Xiv Congreso Internacional Del Clad Sobre La Reforma Del Estado

XIV Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Salvador de Bahia, Brasil, 27 - 30 oct. 2009

El impacto de la crisis global en el mercado de trabajo de América Latina: lecciones de las crisis anteriores Diana Alarcón Si bien el impacto de la crisis financiera en América Latina no ha sido tan pronunciado como en otras regiones, la contracción del comercio internacional, la caída en las remesas, la reducción de los flujos de inversión extranjera directo y la contracción del financiamiento han tenido un gran impacto en la región, con grandes costos sociales.1 Estimaciones recientes de la Cepal apuntan a una contracción del comercio—a abril de este año—del 31,8% en relación con el mismo mes del año anterior, mientras que la caída en las remesas puede ser de entre el 5 y 10 % en 2009 (entre el cuarto trimestre de 2008 y el primer trimestre de 2009) y la caída en la inversión extranjera directa puede ser entre el 35 y 45 %. El deterioro en las expectativas de las familias y las empresas que generan estas tendencias, han provocado un reducción generalizada en el consumo y la inversión del sector privado que redunda en un aumento del desempleo que puede llegar a 9 % hacia finales de este año, que significa más de tres millones de personas adicionales. Aun cuando la economía mundial está mostrando signos de recuperación, las perdidas en el empleo y el costo social que estas generan tienen impactos de más larga duración: “ la recuperación de los índices sociales generalmente toma el doble de tiempo que la de los índices económicos, tal como sucedió durante la crisis de los años 1980, cuando los indicadores sociales tardaron 24 años en llegar a los niveles previos a la crisis y los económicos 12 (CEPAL, 2009, Estudio económico de América Latina y el Caribe 2008-2009). Incluso el escenario optimista que predice la recuperación del crecimiento en 2010, una tasa promedio regional del 3.1% anual, sería insuficiente para revertir el aumento del desempleo y la informalidad que se estiman para este año2. Tendencias en el empleo ILO construyó tres escenarios para predecir los cambios en el empleo que acarrearía la crisis económica a nivel mundial. En enero del 2009 se estimaba que el total de trabajadores desempleados aumentaría entre 29 millones (en el escenario 1) a 59 millones (en el escenario 3) en el periodo que va de 2007 a 2009. Esto corresponde a una tasa de desempleo de entre 6.5 a 7.4 respectivamente. En el escenario intermedio se estimaba que el número de trabajadores desempleados aumentaría en 39 millones en el mismo periodo, lo que corresponde a una tasa de desempleo del 6.8 por ciento y un números totales de trabajadores sin empleo de 220 millones en 2009. En la actualización de estas proyecciones 1Simplemente en remesas, el Banco Interamericano de Desarrollo estima que America Latina y el Caribe van a recibir cerca de $62 billones de dólares en remesas en 2009, lo que representa una reducción del 11 por ciento comparado con el año pasado debido a la contracción en Estados Unidos, España y Japón principalmente. La reducción de las remesas afecta a 4 millones de personas, de las cuales casi la tercera parte están en México, el país con mayor recepción de remesas en la región.

2Las

estimaciones de la Cepal se pueden consultar en:http://www.eclac.org/cgi-bin/getProd.asp? xml=/prensa/noticias/comunicados/7/36467/P36467.xml&xsl=/prensa/tpl/p6f.xsl&base=/tpl/top-bottom.xsl (acceso 7 septiembre 2009). 1

XIV Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Salvador de Bahia, Brasil, 27 - 30 oct. 2009

realizadas en mayo de 2009 se incorporó el deterioro en los mercados de trabajo observados hasta ese momento que implicaron un aumento del número de desempleados a nivel global de 21 millones de personas en el primer escenario, 31 millones en el escenario 2 y 50 millones en el escenario 3. Con la información existente al momento, se estima que el desempleo a nivel global está colocado entre los escenarios 2 y 3.3 Sin embargo, en los países en desarrollo, el desempleo abierto es una expresión muy parcial del deterioro de los mercados de trabajo. El desempleo no es opción para la mayor parte de los trabajadores que no cuentan con seguro de desempleo, ni ahorros que les permita subsistir por periodos muy largos. La pérdida de empleos formales en esos países y/o la falta de acceso a estos para una gran mayoría de trabajadores, se traduce en altas tasas de empleo precario, de baja remuneración y sin seguridad social. En base a las últimas proyecciones sobre el crecimiento del producto a nivel mundial, la OIT estima que la tasa de empleo vulnerable fluctúa entre el 48.9 y 52.8 por ciento del empleo mundial, que se traduce entre 1.5 y 1.6 billones de trabajadores a nivel mundial.4 Si bien América Latina ha transitado la crisis en mejores condiciones dada su solvencia financiera, un buen ambiente de negocios, una acumulación de reservas considerable lograda, en parte, por los altos precios de las materias primas de los últimos años 5, las proyecciones de crecimiento en la región para 2009 se han deteriorado desde el último ejercicio del FMI en enero del 2009. Se espera que las economías de Argentina, Brasil, México y Venezuela se contraigan este año, con una contracción particularmente severa en México.6 Las economías de México, Centroamérica y el Caribe—a partir del turismo—son particularmente sensibles a la debilidad del crecimiento en Estados Unidos, mientras que las economías del Sur de América Latina han sido afectadas por la caída en los precios de las materias primas. En las últimas proyecciones de la OIT, el desempleo en la región afectaría entre 22.6 y 25.7 millones de trabajadores que corresponde a una tasa de desempleo de entre 8.1 y 9.2 por ciento.

3En el primer escenario se utilizan las tendencias históricas que asocian el crecimiento económico a los cambios en el desempleo en cada país entre 1991 y 2008 con las proyecciones sobre el crecimiento del producto del FMI para 2009. El segundo escenario esta generado a partir de la relación entre el crecimiento económico y el desempleo en el peor ano de contracción observado en cada país con las proyecciones sobre el crecimiento del producto del FMI para 2009. El tercer escenario esta construido tomando el peor año de recesión económica en cada país y su impacto sobre el desempleo; asumiendo que este crecimiento en el desempleo ocurre simultáneamente en todos los países desarrollados en 2009 con un efecto retardado en los países en desarrollo. Para estos últimos, se estima que la mitad del aumento en el desempleo ocurre en el año 2009.

4Empleo vulnerable se calcula como la suma de los trabajadores por cuenta propia y los trabajadores en empresas familiares como porcentaje del empleo total. http://www.ilo.org/wcmsp5/groups/public/---dgreports/--dcomm/documents/publication/wcms_106504.pdf Accedido el 7 de septiembre de 2009. 5Estas predicciones se pueden consultar en: http://web.worldbank.org/WBSITE/EXTERNAL/NEWS/0,,contentMDK:22243070~pagePK:34370~piPK:34424~theSiteP K:4607,00.html (Accedido el 7 de septiembre 2009). 6En las ultimas estimaciones nacionales, México calcula una caida del PIB de mas del 10 por ciento anual, mucho mayor que la contracción observada durante la crisis del tequila en 1995-1996. 2

XIV Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Salvador de Bahia, Brasil, 27 - 30 oct. 2009

De manera paralela, se estima que el porcentaje de trabajadores en condiciones de vulnerabilidad en América Latina y el Caribe aumente para colocarse entre el 31.6 y 34.5 por ciento del total de trabajadores en 2009. Respuestas frente a la crisis América Latina es probablemente una de las regiones con mayor experiencia en el manejo de condiciones de crisis. Desde principios de la década de los 80, la región experimentó periodos recurrentes de crisis económicas y financieras que llevaron a la implementación de distintas iniciativas sociales para paliar los impactos más negativos de la contracción económica. La experiencia acumulada en la región para responder a condiciones de crisis, le permitieron, no solamente enfrentar esta crisis global con una posición más sólida desde el punto de vista de la macroeconomía—con mejor estabilidad en las finanzas públicas, tasas de cambio mas flexibles y reservas de divisas que le han permitido activar políticas anti-cíclicas—sino además con una red de protección social que han sostenido una reducción importante en la incidencia de pobreza y mejoras sustanciales en la cobertura de los sistemas de educación básica, salud y nutrición. La innovación en el diseño e implementación de programas sociales en América Latina son ahora referentes importantes para el diseño de programas de protección social frente a condiciones de crisis en otras regiones. La evolución de los programas de protección social en América Latina ha dejado grandes enseñanzas acerca de la efectividad de distintas formas de intervención para proteger niveles mínimos de consumo entre la población más vulnerable y de protección social para impedir daños permanentes en la salud y en la acumulación de capital humano entre la población. Los primeros programas de protección social que surgieron en los años 80 frente a la crisis generalizada de la deuda, estaban diseñados básicamente para ampliar la cobertura de los sistemas de salud, educación y saneamiento. Los Fondos de Inversión Social que se realizaron en muchos países de la región, intentaban ampliar la infraestructura de servicios básicos bajo es supuesto de que este era el factor que limitaba el acceso a estos servicios entre poblaciones vulnerables. La evaluación a este enfoque de “oferta de servicios” volcó la atención hacia los condicionantes de la “demanda de servicios” con una propuesta que intenta generar estímulos a las familias para hacer un mejor uso de los servicios de educación y salud para aumentar la inversión en capital humano y romper, de esta manera, la reproducción intergeneracional de las condiciones de pobreza. En ese sentido, América Latina ha sido una región pionera en la adopción de programas de transferencias condicionadas de ingresos que intentan, no solamente reducir la pobreza extrema en el corto plazo, sino mejorar las condiciones de salud y educación de los niños para colocarlos en condiciones de mejorar su inserción laboral a largo plazo con mejores ingreso que les permita superar su condición de pobreza.

3

XIV Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Salvador de Bahia, Brasil, 27 - 30 oct. 2009

Como puede verse en el cuadro adjunto, hay varios programas en la región diseñados bajo estos lineamientos. Los Progamas de Transferencias Condicionadas de ingresos se han convertido en un elemento central de las estrategias de reducción de la pobreza con pequeñas variaciones pero muchos elementos comunes: • • • • •

Selección de beneficiarios a partir del ingreso familiar (e indicadores geográficos en algunos casos). Extrema pobreza Transferencia condicionada a cambios de comportamiento Tipicamente asistencia de niños a la escuela, atención a la salud y nutrición Atención integral a las familias (no los individuos) Estímulo a la demanda de servicios

País

Argentina

Brasil

Brasil

Chile

Colombia Ecuador

Nombre del Programa

Jefes y Jefas de Hogar Desempleados

Bolsa Familia Programa de Garantía de Ingreso Familiar Mínimo (Renda Mínima) Programa Puente, entre la Familia y sus Derechos Familias en Acción Programa de Protección Social

4

Fecha de Inicio

Abril 2002

Octubre 2003

Junio 2001

Marzo 2002

Noviembre 2000 Junio 2000

XIV Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Salvador de Bahia, Brasil, 27 - 30 oct. 2009

País

Honduras

Jamaica

México Nicaragua Perú República Dominicana

Nombre del Programa Proyecto Pograma de Asignación

Fecha de Inicio

Enero 1999

Familiar BID Fase II Programa de Promoción a través de Educación y Salud (PATH) Programa de Desarrollo Humano Oportunidades Red de Protección Social Transferencias Condicionadas en

Diciembre 2003

Agosto 1997 Enero 2000 Marzo 2005

Efectivo Tarjeta de Asistencia Escolar (TAE)

Octubre 2001

Programs of cash transfers to the poor were introduced in Latin America as a way to reduce the negative impact the economic recession in the 1980s. Their importance increased in the late 1990s, partly in response to the persistence of high levels of poverty in spite of several years of positive growth in the region. But they also reflected more fundamental changes in the conceptualization of poverty and the means to reduce it. Conditional Cash Transfer Programs (CCTP) have gained relevance as a central piece of poverty reduction strategies in many countries throughout the region and they are being considered as promising alternatives in countries that have not yet adopted them. Earlier poverty reduction programs, including those that had a cash transfer component, were conceived as programs of assistance to help families cope with that transitional costs of economic hardship generated by the debt crisis in the mid-1980s and economic reform of the early 1990s. The underlying assumption was that the main mechanism for poverty reduction was job creation and expansion of income opportunities of the poor through economic growth. The central effort of governments in the region was to restore macroeconomic stability and growth. In this view, poverty reduction programs were a temporary safety net to smooth the impact of recession on the poor until growth and stability were restored.

5

XIV Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Salvador de Bahia, Brasil, 27 - 30 oct. 2009

Tight fiscal restrictions also placed greater emphasis on issues of efficiency in the allocation and use of public resources. Strict targeting for the selection of beneficiaries living in extreme poverty became the main tool to avoid leakages to the non-poor. Sectoral programs to improve school attendance, nutrition, basic health or to expand access to basic infrastructure incorporated targeting as a central feature in the designed of social policy interventions. In the late 1990s, it became evident that the relationship between economic growth and poverty reduction was weak, especially in highly unequal Latin American countries. The evaluation of poverty reduction programs also pointed at problems of design and implementation: i) information on which the selection of beneficiaries was based was not systematically collected and quality validated, a problem that re-enforced discretionary selection of beneficiaries based on political considerations in some countries; ii) sectoral focus in the design of poverty reduction programs often resulted in duplication and lack of coordination among institutions with similar programs7; iii) related to this, the delivery of services and benefits was fragmented, often small, and thus failed to accumulate a critical mass of assets to lift people out of poverty; iv) programs focussed on the individual and they failed to see the family as the social and economic unit that determines the living conditions of individuals; v) targeting, even in countries where there were good sources of information, was often bent by political interests with the consequent drain of already scarce resources towards non priority groups. In addition to these points, extreme targeting helped to decrease Type 1 errors by excluding the non-poor as beneficiaries from social programs, but it in the context of constrained resources it increased Type II errors by excluding poor families from the benefits of goods and services for poverty reduction. In the late 1990s, the persistence of high levels of poverty theoretical inspite of several years of modest growth, but also the contributions of Amartya Sen and the Human Develoment paradigm that placed greater emphasis on the importance of expanding the opportunities of families and individuals to facilitate full participation in society and the economy, helped to readdress the design of programs for poverty reduction. The multi-dimensional character of poverty was explicitely incorporated into the design of more integrated programs that focussed on the family as opposed to the individual and were based on the principles of subsidiarity and co-responsibility. More recent programs aim at two objectives: i) in the short run, decreasing poverty levels through direct cash transfers to increase current consumption; ii) in the long run, breaking the vicious cycle of poverty by estimulating families to increase their investment in human capital (education and health) as a condition for the transfers of resources. CCTP stimulate the demand for services, as opposed to a narrow focus on the supply of basic services. Improvement in the frequency and quality of household surveys and registration of beneficiaries has also helped to improve targeting reducing discretional selection of beneficiaries. Substantial expansion of resources in the budget of these programs has also helped to reduce the exclusion of beneficiaries among the poor.

7A good example of the duplication that prevailed in early programs of poverty reduction can be found in the review that De la Breire and Lindert do of the flaws of the data registry of beneficiaries in Brazil. De la Breire and Lindert, 2005. 6

XIV Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Salvador de Bahia, Brasil, 27 - 30 oct. 2009

Since the late 1990s Conditional Cash Transfer Programs (CCTP) have been adopted as a central part of poverty reduction strategies in Latin America. With important variations from country to country that reflect different contexts but also the particular evolution of poverty reduction programs, they all share important common elements. Governments grant cash transfers to families in extreme poverty (usually identified by level of income and few other indicators for geographical location and socio-economic characteristics of families). In exchange families are committed to send children to school on a regular basis and attend health clinics to receive basic prevention. Different conditionalities have been incorporated in various countries (attending talks to parents to improve health habits and nutrition, accepting advice to improve the dynamics of family life, etc.). But conditionality typically includes a behavioral change towards education and health. The underlying assumption is that by estimulating greater investment in human capital, labor productivity of young entrants to the labor force will improve their opportunities to increase income breaking the vicious circle of poverty. To the extent that CCTP incorporate interventions to simultaneously improve the level of income of beneficiaries, school attendance and health care they offer an opportunity to improve coordination among different levels of government and other stakeholders. Some of the programs designed around these criteria include: Chile Solidario, Oportunidades in Mexico, Bolsa Familia in Brasil, Familias en Acción in Colombia, the Program Asignación Familiar in Honduras, Program of Advancement through Health and Education (PATH) in Jamaica, Red de Protección Social de Nicaragua, Programa Beca Futuro in Bolivia, Bono de Desarrollo Humano in Ecuador. An interesting feature in the design of CCTP is the construction of sound information systems throughtout implementation of the program that has helped improve monitoring and the evaluation of results8. Evaluation of some of the earliest programs (Brazil and Mexico, but also Colombia and Nicaragua more recently), show significant improvements in school enrolment, especially in secondary school; improvement in assistance to primary health care facilities and greater consumption among families in extreme poverty. Impact of these programs in current poverty reduction, obviously depends on the coverage and the level of income transfered to families. The long term impact of these programs breaking the vicious cycle of poverty (an explicit and important objective of these programs) is more difficult to evaluate. It will have to wait a few years when children start entering the labor markets. Estimation of long term impact at present crucially depends on the assumptions made on the returns to the education and health of children in future labor markets.

8Some of these programs have used rigorous methodologies to measure results building experimental models (within the program Oportunidades in México for example) or through the construction of control groups, including in some cases the ex-ante evaluations to support evaluation of alternatives in program design. Some of these studies include: Gertler, 2000, Hoddinott, Skoufias y Washburn, 2000, IFPRI 2000 y 2002, Schultz 2000, Skoufias and McClafferty, 2001, World Bank, 2001, Bourguignon, Ferreira and Leite, 2003. 7

XIV Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Salvador de Bahia, Brasil, 27 - 30 oct. 2009

CCTP have become a central component of policies adopted by many Latin American countries ro reduce poverty. Resources allocated fluctuate from 0.12 per cent of GDP in Colombia in 2002, to 0.32 per cent in Mexico and Jamaica in 2002 and 2004 respectively. Level of transfers in other countries has been smaller, but almost all countries are increasing the amount of resources committed to these programs to become a central component of government’s strategies to reduce poverty (Rawlings, 2004). In Brazil, for instance, the government consolidated previously dispersed transfer programs to support education, nutrition and gas consumption (Bolsa Escola, Bolsa Alimentacao and a Gas Subsidy) and is turning Bolsa Familia in the largest CCTP in the world, both in terms of the number of families covered and the amount of financial resources transfered. While the design of CCTP addressed many of the problems identified in previous transfer programs, there is still controversy around some of the following issues: i) CCTP are oriented to estimulate the demand of poor families for basic services in education and health, but they are not necessarily helping to improve the supply of such basic services. In the worst case escenario, without an increase in resources or substantial reforms in service provision, a larger demand estimulated by CCTP would contribute to overcrowding and further deterioration of already thin public services; ii) conditionality runs the risk of excluding the poorest of the poor in countries with a large incidence of poverty and geographical dispersion where there are no schools or health posts to meet the conditions for participation in the program; iii) changes in the power relations within the family (typically cash transfers are disbursed to mothers) are create tensions in some countries; iv) especially among indigenous and traditional communities, the selection of beneficiaries based on the socio-economic indicators of individual families is challenging traditional leadership and may become a source of tension within the community; v) cash transfers may generate negative incentives towards work by parents9; vi) with the exception of Brazil, management of CCTP is highly centralized within the government and may counter efforts towards the descentralization of public services; vii) better use of good quality statistical data for the selection of beneficiaries, has reduced discretionality in the selection of beneficiaries, but criticism remains about the political use of information and benefits10. An additional factor that will have to be included in the evaluation of these programs as effective mechanisms for poverty reduction is their cost effectiveness. A recent study of Oportunidades in Mexico (Coady, Pérez and Vera-Llamas, 2005) found that the administrative costs of the program, which represent about 11 percent of total resources allocated to the program in recent years under-estimates total costs because they do not incorporate the private costs incurred by beneficiaries to access cash transfers and comply with conditionalities11. The estimation of costs in relation to the total budget is not straightforward. Cost reporting varies from country to country and not all costs are fully accounted for. 9In the case of Mexico, the evaluation ruled out the existence of such substitution effects...... But it is not an issue that has been sistematically evaluated in other countries. 10An illustration of this point is the controversy generated in Mexico by the appointment of the former Minister of Social Development to lead the presidential campaign of her political party for her knowledge of the operation and access to the information of beneficiaries of Oportunidades. 8

XIV Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Salvador de Bahia, Brasil, 27 - 30 oct. 2009

Assumptions made about the time duration of the programs will also be critical in the estimation of administrative costs. The only point we are trying to raise here is that any effort to assess the impact of CCTP vis a vis alternative interventions would need to include proper estimation of private costs, an issue that deserves further evaluation. Comprehensive systems of monitoring and evaluation of CCTP have helped to refine the design and implementation of CCTP improving their impact. They have contributed to give continuity to these programs inspite of political renovation in countries where opposition governments have win elections (this is the case of Oportunidades in Mexico that took after Progresa and Bolsa Familia in Brazil that grew out of Bolsa Escola). Yet, difficult challenges remain. In the short term, creating mechanisms to improve inter-sectorial coordination to make sure greater demand for services is consistent with the provision of quality public services. In the long term, breaking the vicious cycle of social exclusion and lack of opportunities to revert the inter-generational transmission of poverty will remain a difficult challenge. Increasing years of schooling and their health status are critical factors in the early development of children and expands their opportunities for an active participation in society. But the extent at which better schooling and health will expand their opportunities to improve labor earning will depend on the relevance of education and the dynamism of labor markets to create better remunerated jobs to accomodate the improved skills of young entrants. It will also depend on the opportunities young workers have to access credit, product markets and technology to support more dynamic growth of micro and small entrepeneurs. The intersectoral approach CCTP have incorporated in their design opens up new opportunities to improve policy coordination in areas that are critical to the success of long term poverty reduction: improvement in the quality and relevance of education, the expansion of job opportunities to new entrants, deepening of credit markets to respond to the needs of small scale producers, the creation of new platforms for technological innovation in the products and processes that are relevant to the poor, the creation of new forms of association among small producers and between small and larger companies, expansion of social and economic infrastructure to improve market access to name a few areas that remain constrains to the expansion of employment and productive opportunities to the poor restricting their earnings and contributing to the inter-generational reproduction of poverty. By dissagregating the various sources of income changes in Brazil, Chile and Mexico since the 1990s, this paper seeks to provide empirical evidence of the relative importance of “other sources of income” for the poor (where resources from CCTP would be captured), vis a vis earnings from labor of the current generation of beneficiary families. In the next section we briefly review the characteristics of CCTP in Brazil, Chile and Mexico with the aim of providing the policy framework that may help to explain the origen of “other sources” of poor household 11Administrative costs in the study include the cost of identifying beneficiaries, disbursement of cash transfers to families and verification of compliance with conditionalities. Actual administrative costs depend on the time horizon of these programs. Identification of beneficiaries as one of the most expensive components of the program is substantially reduced over the long term. Private costs include the income of working children and transportation costs to the health unit, the school (especially relevant at the secondary level) and the post to collect payment when applicable. A low cost estimate of private costs in Mexico amount to 20 percent of the total cost of the CCTP in the short run and 36 percent in the long run, when some of the most expensive components of the administrative costs are reduced. 9

XIV Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Salvador de Bahia, Brasil, 27 - 30 oct. 2009

income identified in recent surveys. A hypothesis we are contemplating in this paper is that improvement in the design of CCTP in these countries has helped to concentrate cash income among the extremely poor, explaining the increasing importance of “other sources” in the income of those at the bottom of the distribution. Chile Chile is probably the country with the longest experience of highly targeted social programs. The macro-economic orthodoxy of the military government placed a heavy emphasis on the creation of markets even in basic services that were once provided and or regulated by the State. Extreme targeting for the delivery of basic service to the poor were refined over time and backed up with improved systems of information to identify the poor. Sectoral policies included specific interventions whose chief objective was to avoid leakages of limited resources to the non-poor. Recovery of democracy in 1990 broadened the constrains of strict targeting of social policy and redifined its role as an instrument to reach the poor in order to meet the goal of universal provision of basic social services. During the 1990s macroeconomic stability and rapid growth helped to reduce the level of poverty in the country. Absolute poverty almost halved from 38.6 percent in 1990 to 20.6 in 2000. By 1996 extreme poverty was also reduced by half; from 12.9 to 5.7 percent (Mideplan, 2004) but no substantial progress was made thereafter. In the year 2000 the government of Ricardo Lagos introduced the program Chile Solidario to estimulate the demand for basic education and health care among families in extreme poverty. The program is centered around the family, not the individual, as the unit of intervention. It includes an element of “personalized” support to families in extreme poverty to identify their specific needs and match them with the supply of government sponsored services. The objective is to work directly with families to expand their opportunities to reach markets, develop skills and strenghten the dynamics of family life to estimulate shared family responsibilities and promoting their autonomy for the generation of income and their social integration to community life. Chile Solidario is designed to reach 225 families (800 thousand people) in extreme poverty, including 15.6 thousand elderly in extreme poverty and living alone. It consolidates former programs that subsidized water consumption and sanitation and replaces other State sponsored pension systems. Not all services included in Chile Solidario represent direct subsidies to families. Secondary school attendance is estimulated through subsidies to school providers. A family that qualifies to participate in Chile Solidario receives a small cash tranfer for the first 2 years to guarantee minimum levels of income. Participation in the program is conditioned upon compliance with 52 conditions estipulated in the contract subscribed between the government and each family. Once families graduate from the program they qualify for an exit bonus for 3 more years to support a minimum consumption while they consolidate their independence to generate autonomous income. Subsidies for the elderly and handicaped in extreme poverty are permanent. Once a family has qualified to be included in Chile Solidario, they have preferencial access to government services in the areas where assistance is required within the development dimensions that are part of the program: identity cards, health, education, housing, jobs, income and family dynamics.

10

XIV Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Salvador de Bahia, Brasil, 27 - 30 oct. 2009

The first phase of Chile Solidario started operations in January 2002 with benefits for 56 thousand families. It became a law in 2004 and by the end of 2005 it had reached its target of 225 thousand families (Mideplan, 2006). It has been gradually replacing resources transfered to the poor through other programs. Resources for this program increased significantly; from 101 million pesos in 2002-2003 to 11,786 million in 2003 and 40,864 million in 2004 (almost 66 million dollars at a 620 pesos conversion rate). But even with this increase in resources Chile Solidario still represents a small portion of monetary subsidies for poverty reduction. In 2003, 57 per cent of social investment channeled through Ficha CAS–main instrument for the identification of the poor, currently called Ficha Familia) —were cash transfers (excluding resources from Chile Solidario). It is difficult to have the exact dollar amount of cash transfer programs because not all sectoral programs from past policies have been replaced. Table 1 offers an approximate figure estimated for those social programs that use the Ficha CAS (currently Ficha Familia) to target their beneficiaries. In 2003 monetary transfers to families were about 444.8 million dollars (assuming most resources from Chile Solidario took the form of direct cash subsidies). Direct cash transfers in that year were close to 60 percent of total social investment allocated through Ficha CAS. And almost 70 per cent such transfers corresponds to subsidized pensions to the elderly. Mechanisms for targeting in Chile are fairly efficient; 56 per cent of subsidized pensions for the elderly and 67 of the subsidies for poor families are allocated to the 20 per cent bottom of the distribution. And between 78 and 87 per cent of respective subsidies are captured by the poorest 40 per cent (Tables 2 and 3). Two points are evident from information provided in this brief review: Chile has a long tradition in the use of targeting for the delivery of goods and services. Social programs have good systems of information and relevant managerial experience to reach the poor. Cash transfers for poverty reduction have succesfully reached the intended population helping to improve the level of income and consumption of the poor. But the size of transfers is small when compared to other programs in Latin America. In 2003 total cash transfers that can be documented from available information represent 0.15 percent of GDP and they are intended to reach no more than the 5.7 percent of the population in extreme poverty. Once we add the transfer of subsidized pensions, total transfers represent 0.5 percent of GDP. Brazil Adopting the poverty line proposed by the World Bank to estimate extreme poverty (US$1.08 a day adjusted by the cost of living of the various regions), 27.8 percent of people in Brazil (21.9 percent of families) are poor. Extreme poverty is greater in rural areas, affecting 46 percent of people. In June 2004, 4.1 million families (approximately 16.5 million people) benefited from Bolsa Família the largest cash transfer program in the world. In 2005 Bolsa Familia reached 8.7 million families with a budget of 6.5 billion reales. In 2006 coverage is expected to increase to 11.2 families12.

12Interview with Marcia Lopes, Executive Secretary in the Ministry for Social Development and Reduction of Hunger, Folha de Sao Paulo, January 2006). 11

XIV Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Salvador de Bahia, Brasil, 27 - 30 oct. 2009

Bolsa Familia consolidates other cash transfer programs from previous governments in Brazil: Bolsa Escola, Bolsa Alimentacao, Cartao Amimentacao and Auxilio Gás were designed to estimulate school attendance, better health care, food consumption and a subsidy to gas consumption to compensate for the cost of adjustment. Administration of these programs was the responsibility of different government offices creating duplications, gaps and inefficiencies. Bolsa Familia was designed to resolve these problems but is not the only source of cash transfer. Other programs include guaranteed minimum income to rural residents; PETI, BPC – PPD, BPC – Idosos and Agente Jovem which include direct cash transfers for children, young people and people living with AIDS. But Bolsa Familia is, by far, the largest cash transfer program within Brazil. Making a rough estimate from information available and assuming that all resources allocated to the other programs (excluding rural minimum income) take the form of transfer payments to families, they represented about 6.7 percent of resources transfered by Bolsa Familia in 2005 (MDS, 2005). Bolsa Familia started operations in 2002 with five explicit objectives: “(a) consolidating and rationalizing existing federal conditional cash transfer programs; (b) promoting efficiency in the use of public resources; (c) improving the system for identifying the target population; (d) leveraging synergies from jointly promoting education, health and nutrition incentives; (e) strengthening monitoring and evaluation; and ( f ) leveraging opportunities to promote vertical integration in the social safety net between federal and sub-national programs” (quoted in World Bank, 2004). The priority within Bolsa Familia are families in extreme poverty (with a per capita income below R$ 50 a month or US$ 17). But it also targets the moderately poor defined as those families with a per capita income below R$ 100 (US$ 34) a month. Cash transfers range from R$ 15-95 (US$5-33) depending on the composition and level of income of families. The average transfer per family has substantially increased every year. In 2003 average transfers were R$ 24.75 a month, it increased to R$ 66.93 in 2004 and to R$ 64.67 in 2005 equivalent to US$ 22.3.(MDS, 2005). This amount is significantly higher than the average benefits received from previous programs but it is a fraction of the poverty line. It was about 17 percent of the poverty line used by the World Bank, 10.8 percent of IPEA’s and 5.4 percent of the minimum wage, based on the average transfer per family in 2005. The consolidation of previous programs for poverty reduction and substantial increase in resources available through Bolsa Familia was estimulated by the persistence of high poverty incidence in Brazil that remained at around 30 percent since the mid-1990s; compounded by the recent economic slowdown which brought a sharp increase of poverty, especially in the metropolitan areas of the Southeast. Young people entering the labor market, unskilled workers, indigenous people and afro-descendents are markedly affected by high, often increasing, incidence of poverty. Similar to other programs in Latin America, Bolsa Familia was conceived as a program to support minimum levels of consumption among the poor, while they invest in the human capital that would help them generate autonomous income in the future.

12

XIV Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Salvador de Bahia, Brasil, 27 - 30 oct. 2009

The coverage of Bolsa Familia has met (sometimes exceeded) its own targets every year. In 2003 it had reached 3.6 million families and it almost doubled in 2004 with transfers to 6.6 families. By October 2005 there were 8 million beneficiary families in Brazil. Remanents from other programs (Bolsa Escola, Cartao Alimentacao, Auxilio Gás and Bolsa Alimentacao) add another 3 million beneficiary families. Table 4 shows a substantial increase in resources since 2002 and the growing importance of Bolsa Familia as it replaces previous transfer programs. Relative to GDP total resources transfered under these programs are estimated to increase from 0.2% in 2003 to 0.5% by 2006. An early evaluation of Bolsa Escola (Bourguignon, et. al. 2003) found a significant impact on school enrolment, especially at the secondary level, but they found no impact on poverty reduction or inequality. The size of transfers was rather small. Once cash transfers in Brazil were substantially increased with the adoption of Bolsa Familia, a recent study by the World Bank estimated that an average transfer of R$71 (US$24) to 11.2 million families by end 2006 could result in a large reduction of extreme poverty, from 14.5% to 8.1% and a reduction in overall poverty (extreme + moderate) from 33.3% to 29.9%. This transfer could reduce the Gini Coefficient of income inequality from 0.595 to 0.578 and would help to improve the years of schooling among children by 3.4%. Results are predicted assuming there is perfect targeting of subsidies to the poor and no change in any other source of income for the poor. Mexico Oportunidades is the continuation of Progresa, introduced in 1997 to support rural families. Oportunidades became the flagship program for poverty reduction in the administration of president Fox. In the year 2000 coverage was extended to semi-urban and urban areas. By 2005, almost 5 million families in Mexico received benefits from Oportunidades; 69 percent in rural areas and another 17.4 percent in larger rural areas, classified as semi-urban (Table 5). As a conditional cash transfer program families in extremely poor are responsible to enrol children at school and supervise their attendance for at least 85 percent of school days. They are also expected to assist to health facilities on a regular basis to receive primary health care. In addition, mothers are expected to attend workshops where they receive nutritional information and advice on other health issues. Families receive a cash transfer for each child attending school, from third grade of primary education to high school. The amount of grants for school attendance increases with the level of education; it is estimated to cover the income contribution of working children to the household and it is slightly higher for girls at secondary level when women are more likely to drop out of school. A small stipend is also provided for books and school materials twice a year (Table 6). In addition, families who qualify for Oportunidades receive a small cash transfer to support food consumption. Failure to meet any of the conditions in the program results in the suspension of stipends transfered to families every two-months13.

13In the third quarter of 2004, only 4.4 percent of beneficiaries were identified as non-compliance on the assistance to health facilities. Sedesol, 2004. Impact Evaluation 13

XIV Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Salvador de Bahia, Brasil, 27 - 30 oct. 2009

Coverage of the program had a rapid expansion since 2000. Five million families in 2005 are slightly over one fifth of the total number of households in the country and over one fourth of the total population. Sedesol (the Secretariat for Social Development responsible for the administration of the program) estimates that there are 5.6 million households in extreme poverty in Mexico. Assuming perfect targeting, coverage from Oportunidades would reach 89.3 percent families and 81 percent of individuals in extreme poverty. The budget allocated to Oportunidades has also grown fast. In 1999 resources for Progresa were 0.16 percent of GDP. By 2005 the relative importance of Oportunidades more than doubled to represent 0.38 percent of GDP. Direct transfers to families, however, is much smaller. A large part of the budget supports the expansion of services within the Ministries of Health and Education to accomodate an increasing demand for services. Part of the budget goes to the purchase of a nutritional supplement for children under 5 years of age and for pregnant and lactating mothers14 distributed by the Mexican Institute for Social Security (IMSS). CCTP in Mexico havebeen one of the best evaluated programs of this kind, since its early introduction as Progresa15. Positive results in promoting school attendance documented in external evaluations were critical to give sustainability to the program, even in face of the comming to power of an opposition government in the year 2000 for the first time in 70 years. More recent evaluations report a substantial increase in the visits of beneficiaries to health clinics in urban areas and rural areas, including attention to pregnant women, inmunizations, family planning, basic treatment for diarrea and respiratory problems and preventive check ups. In 2003 Sedesol reported that resources transfered through Oportunidades represented 22 percent of the total consumption of rural families and 15 percent consumption of urban families16. Referencias Bourguignon, François; Ferreira, Francisco and Leite, Phillippe G. 2003. Conditional Cash Transfers, Schooling, and Child Labor: Micro-Simulating Brazil’s Bolsa Escola Program. The World Bank Economic Review, Vol. 17, Nº. 2 229-254. Washington, D.C.: World Bank Coady, David; Pérez, Raúl and Vera-Llamas, Hadid. 2005. Evaluating the cost of Poverty Alleviation Transfer Programs: an Illustration based on PROGRESA (now Oportunidades) in Mexico. Economic and Sector Study Series, Region II. RE2-05-006, December 2005. Washington, D.C.: Inter.-American Development Bank De la Brière, Bénédicte and Lindert, Kathy. 2005. Reforming Brazil’s Cadastro Unico to Improve the Targeting of the Bolsa Família Program. Social Protection Discussion Paper Series, 32757, Nº. 0527, June 2005. Human Development Network. Washington, D.C: World Bank 14The supplement is distributed to all children under 2 years of age, pregnant and lactating mothers that qualify for Oportunidades. Children who are 2 years of age and less than 5 also qualify for the nutritional supplement when they suffer from malnutrition or are recovering from it. 15One of the first external evaluations was performed by IFPRI. 16Sedesol, 2004. Evaluación de Impacto. 14

XIV Congreso Internacional del CLAD sobre la Reforma del Estado y de la Administración Pública, Salvador de Bahia, Brasil, 27 - 30 oct. 2009

División de Planificación Regional, Ministerio de Planificación y Cooperación. 2004. Gasto Público Social Regional 1990-2003. Chile: Gobierno de Chile Document of The World Bank. 2001. “Brazil, As Assessment of the Bolsa Escola Programs”. Report Nº. 20208-BR, March 15, 2001. Human Development Sector Management Unit, Brazil Country Management Unit, Latin America and the Caribbean Regional Office. Washington D.C: World Bank Document of The World Bank. 2004. “Project Appraisal Document on a Proposed adaptable Program loan in the amount of us$572.2 million to the Federative Republic of Brazil for a Bolsa Familia Project in support of the first phase of the Bolsa Familia Program”. Report Nº: 28544-BR. May 25,2004.Washington, D.C.: World Bank Ministerio de Planificación y Cooperación, MIDEPLAN – Sistema de protección social Chile Solidario. División Social. 2004. “Conceptos Fundamentales. Sistema de Protección Social “Chile solidario””, diciembre 2004. Chile: Gobierno de Chile Ministerio de Planificación y Cooperación, MIDEPLAN – Sistema de protección social Chile Solidario. División Social. 2004. Ficha Familia “Más y mejor focalización en una nueva etapa del desarrollo social del país”, septiembre 2004. Chile: Gobierno de Chile Ministerio do Desenvolvimento Social e Combate à Fome. 2005. Balanço MDS. Programas Sociais. Brasil: http://www.mds.gov.br/ascom/ Palma, Julieta and Urzúa, Raúl. 2005. The “Chile Solidario” Experience. Anti-poverty Policies and Citizenry: Policy Papers No. 12. Francia: UNESCO Rawlings, Laura. 2004. A New Approach to Social Assistance: Latin America’s Experience with Conditional Cash Transfer Programs. Social Protection Discussion Paper Series No. 0416. Washington, D.C.: World Bank Skoufias, Emmanuel. 2005. “PROGRESA and Its Impacts on the Welfare of Rural Households in Mexico”. The International Food Policy Research Report, 139. Washington, D.C.: International Food Policy Research Institute. Sadoulet, Elisabeth; Finan, Frederico; de Janvry, Alain and Vakis, Renos. 2004. Can Condicional Cash Transfer Programs Improve Social Risk Management? Lessons for Education and Child Labor Outcomes. Social Protection, The World Bank, SP Discusión paper Nº. 0420, December 2004. Washington, D.C.: World Bank.

15