek to circumvent costly and complicated requirements by hiring workers informally. This particularly affects young and less- '"Montes.Santaman'a, and Bendini, 2004. 14 educated workers. Labor market regulations may impede the filling of vacancies on merit and deter formal employment, to the disadvantage of the poor. However, the 1997 Social Security reform, which replaced the pay-as-you-go pension system with one based on privately managed individual accounts and overhauled the health financing system, appears to have stimulated formal employment generation, mainly because of two reasons. On the one hand, the reform increased public contributions to the financing of both systems (as a basic pillar), effectively reducing the burden on employers and employees and introducing clearer incentives for them to make contributions "on top" of this basic pillar. In line with this. contributions in general were made more uniform across wage levels, scrapping the existing "levy" on human capital under the former system. On the other hand, the reform strengthened the link between contributions and benefits. For example, the portability of benefits across jobs increased in the pension system, which raised the incentives to contribute to the system and, thus, to become formal. Statistical analyses conducted for this report (Santamaria and Montes, 2005; Kaplan, 2004) show that the formal sector grew more than the informal sector in the period right after the reform, which provides evidence that the reform may have had a positive impact on formal employment generation. Other studies conducted in the country, however, do not find suchevidence. The Rural Poor need to explore opportunities in the RNF sector Since the 1990s, Mexico's rural economy has seen a decline in the importance of agriculture and its prominence as a source of income for rural families has decreased. The share of the rural workforce employed in agriculture declined from 63 to 56 percent between 1995 and 2003. The share of income that rural households derive from agriculture dropped by half, falling from 51 to 24 percent between 1992 and 2002, largely due to a sharp reduction in income from independent farming. Instead, the rural workforce became more dependent on the RNFsector: income share from RNFwage labor increasedfrom 20.4 to 36.1 percent, and mostly in high-returnsectors. But the rural poor remain more dependent on agriculture and low-return activities inthe RNFsector than the non-poor, and the gaps have increased over time. The poor have participated in the transformation of the rural economy, but to a lesser degree than non-poor households. For example, from 1992 to 2004, income from non-agricultural labor increased from 19.6 to 23.9 percent for the poor, while it increased from 28.6 to 46.4 percent for the non-poor. The share of extreme poor households' income derived from agriculture fell significantly, from 67.7 to 38.7 percent. But relative to the average household, very poor households experienced a smaller increase in income from RNF wage labor, especially in high-returnactivities, which was compensated by a larger increasein transfers. 15 Table 6. Income Shares in RuralMexico 1992 2002 Income Shares All Extreme All Extreme Households Poor Households Poor IndependentFarming 38.5 38.1 12.6 16.8 Agricultural Wage Labor 12.3 19.6 11.3 21.9 Su6-tota1Agriculture 50.8 57.7 23.8 38.7 IndependentNon-FarmActivities 8.1 4.8 5.7 6.8 Non-FarmWage Labor 20.4 15.9 36.1 17.2 Highreturn 4.9 1.3 23.8 4.4 Low return 15.5 14.6 12.3 12.8 Transfers 8.0 6.0 16.5 25.4 Other Sources 12.6 15.5 17.8 11.9 Sub-total Non-Agriculture 49.2 42.3 76.2 61.3 Note: Occupations providing average earnings below the moderate poverty line are classified as "low return", those above as "high return." Source: Mexico: A Study of Rural Poverty, World Bank, 2005. The poor have been unable to take full advantage of the process of agricultural modernization and increasing productivity. Both land and labor productivity rose in the 1990s at a rate above 2 percent, a reasonableif not impressive performance. But agricultural growth has favored the Northem states, where commercial farming and crops are concentrated and farm sizes are larger. Average gross profit per hectare, after deducting family labor, i s increasing in farm size, with negative gross profit for farms less than 2 hectares and positive gross profit for larger farms. This suggests that Mexico's agricultural sector is also becoming increasingly dualistic. On the one hand, large, commercial, irrigated farms appear to experience increasing productivity. On the other hand, the productivity of subsistence farmers in marginal or isolated areas i s stagnant, as these farmers are unable to switch to export crops and modem agricultural techniques. Agriculture needs to continue to play an important role in poverty reduction. International evidence suggests that high-productivity agriculture goes hand in hand with growth in higher-productivity RNF activities. Among other things, higher agricultural income results in higher demand for RNFproducts, stimulating the rural economy, and there can be strong linkages between agriculture andnon-agriculturalindustries and services. There is a need to increase the efficiency of resources spent on upgrading and supporting the agricultural sector. However, Mexico's public expenditures in rural development in production-related areas i s high by any standards and, as a share of public spending, the highest inLatin America (see Figure 7). The average 1996-2000public expenditure in agriculture as a share of total public expenditure is over 8 percent, way above Argentina (less than 0.5 percent), Chile (above 2 percent), or Brazil (2 percent)". The Progruma Especial Concurrente, which includes agriculture, social, environment, and infrastructure spending by federal agencies in rural areas, i s equivalent to some 30 percent of agricultural GDP. The lack of dynamism in the agricultural sector cannot be attributed to lack of public resources but rather to the needto increase their effectiveness and targeting, along with other structural reforms. "Mexico:A Studyof Rural Poverty,WorldBank.2005. 16 Figure 7. Labor Productivityand Public Expenditurein Agriculture ._ a L P 10,000 4 u < & I. 8,000 L. ai 6,000 U Note: Labor productivity defined as agricultural value added divided by the agricultural labor force and measured in US Dollars of 1995. Average expenditure figures refer to expenditure i n production-related programs for agriculture only, not to all rural development expenditures. They are in current US dollars. Source: Mexico: A Study of Rural Poverty, World Bank, 2005. Low levels of education and lack of access to physical assets hamper the ruralpoor's ability to apply technical upgrading in agriculture and to diversify into RNF activities. Research shows that additional schooling promotes technical change in crop production, participation in RNF activities, and increased household income. However, average levels of schooling achievements are still low, especially for the poor. While only 55 percent of the extreme rural poor aged 12-14 were enrolled in lower secondary in 2002, almost three-quarters of the non-poor were enrolled." Physical capital assets tend also to be associated with higher income from agricultural activities. The poor are still unable to access higher-quality jobs in the formal sector or in higher-yielding agricultural activities. The functioning of the labor market is key to understanding the sources of poverty. The situation of the urban poor in Mexico deteriorated in the 1990s, as evidenced by falling real wages compared to the early 1990s. As a consequence of falling incomes - and growing unemployment during the Tequila Crisis - the poor moved to very precarious informal jobs in sectors and occupations that offer low salaries, grow slowly, and are highly vulnerable to recessions. By 2003, poverty had declined slightly because the poor were working more, but mostly in those "low-quality" jobs. The economy of rural Mexico, on the other hand, is undergoing a significant transformation, with important implications for the rural poor. The labor force has experienced modernization and agriculture i s declining in importance, even for the poor. Agriculture, of course, remains important, but its development has been uneven. Large, commercial, irrigated farms experience increasing productivity, whereas the productivity Poverty in Mexico. World Bank, 2004. Note, however, that the definition of rural for these estimates was based on INEGl'sclassificationwith populationsof less than 2.500 people. of small or subsistencefarmers in marginal or isolated areas may be stagnant, as these farmers are unable to switch to export crops and modem agricultural techniques. An increase in transfers allowed the rural poor to compensate for this negative development, but there appear to be opportunities that the rural poor can seize, if helped by adequate govemment interventions. These opportunities, by and large, are in the RNF sector. Thus, raising the productivity in the economy as a whole (in order to increase the amount of "good" jobs available) and raising the productivity of the poor (in order to increase their chances of accessing a good job) must be cornerstones in the poverty reductionstrategy. Vulnerability While important progress has been made in helping certain groups to manage risks, there are still vulnerable groups who have limited access to safety nets, such as the elderly poor and low-income groups facing unemployment or major health risks. Mexican individuals and households face a diverse set of income risks. The analysis shows that Mexico has made important progress in reaching certain at-risk groups, particularly the young rural poor population. However, several key vulnerable groups can be identified for whom the frequency of risk and severity of loss compels a reexamination of government policy. Among these are the elderly living in poverty - a problem of considerable scope in Mexico, especially when assessed in a regional comparison - and the low-income population, which potentially faces very high costs associatedwith health care and unemployment risks. In addition to these idiosyncratic (individual) risks, Mexicans are also periodically exposed to aggregate shocks, which include economic shocks and natural disasters. The evidence suggests that the overall variation of the households' consumption in the face of these risks i s small because significant consumption smoothing i s taking place.13 Adjustment to macroeconomic shocks has largely taken place via falling wages, such that income losses are diffused across the working population, as opposed to resultingin sector-specific unemployment. Private risk management i s an important component of households' responses to the above risks, but its limits are also apparent. Household characteristics -education, family composition, and self-employment - affect the distribution of shocks in urban and rural areas in similar ways. Coping mechanisms such as loans, donations, sale of assets, and increased labor market participation appear common in both urban and rural areas. But these responses may not be very useful vis-8- vis covariate (aggregate) shocks, when risk pooling i s not possible due to the possibility of general equilibrium effects on labor markets. Moreover, as mentioned above, the importance of coping mechanisms might differ between urban and rural areas because of the different economic context: the labor market affects the urban poor more; conversely, the rural poor benefit from safety nets such as subsistenceagriculture. Despite recent progress. the Mexkan social protection system has historically been - and still is - inequitable and leaves many vulnerable citizens unprotected. High rates of poverty among the elderly coexist with a very high fiscal burden for public pension systems, which points to a serious mismatch between programs and needs. The Mexican protection system for private workers is, after the 1997 IMSS reform. healthier and financially sound in the long run. However, the challenge of increasing coverage to large segments of the population remains. l3Overall, there is some evidence of successful {butperhaps otherwiie costly) consumption smoothing- !arge drops in income do not translate into large drops in consumptifin. Certain coping strategies, such as runninp d o w health or !he fact that mothers cannot stay at home to takc care of their children. may protect present consumlition levels but c:in have negative effectson the consumptionpath in the longer run. I S While health costs constitute a major risk for low-income households, formal health insurance i s limited and regressive across income deciles. By not covering the poorest households, public social security institutions have failed to mitigate the inequity that persists in Mexico's society. As a result, a great majority of the poor have few means of managing risks. Individual shocks, risks, and vulnerability -old age and health risks Poverty rates have been higher among the elderly in Mexico than among the general population. At 38 percent, the poverty rate among older adults is much higher among the elderly than the national average and at a level close to that seen in countries of lesser economic development than Mexico, much higher than in countries such as Brazil, Chile, or Colombia, and considerably higher than the national average. This gap in poverty rates i s particularly pronounced, but not unique to Mexico (Table 7). Table 7. Poverty Rates amongthe Elderly in LatinAmerica EntirePopulation 65 and older Bolivia 30.5% 47.5% Brazil 24.6% 18.5% Chile 20.8% 23.9% Colombia 24.0% 32.9% CostaRica 21.7% 29.1% Guatemala 19.1% 27.1% ElSalvador 27.4% 38.0% Mexico 22.1% 37.6% Source: Gill, Packardand Yermo (2004). Although public pension programs absorb a large share of public expenditures, the elderly poor lack significant coverage. Whereas over 20 percent of the urban population over age 65 receives a pension, only 7 percent of the urban elderly poor have access to a pension. The rural poor are virtually uncovered, with pension incidence among the elderly at less than 1 percent (Table 8). An immediate result of lacking income security at old age i s that a large share of the elderly poor continues to work, particularly in rural areas. A range of social assistance programs and transfers reach h e elderly poor, including Procampo, Oportunidades (600,000 households), Liconsa,Programapara Adultos Mayores (covers 98 percent of people aged 70 and over in Mexico City), Acuerdo Nacional para el Campo, and Programa Alimentari~.'~ fact, as In shown above, transfers have been a major source of income growth for the rural poor. Yet high poverty rates among the elderly are in and of themselves evidence of the limited reach of the public social protection system. 'Coverageot'pensionssystems and other forms of income SLIP~:~-among Pvlexico,'selderly is discussed in inore derail ~ inMexico: An Oveniew of Social ?r-vtection,WorldBank. 7005. 19 Urban Rural Extreme poor All Extreme poor All Receives pension 1996 7.9 21.2 0.9 5.5 2002 II 6.7 22.1 0.8 5.3 Works and does not receive a pension 2002 29.8 24.9 57.9 49.7 Informal risk management strategies rely on extended families (social networks), private transfers (remittances), and accumulation of assets (savings or investments). Accumulated assets tend to increase with age, which would tend to make the elderly better off relative to other groups. However, the value of assets held by the elderly depends critically on two institutions where much progress is still needed in Latin America, in particular in their limited reach to the poor population: (i)legal institutions that protect property rights; and (ii) financial institutions that allow households to convert illiquid assets into income for consumption in old age. In addition, it is important to better understand the existing informal risk management and private transfers (inter-generational) in place and to understand how these are affected by socio-demographic changes. Private strategies to cope with income-risk in old age may have adverse long-term effects. One example is the effect on land inheritance patterns in rural areas. The main risks perceived by old land-owning men, whose possibilities of migration or participation in the labor market are highly diminished by their age, are being left alone, and being unable to earn enough income to survive. Common responsesto this situation observed in ejidos and comunidades are: (i) ownersclingtotheirland,resistinganypressurestopassitoninlife,(ii) ordaughters old sons stay with parents in the household, looking after them and helpingtilling the land, often with the promise that they will inherit it; (iii) parentsinstill uncertainty about inheritance rights as a means to maintain family unity and thus ensure that transfers keep flowing from distant children. As such, the resistance of small landholders of advanced age to pass on their lands to the young generation is part of a broader survival-cum-risk management strategy identified in field studies. However, it results in reduced agricultural productivity, youth exodus, and demographic imbalances. The ageing of the population means that old-age poverty poses one of the major social protection challenges facing Mexico. A recent study of demographic trends in Mexico found that the percentageof the population above 60 years of age rose from 6.6 in 1989 to 8.6 in 2OO2.I5 As shown in Figure 8, the increase in the average age of the household head i s particularly pronounced in rural areas. Higher life expectancy i s increasing the proportion of elderly relative to young, putting pressure on private risk management strategies - like extended families and inter-generational transfers - and formal pension systems alike. l5Ariza and de i)li\eirn 2004 20 Figure 8. Age of Household heads, 1992-2002 50 1 46 4 1I 44 j Y 42 40 ! I I 1992 1996 2000 2002 +Urban poorest 10% +Rural poorest 10% .- Source: Mexico: An Overview of Social Protection,World Bank (2005). Exposure to health shocks combined with insufficient capacity to manage them is a major risk associated with poverty. The low-income population in Mexico - a majority of which does not have access to health insurance - faces very high costs associated with health care. More than 5 million Mexican citizens face catastrophic health expenditures each year, causing at least 2 million of them to fall into poverty.16 In addition, the distribution of out-of- pocket expenditure is regressive, as the poor show much higher levels of out-of-pocket expenditure than the rich. This translates into a higher frequency of catastrophic health shocks and suggests that available risk-pooling mechanisms do not reach the poor andor do not offer efficient protection. There are significant inequalities in public health expenditure, and the poor are in general uninsured or, inthe best of cases, partially insured. Consequently, they frequently pay out-of-pocket for health services. Households in the lowest decile directly spend around 11 percent of income on health care while the richest spend less than 4 percent, pointing to important problems of efficiency and equity of public health s~bsidies.'~ A recent study found that while only 9 percent of insured households fall below the poverty line after catastrophic health care expenditures, 40 percent of uninsured households are impoverished when suffering a health shock." In2002, 73 percent of households impoverished from health expenses were not insured. There i s also an urban-rural gap, as 60 percent of ruralhouseholds fall below the poverty line as a result of catastrophic health care expenditures, while 17 percent of urban households face the same problem. There is some evidence that Oporrunidudes, which combines cash transfers with free access to healthclinics, can have an important role in protectingruralhousehold income from health shocks (Box 1). However, some consumption smoothing mechanisms that protect household consumption from health shocks are in place, even if uncovered by Oportunidades. The recent introduction of Seguro Popular - which i s targeted at poor families - i s therefore a positive development. A forthcoming impact evaluation will shed light on the extent of the program's effectiveness. It1Povern. in Me~ricti,World Bank. 7004. Catastrorhic health shock defined as more than 50 percent of household income net of basic nutritional