Report No. 24500-BR Brazil Strategies for Poverty Reduction in Ceara The Challenge of Inclusive Modernization (In Two Volumes) Volume 1: Main Paper April 10, 2003 Brazil Country Management Unit Latin America and the Caribbean Region Document of the World Bank CURRENCY AND 1EXCH8ANGE IRATtES (1R$/US$) Currency Unit - Real (R$) December 2000: R$1.95 December 2001: R$2.31 December 2002: R$2.65 WEOGHTS AND MEASURES The Metric System is used throughout the report. FOSCAL YEAR January 1 to December 31 ABBREVOATOONS AND ACRCNYNS CREDE : Centro Regional de Desenvolvimento da Educac,o FUNCEME : Fundag&o Cearense de Meteorologia FUNDEF : Fundo de Manutengao e Desenvolvimento do Ensino Fundamental e de Valorizac,ao do Magist6rio IBGE : Instituto Brasileiro de Geografia e Estatistica ICT : Information and Communication Technology IPLANCE : Fundac,o Instituto de Pesquisa e Informac,o do Ceara LDB : Lei da Diretrizes Basicas MEC : Ministerio da Educaqao PETI : Programa de Erradicac,o do Trabalho Infantil PIB : Produto Interno Bruto PNAD : Pesquisa Nacional por Amostra de Domicilios PROGERIRH : Programa de Gerenciamento Integrado de Recursos Hidricos PSJ : Programa Sao Jose RMF : Regiao Metropolitana de Fortaleza SDR : Secretaria de Desenvolvimento Rural SEBRAE : Servic,o de Apoio as Micro e Pequenas Empresas SEDUC : Secretaria de Educac,o Basica SENAI : Servico Nacional de Aprendizagem Industrial SEPLAN : Secretaria de Planejamento e Coordenac,o TVC : TV Ceara UFC : Universidade Federal do Ceara UFPE : Universidade Federal de Pernambuco Vice President LCR : David de Ferranti Director LCC5C : Vinod Thomas Lead Economist : Joachim von Amsberg Task Manager : Michael Walton Brazil: Strategies for Poverty Reduction in Ceara TABLE OF CONTENTS ACKNOWLEDGMENTS .......................................................... i EXECUTIVE SUMMARY .......................................................... -ii I. MOTIVATION ...........................................................1 II. ASSESSING THE PAST .......................................................... 3 THE CONTEXT- STRUCTURAL DEFICITS AND EXTERNAL CONDITIONS .....................3 HEALTH, SKILLS AND LIVING CONDITIONS ........................................................... 5 GROWTH, INEQUALITY AND INCOME POVERTY ...................................................... 11 VULNERABILITY .......................................................... 17 GOVERNANCE, CULTURE AND INCLUSION .......................................................... 19 OVERVIEW OF SPENDING .......................................................... 21 III. POLICY OPTIONS FOR INCLUSIVE MODERNIZATION ........................ 23 THE CONTEXT- INCLUSION AND MODERNIZATION IN A GLOBALIZING WORLD .......... 27 POLICIES FOR AN EQUITABLE GROWTH PATH ...................................................... 28 TACKLING CEARA'S EDUCATION DEFICIT - A MEDIUM TERM PROPOSITION .............. 32 FASTER POVERTY REDUCTION THROUGH INCOME TRANSFERS ............................ 37 GOVERNANCE AND INCLUSION .......................................................... 47 SPATIAL IMPLICATIONS OF A RENEWED ATTACK ON POVERTY AND INEQUALITY ...... 48 FINANCING AND THE LEGACY OF GOOD FISCAL GOVERNANCE .............................. 50 THE PROCESS OF POVERTY STRATEGY DESIGN ................................................... 52 REFERENCES ................................................................ 61 Tables Table 1: Growth in enrollments in Ceara, the Northeast and Brazil, 1995-2000 (in percent) ..................................................................9 Table 2: Ceara - real expenditures by category, 1990, 1995 & 2000 (index 1990) .......... 22 Table 3: Summary of key policy conclusions .......................................................... 26 Table 4: Poverty reduction impact of economic growth: projected poverty rates for Ceara in 2004 and 2020 ................................................................. 28 Table 5: Options for Bolsa Escola in Cearc: projected poverty impact and costs ........... 39 Table 6: Projected poverty and indigence rates for Ceara in 2020 .................................. 40 Table 7: Spatial dimensions of policy recommendations ................................................ 49 Table 8: Summary of policy recommendations for equitable growth .............................. 54 Table 9: Education - recommendations ................................................................. 57 Table 1 0:income Transfers - recommendations ...................................................... 59 Figures Figure 1: Ceara's historical inheritance: a vicious cycle of disadvantage ..........................4 Figure 2: The decline in infant mortality in Ceara, the Northeast and Brazil 1950-2000 ....6 Figure 3: Rising access to water in Ceara, the Northeast and Brazil 1992 and 1999 .......7 Figure 4: Growth in access to sanitation in Ceara, the Northeast and Brazil, 1992 and 1999 ..................................................................8 Figure 5: High but falling illiteracy rates in Ceara, Northeast and Brazil, 1986-1999 .........8 Figure 6: Inequality in Ceara and Brazil-the Gini coefficient, 1989 to 1999 ................... 11 Figure 7: Cumulative growth in GNP, Ceara and Brazil, 1985-1999 ................................ 12 Figure 8: Evolution of the incidence of poverty, Ceara & Brazil, 1981-1999 ................... 13 Figure 9: The growing urbanization of poverty in Ceara, 1981-99 .................................. 15 Figure 10: Agricultural production and rural poverty in Ceara . . 18 Figure 11: Getting on to a virtuous cycle of redistributive action . . 25 Figure 12: Projected years of schooling for different age groups with an aggressive program of education and literacy, 2000-2020 ........................................................ 34 Figure 13: Distribution of earnings of agricultural workers and farmers in 1999 in relation to the minimum wage ......................................................... 45 Boxes Box 1: The health agents program ..................................................................7 Box 2 Education reform and progress in Ceara ............................................................. 10 Box 3. Projeto Sao Jose ................................................................. 16 Box 4. Better services for the poor via better governance ............................................... 20 Box 5. Key elements of an accelerated strategy of reducing income poverty and inequality ............................................................................................................................ 25 Box 6. The role of redistributive transfers in the modernization process ...................... 38 Box 7. The process of design of policies for poverty reduction and development ............ 52 1. ACKNOWLEDGMENTS This is a product of joint work between IPLANCE/SEPLAN of the Govemment of Ceara and the World Bank. The joint work was led by Michael Walton for the World Bank and Alex Araujo, then of IPLANCE. Principal contributors included Gisella Colares, Annuzia Pontes and Amelia Costa (IPLANCE) Fatima Falc&o (SEPLAN), and Gillette Hall (World Bank). Madalena dos Santos and Joachim von Amsberg (World Bank) provided key inputs and advice. M6nica Amorim and Jair Amaral (SEPLAN/CED) provided inputs on options for equitable growth. Jos6 Ricardo Bezerra Nogueira and Rozane Bezerra de Siqueira (Federal University of Pernambuco) undertook the simulation work on transfers. Francisco Ferreira and Phillippe George Leite (then of PUC-Rio, currently World Bank) prepared the background analysis on the effects of education expansion on distribution. Mark Thomas (World Bank) prepared the Annex to chapter 2 in the background material on how multiple dimensions affect poverty and poverty trends. Paul Siegel (World Bank) provided the main inputs for the section on drought management. Antonio Magalhaes and Dorte Verner (World Bank) provided ongoing inputs and advice. Judith Tendler (Massachussetts Institute of Technology) provided invaluable general advice. Charles Kenny and Juan Navas-Sabater (World Bank) prepared inputs on information and communications technology. Christina Alquinta (World Bank) provided research support. It draws on extensive meetings in Ceara and on work by Francisco Ferreira Alves (IPLANCE), Na6rcio Menezes (USP) and Mark Thomas (World Bank). The work was undertaken under the overall supervision of Gobind Nankani (Country Director for Brazil, World Bank during the bulk of report's preparation) Vinod Thomas (current Country Director for Brazil, World Bank) and M6nica Clark (Secretary of Planning, Government of Ceara, during the report's preparation). Particular thanks to Senator Tasso Jereissati (Govemor of Ceara until 2002) for posing the questions that motivated this report and for his active engagement in the process of preparation. Drafts of this report have been discussed with the Government of Ceara and have been shared with the Federal Govemment of Brazil. The analysis of the report was undertaken in the period to August 2001, and includes information that was available up to that time. While the work was a product of a collaborative process, views expressed in this document are those of World Bank and do not necessarily reflect the views of the Govemment of Ceard, the Federal Government of Brazil or the Executive Directors of World Bank's Board. This report was prepared at the request of the Government of Ceara on the basis of the work of a joint team involving the World Bank and IPLANCE/SEPLAN. The government posed two types of questions to the team, conceming the past and the future: Why is there still severe deprivation in Ceara after so many years of active development efforts? Have government programs been well-designed? Mexico office JB3-100 What can be done differently in the future? How can a more inclusive strategy be combined with the modernization effort? With respect to the analysis of the past, the report concludes that there has in fact been major progress on many dimensions of poverty in Ceara. This can be seen in terms of sharply reduced mortality, lower illiteracy, increased access to schooling, water, electricity, roads and communications. In all these areas the legacy of history in the 1980s was of severe deprivation-especially given Ceara's average income level. But the pace of progress in the past fifteen years has been impressive in many areas. Indeed, in most cases advances have been faster than in other states in Brazil, transforming both current conditions and future prospects of many of the poor. This is a tribute to the effectiveness of the state, often working with the Federal govemment and other societal actors. Where progress has been disappointing has been in income poverty and inequality. There has been only a slow decline in income poverty, and no significant change in inequality since the late-1 980s. Why was this? The report concludes that this was primarily due to the influence of history and of extemal conditions. First, the historical inheritance was of a vicious cycle of disadvantage, in which an unequal structure of work and physical assets, of education and human assets, of power and culture, and of vulnerability interacted in a fashion that perpetuated income inequality. It is a common international phenomenon that income inequality changes very slowly because of the difficulty of breaking these cycles quickly. Second, Ceara experienced adverse extemal developments due to Brazil-wide conditions, especially with respect to slow growth and macroeconomic instability and the widening in the retums to skills in the past ten years owing to the pattem of technological change and global developments, with stagnant retums to the unskilled. Where the state was active in areas affecting incomes, it was effective, notably with respect to fiscal consolidation, in sustaining a growth rate above the (low) Brazil-wide average, in attracting industry, and in programs, such as Projeto Sao Jos6, that have brought modest but important benefits to a large proportion of poor rural-dwellers. But ii the depth of the inherited, vicious cycle of disadvantage and the adverse external conditions meant that change has been slow-especially with respect to income inequality. What can be done in the future? Is a more inclusive strategy consistent with the modernization challenge in a world in which global forces are increasing competitive pressures? The report argues that a well-designed redistributive strategy is not only consistent with modemization, but is an important complement. Societies that effectively manage inequality and insecurity typically have deep and broad educational systems, and responsive and inclusive political and institutional structures. These can be sources of competitiveness and resilience in the global economy. But there is no magic solution. Success will depend on a range of complementary actions to tackle the various areas of disadvantage to create a virtuous cycle of change that is both redistributive and growth-oriented. In particular, the report finds that growth alone will be a weak instrument of poverty decline, precisely because of the extent of inequality. Thus the report proposes complementary efforts in four areas: * Achieving a more equitable pattem of income and job growth, especially through continued support for intensifying agriculture, a shift in industrial policy to support areas of existing potential in industries and services, a broader strategy to support micro, small and medium enterprises, and support for city strategies, including in the interior. * Continuing and deepening the effort to get equitable skills development, through ongoing plans to upgrade the quality of schools for the poor, increased support for early childhood development, and incentives for poor children to stay in school. * Use of redistributive transfers for the poor and indigent, as a means of complementing human capital development-building on Federal Bolsa Escola plans--and as part of a reformed approach to drought management Strengthening institutional change, especially at municipal level, .to tackle problems of local capture or inefficiencies, combined with special efforts at reaching the socially excluded. Finally the report emphasizes the importance of the process of policy formulation and implementation. There is a great deal of development activity going on in Ceara, but there is also a great deal that is not known about both the direction of this effort, and even more, the consequences. A process of ongoing updating of the vision, formulation of goals, experimentation and evaluation, integrated with participatory debate on what does and does not work, is both desirable for Ceara, and increasingly recognized as integral to the process of development in rich and poor countries. iii Brazil: Strategies for Poverty Reduction in Ceara - The Challenge of Inclusive Modernization 1. MOTIVATION 1. This report was prepared at the request of the Government of Ceara in response to a central developmental puzzle. For over a decade the govemment had been pursuing an active development agenda that sought to increase the well-being of all the population. The government has combined successful efforts at fostering economic growth - at 4%, the state's average annual growth rate has exceeded the national average for Brazil throughout much of the 1990s - with significant improvements in fiscal management. It has re-oriented expenditure towards investment in basic services such as health, education and infrastructure. It has earned a reputation as a dynamic, modernizing state, attracting new industry with pro-business tax incentives on the one hand, while experimenting with innovative efforts to raise efficiency of service delivery through local participation and outreach among the poorest on the other. The state has clearly been effective in a range of central arenas of economic and social policy. 2. There has been major progress in many dimensions of well-being including amongst the poor, notably in health and education. Yet deprivation has persisted, especially in terms of income poverty and inequality. This was brought home sharply by the unexpectedly early arrival of drought in 2001, that threatened the livelihood of some of the poorest families in the state. 3. This contrast between multiple programs and persistent deprivation is a feature of the whole of Brazil, but it is more marked in the North East where levels of deprivation are substantially greater. In the past, Ceara has been a leading state in Brazil-and internationally-in innovation in development solutions. 4. The government in 2000 posed the question of how to explain the coexistence of development activism and persistent material deprivation, and of what could be done differently. It asked the World Bank to participate in the work and prepare the report, both to assure greater objectivity-not least in the context of an often highly politically charged discourse over poverty and inequality in Ceara-and to bring lessons from cross-country experience. The work was undertaken jointly with IPLANCE and SEPLAN. Parallel work by an Israeli team focused on the development of the interior of Ceara, with particular attention to secondary cities: the two reports are highly complementary. 5. The issue posed by the Govemment can be broken down into two kinds of question: Why is there still severe deprivation after so many years of active development effort? Have government programs been well-designed? What can be done differently in the future? How can a more inclusive strategy be combined with the modemization effort? 6. Section 11 outlines the assessment of the first question. The basic answer is that there has been major progress in many @ear ha dimensions of well-being, and government programs and policies MaIde MWSior have, for the most part, been well-designed in support of this. pSOgress in Slow progress, especially in income poverty and inequality, has §nany largely been due to the sheer weight of the structural inequalities d and problems that flow from Ceara's and Brazil's history, @1 f- $ ng$ and combined with setbacks from external factors. goverammeMn progff)S!1ga haew 7. Section III then turns to the future, and the second question. been A severe inherited neglect of the skills, health and overall living conditions of the bulk of the population, especially of the poorest in rural areas; > An unequal structure of asset ownership and distribution of employment, with a large fraction of the population living in semi-arid rural areas with inadequate complementary land and material resources-providing a reserve army of poor unskilled migrants for urban areas; > Unequal and weak institutions, especially a history of concentrated power, cultures of inequality and, in the 1980s, a bloated public employment payroll that was a source of fiscal problems; and > High levels of vulnerability to the weather, to economic conditions in the rest of Brazil, and to the violence and insecurity associated with rapid urbanization. Overlaying these were two adverse external influences for poverty or distribution: > Slow macroeconomic growth and instability in Brazil, that contributed to an environment of slow average growth for the past two decades, and recession in 1998-99 after the East Asian-Russian financial crisis; > Rising premia on skills Brazil-wide combined with a flattening of returns to basic education--implying stagnation in unskilled wages-- which is generally attributed to technological change, perhaps in combination with international integration. 4 11. It is important to assess the government's past actions to reduce poverty and inequality in the context of these historical conditions. This is done in relation to each of the four areas of structural deficit, while recognizing that there are powerful complementarities for poverty and inequality between the various areas. We finally provide an overview of spending that cuts across these categories. HEALTH, SKILLS AND LIVING CONDITIONS Over the past 12. The capability to live a long and healthy life and have the skills two decades, the to participate fully in work and living are fundamental aspects of infant mortality freedom from poverty. Ceara's historical legacy was bleak for rate has fallen both, but gains have been substantial and state efforts impressive, from 150 to 50 This is probably the area of greatest advance in tackling parts of per 1,000... a the inherited cycle of disadvantage, in part through use of remarkably fast innovative means of dealing with aspects of the historical legacy of unequal institutions. iner by international standards 13. With respect to health, the infant mortality rate-one of the most reliable overall indicators of health status-was historically extremely high in Ceara and elsewhere in the Northeast. Moreover, progress in Ceara had been significantly slower than elsewhere between 1950 and 1980 (Figure 2). The 1980 rate of 150 deaths per thousand was substantially higher than much poorer countries in the same year, including India (115), Haiti (123) and Tanzania (108).3 However, there was a major acceleration in progress in the 1980s and 1990s, with a reduction to about 50 by 1999, catching up with the rest of the Northeast.4 This was a remarkably fast decline by international standards. 3 See World Bank. 2000. World Development Indicators. 4 This uses IBGE's estimates from Census and PNAD data, rather than births and death registration, as used by the Secretary of Health. Both methods can be defended, but the former was used in order to have comparable data with the rest of Brazil. There remains significant uncertainty over the quality of the underlying data, so interpretations have to be made with some caution. 5 Figure 2T he decline in infant mortality in (Cears, the boriheast and l3ra2il 1950 - 2000 EvoIugio da Moritalidade infantil em mortes por mil nascidos vivos 250,0 200,0 - 50,0 --- --------- --- ------ 1950 1960 1970 1980 1990 2000 Brasil Norde-ste Cear Source: IBGE/Diretoria de Pesquisas/Departamento de Indicadores Sociais 14. What lies behind this remarkable decline? Intemationally the major influences on infant mortality are typically incomes (especially of the poor), education (especially of women), water and sanitation, and preventive health services. Since incomes were only growing slowly, and the big education push came later, it is plausible that factors in environmental and basic health were of particular importance. Of special note in Ceara has been the highly innovative program of health agents, introduced in the late 1980s, which brought core preventive and even basic curative services into the community, with a strong emphasis on reducing infant and child mortality (See Box 1). This has had multiple effects on health and beyond. As one indicator, through this and complementary programs, under-5 vaccination was virtually universal by 2000, compared with 25 percent for measles and polio in the late 1980s. 6 Box 1: The health agents program The health agent program, introduced in 1987, is a remarkable example of the scope for innovative state action to create a highly effective cadre of field workers, closely tied to communities, and with a major impact in terms of results. By the early 1990s, infant deaths had dropped sharply (see Figure 2), vaccination for measles and polio had jumped from 25 to 90 percent, and 7000-odd agents were visiting 850,000 families every month. The program is still going strong, even as some of the functions of the health agents have shifted--for example, out of basic curative care as the number of higher health professionals has expanded, and into areas dealing with broader aspects of family welfare. The real question is how the government could create, so quickly, such an effective and motivated cadre of field workers. The answers are complex, but it is clear that action by the state government was key, especially by a new leadership team in the state Department of Health who came in with a new vision (inspired in part by the thought of Paulo Freire) and designed innovative mechanisms for implementing this. While the program was decentralized, key functions were centralized, including a highly competitive recruitment process, active radio campaigns, and a major focus on-and celebration of- results. This required going against the traditional interests of mayors and health professionals. And while there was a high degree of state activism, implementation was in part demand-driven--communities increasingly demanded of non-participating mayors that they join the program. The results in health were brought about via the effects of the program on worker motivation and trust from the community. It remains a model, not only within the health sector, but for many categories of public sector action. Source: Tendler, 1997, Good Government in the Tropics, and discussions in Ceara. 15. Water and sanitation are also important factors behind health status-as well as affecting the quality of life directly. Here too Ceara started with higher levels of deficit, but has been catching up fast. This is especially the case for water (Figure 3) where coverage is now over 80 percent, with access to sanitation also catching up from a much lower level (Figure 4). Expansion in these areas is primarily due to government programs. Figure 3: Rising access to water in Ceard, the Northeast and Brazil 1992 and 1999 Condis6es Habitacionais Abastecimento de agua adequado 100.0 92 4 89.1 88.6 90.0 _ 80.5 80.0 _ atky _80. 80.0 :˘ 69.9 70.0 60.0 50.0 1992 1999 M Brasil 3Nordeste l:Ceara Source: IBGE/Diretoria de Pesquisas/Departamento de Indicadores Sociais 7 Figure 4: Growth in access to sanitation in Ceark, the Nlortheast and Brazil, 1992 and 1999 Condlies Habitacionais Esgotamento sanitario ligado a rede coletora 60.0 52.5 50.0 48.0 40.0 30.0 ~~~~~~~~ 28.1 30.0 ~~~~~~~~24.0 20.0 4$t~l 19.3 4 8.5 10.0 1 1992 1999 |lBrasil ONordeste OCeara Source: IBGE/Diretoria de Pesquisas/Departamento de Indicadores Sociais 16. In education the deficit from historical neglect was of equal severity throughout the Northeast. In the mid-1980s, adult illiteracy was 40 percent throughout the region-in this case comparable to India's male illiteracy rate in the 1980s (though better than that of Indian women), but significantly worse than Indonesia, where illiteracy was less than 30 percent, even though Indonesia's average income was much smaller than Ceara's at that time. Since the 1980s, illiteracy has fallen substantially to the mid-20s-at about the same pace as elsewhere in the Northeast. Figure 5: High but ffafing illiteracy rates in Cears, the Northeast and Brazfl, 1986-1 999 Evolug;o da taxa de analfabetisnmo Populagao com 15 anos e mais - em % 45 40 38.7 38.3 35 32.7 3. 30 26.6 27.8 25 20.0 20 17.2 15 5 1986 1992 1999 OBrasil ONordeste OCeara 8 17. A major push on basic education occurred in the 1990s, in Ceara as in the rest of Brazil. Enrollments in basic education jumped from 57 to 98 percent between 1990 and 1999. This reflects major efforts and increased spending by both the state and federal government, with more rapid growth in enrollments in both basic education (ensino fundamental) and upper secondary schooling than the Brazil and Northeast average, notably in the last five years (see Table 1 and Box 2). Illiteracy rates move much more slowly because the bulk of the adult population continues to suffer from the severe educational neglect of previous decades. There have been efforts at improving adult literacy, but, as elsewhere in the world, the results are less than clear-it is much more difficult to teach adults basic skills than children. Table 1: Growth in enrollments in Ceara, the Northeast and Brazil, 1995-2000 (in percent) Basic education Upper secondary education Ceara 34.5 99.0 Northeast 23.3 68.0 Brazil 9.3 52.4 Source: SEEC/MEC 18. The remaining challenge. While progress has been impressive, there is a continuing challenge in both reducing the historical deficit, tackling continued inequalities in deprivations in skills and health, and upgrading human capital for new demands. Mortality rates are much higher for the income poor, and there are also significant correlations between income poverty and low skills. In particular in education there is both a problem of high levels of illiteracy of the existing workforce (still 17 percent for 15-39 year-olds, compared with 16 percent for the Northeast) and weak skill acquisition and dropouts within schools owing to the interaction between low quality and, for the poor, continuing demands for child labor. 9 Box 2. Education reform and progress in Cearai The Govemment of Ceara has been a forerunner among other states for its innovative and committed approach to education. Beginning with a state program called "Everyone for Quality Education for All" (Todos pela Educaqdo de Qualidade para Todos) in 1995, Ceara has continuously implemented initiatives to improve both the quantity and quality of its education system. In 1996, three programs were launched to target the quality and efficiency of the state's basic education system. Ceara also has attempted to administratively streamline the Education Secretariat by creating 21 Regional Educational Development Centers (CREDEs) in 1996 to coordinate municipal educational policies for the state's microregions (21). Ceara's strategies to modernize school management have been disseminated to other states in the region. Between 1990 and 1999, educational enrollment rose by 42% and today Ceara's school system includes almost 100% (98.3%) of the state's children between the ages of seven and fourteen. This evolution has been significantly faster than the regional and national average (Table 1). For example, between 1995 and 2000, enrollment in Ceara's basic education cycle (ensino fundamental) rose 34.5% while in the Northeast region the average was 23.3% and in Brazil only 9.3%.5 Measures are currently being taken to resolve historical problems of general capacity constraints and the inadequate quality of the education system.8 First, policies have been oriented to address the persistence of high repetition rates and age-grade distortions. Only 46% of the students complete the 8 grade within the expected time-frame of eight years and 56% of the students are two or more years older than they should be for their respective grades. Improvements in this area have been ongoing and are notable. Currently, the distribution of students within each grade has become more balanced with an average of 12.5% of the students at each level in the basic cycle. Second, teachers' professional training and development is being advanced. Professional and pedagogical training has been improving since the original reforms were implemented in 1995. The number of professors with university degrees or higher has risen from 44% to 62% while teachers with only a secondary and/or a technical education has declined from 56% to 38%. Nevertheless, 29% (11,406) of Ceara's basic education teachers, mostly in rural schools, are considered 'lay teachers" that do not meet the 1996 National Education Law (LDB) requirements.8 Third, efforts are improving synergies between the state and municipal education systems. The delivery of education has been complicated by the overlapping oi municipal and state education systems despite the government's attempt to transfer education responsibilities to the municipalities. For example, 66% of public school enrollment in the basic education cycle is supplied by municipalities while the state continues to serve 23% of the students. The state govemment knows that its noteworthy advances and trend setting educational policies require further advances, not only to improve the education system, but to address poverty and broader social issues. In Ceara, where 84% of the poor live in households headed by a person with less than four years of schooling and 30% of the population over 15 years of age is illiterate, education reform is seen as the key instrument to end the cycle of poverty. As a result, the Govemment has implemented a comprehensive state-based strategy called the "Ceara Strategic Human Resource Qualification Plan, 1999-2003" which is being funded by state, federal and international agencies. 9 This strategy has been designed to improve the performance of the overall education system. 5These figures are from: SEDUC. 2000 A Escola do Novo Mlemo State of CearA, SEEC, MEC 6ThIs figure is taken from INEP 1999 fnformago5es Educacionais Estado do Ceard Reuniao dos Secretdnos Estaduats de Educa,do, 4-5 February 1999. 7 These figures are from. SEDUC. 2000. A Escola do Novo MilWn,o. State of Ceara, SEEC, MEC. t Source: World Bank 2000. Report Number 21428BR: Project Appraisal Docunent. State of Ceardt. 29 November 2000. 9 This framework, presented by the Govemo do Ceara in 1998, was prepared by an inter-institutional task force with representatives from four state secretariats. A 12-member team, formed from these four secretariats, was established by the Govemor While its activities have been initiated, problems are centered around budget hlmitations (World Bank 2000) 10 GROWTH, INEQUALITY AND INCOME POVERTY 19. Progress on income poverty is a product of the interaction between growth and inequality-the more unequal a society the smaller the gains to the poor from growth, unless the incomes of the poor grow significantly faster than other groups. Ceara's legacy is one of extreme inequality. Brazil is one of the most unequal countries in the world-with a Gini coefficent (a synthetic measure of inequality) of 0.6. Ceara, as well as the rest of the Northeast, has actually slightly higher inequality than the Brazil average. And there has been no change of significance, either in Ceara or Brazil, in the past 15 years (See Figure 6). This is a largely a reflection of the great difficulty in breaking the cycle of disadvantage with respect to incomes-something that Ceara shares with most unequal societies in the world. Figure 6 Inequality in Ceara and Brazil-the Gini coefficient, 1989 to 1999 Coeficiente de Gini 0.750 0.700 --- - - -- - - - - -- - ---- Basl- - -- - - -- - ' : ' ' ' ' ' {I ~-CF Ceara 0.650 -- - - - - - -- - - -- -- -- -- _-----_L___-- ____4 __L___ _ 0.600 --------0 ___ 0.500 - 1989 1990 1992 1993 1995 1996 1997 1 998 1999 Source: IBGE, microdata from PNAD 20. In the past 15 years Ceara succeeded in growing above the Brazil average, despite the high levels of integration in national financial, product and labor markets (Figure 7). Between 1985 and 2000 Brazil grew at 2.5 percent per annum, while Ceara grew at 3.6 percent annually. This was an important achievement, that can be linked to the effective fiscal consolidation in the late- 1980s, infrastructure expansion, success in attracting industrial investment and sectoral policies in tourism and urban growth.10 10 This is consistent the empirical analysis of differential growth and poverty performance across Brazilian states by Naercio Menezes and Mark Thomas. I I I I I~~* Figure 7: Cumulative growth in GNP, Cearci and lBrazil, 1985-1999 Evolluio Real do PIB 1985=100 170.0 . I I I I I I I I I I iO -eI 150.0 ------ t '--- - - --- , | , t , : W | , _ , , ~~~~~~~~~~~BrasilI 140.0 -- - - - -4 _1 / ,,Nordeste a' a' a' a' _o_aE a' a' a' a' a' ' - Ceara 1 30.0 ---,--*--- : * , 120.0 - ---- r-~n-<~ ------t-- 1 10.0 - _v -. - _L_ J---L_ Sources: 85/98 IBGE; 99/00 Brasil IBGE, CearI Iplance 21. By contrast, as shown in Figure 6, there has been no aeSr grew a a significant change in inequality in Ceara. This is a common rat@e of 36% Per phenomenon in the world: changes in inequality are typically very yearO bet ween slow, except in periods of radical social and institutional change. V 985 and 2000 Where inequality has fallen in "normal" times in other societies, this y coan,g has usually been associated with major expansion and hre hlas b,een equalization in educational attainment, especially when combined with rapid growth in unskilled or semi-skilled work in agriculture, ' a2.,@ dn industry and services.11 Ceara's expansion in education has been llequal'ty, sudch too recent to have a significant effect on the composition of skills that desPite and has occurred in a period in which the overall returns to high growth, income levels of skills were rising and returns to basic skills were falling in poverty hag Brazil. At the same time agricultural productivity has remained laflen very very low, and growth in non-agricultural activities would have had a to have been at the very high levels experienced during the fast slowy0 growing episodes of East Asian countries and Chile to transform employment prospects. The classic recent examples are Korea and Malaysia in the 1970s or 1980s and the United States in the middle of the century; Colombia is notable for being a Latin America country that experienced a significant decline in inequality in the 1970s and 1980s. It should be noted, however, that in Colombia, Malaysia and the United States, these declines were followed by periods of rising inequality. 12 22. The net effect on income poverty was a slow long-term decline-with an acceleration in the wake of the Real Plan, but little change since 1996. (See Background Material Chapter 2, especially the Annex, for further discussion.) There are many poverty lines used in Ceara and Brazil, but all those that are held constant in real terms display the same trends (See Figure 8) The trend pattem can be divided into three periods: cyclical changes in the 1980s, a significant decline between 1990 and 1995, and stagnation between 1995 and 1999. For the whole period Ceara's pace of poverty decline has been the same as for Brazil, but has been better than the Northeast average. Figure 8: Evolution of the incidence of poverty, Ceara and Brazil, 1981-1999 (in percent) EvoluVao da proporgao de pobres 90.0 8 0 .0 4 - ,~>y+T4__4_4 ,_, 40.0 - - - - r - - - - a - - - - - -- - - -r --r-- 40 .o I- - - -- - - 4 A, - - ---,r .,,, 3o.0 N M V Ln tD r | oD ci o N m nwBai 200 -- r- r- -4 -4 -4 - - 4-: - 4 - 1- 1- 4 - r - -r--r- -4 _-4 - n-G- Cea- r6- Note: the estimates for 2000 are preliminary and indicative since there was no PNAD for this year; they are based on the 2000 urban employment survey and a statistical matching of the past relationship between this survey and PNAD. Source: IBGE/PNAD, on the basis of the poverty line used by IPEA, of R78 in prices S ,o Paulo in 1999. 23. The 1995-1999 stagnation was a product of mixed trends within Cearci. Rural income poverty continued to decline slowly over the period - bolstered by a state-run supplement to the Federal drought relief program (during the drought year of 1998, rural poverty rates for the first time remained stable). Urban poverty, however, rose, with an upswing in 1999 reflecting the Brazil-wide slowdown for this year. There was probably a further decline in poverty in 2000 with both overall growth of 5 percent and a good harvest; but this is unlikely to 13 have been sustained in 2001, owing to the overall growth slowdown and drought.12 24. While there has been progress, income poverty remains broad and deep, and improved much more slowly than social indicators of well-being during the 1990s. According to govemment of Ceara (IPLANCE) estimates, roughly half the population lived below a poverty line of half a minimum wage per person in 1999, and over one fifth lived below an indigence line or a quarter of the minimum wage per person.13 The great depth of poverty is indicated by the gap between the mean incomes of the poor and the poverty line: average per capita incomes of the poor are over a third less than half a minimum wage in rural areas, almost 40 percent below in urban areas in the interior, and over a quarter below in Fortaleza.14 There is a range of altemative means of estimating the poverty line which yield slightly different estimates for current poverty rate, but what really matters is trends over time (whether poverty is rising or falling) and the structure of poverty (who is poor). 25. While there is significant heterogeneity amongst the income poor, there are some clear overall features of the structure. The deepest poverty is in rural areas, where roughly three-quarters of the population live in poverty, in households primarily dependent on subsistence agriculture and landless agricultural workers. Families whose household head is a small farmer have a poverty rate of almost 80 percent and those whose head is a agricultural laborer an even higher rate-of almost 85; where the head of the household works on rural non-farm activities the rate is almost 60 percent The corresponding urban poverty rate is about two-fifths. However, in terms of numbers, more poor people live in urban areas (2 million) versus rural areas (1.7 million), a reflection of Ceara's on-going process of urbanization. As Figure 9 illustrates there has been a steady urbanization of poverty over the past two decades, following the overall process of urban growth. While the metropolitan area of Fortaleza has the lowest share of the poor at around 25 percent, this share has been rising over time. A concentration of poverty among the young is also apparent from the surveys: in both rural and urban areas poverty rates are highest among children and youth, and the population ages 0-18 makes up over half of all the poor in Ceard.'5 Poverty is also strongly associated with the level of education-though here the direction of causation can be in both directions. 12 Note the 2000 poverty numbers are preliminary-see note to the Figure 8. 13 This report draws on poverty estimates calculated by the government of Ceara policy research institute (IPLANCE). It does not endorse a particular method of calculating the poverty line. For more details on this see the background material, World Bank 2001e. 14 The gap used here is the mean distance between the per capita income of poor households and the poverty line used--in this case half a minimum wage. (This is technically the ratio between the "poverty gap", Pl and the poverty line Po in conventional usage.) 15 Note, however, that this is on the basis of a simple per capita income calculation with no allowance for different consumption "requirements" of children, economies of scale, or distribution within the household. Allowing for these would be expected to change the quantitative results, but not the qualitative message. 14 26. As a final comment on trends, in assessing the overall level of well-being of the poor, it is important to recall that income poverty is onl K one factor. All analyses of poverty-and the views of the poor themselves --emphasize the multi-dimensional nature of well-being. As the previous section illustrated, in dimensions relating to health, education and a variety of areas of service delivery, there has been significant progress. While there are complexities in aggregating these into a synthetic index of poverty, since income poverty improved very slowly and other aspects substantially, there was clearly progress on average for the poor in the 1990s. (See Annex to Chapter 2 of the background papers). Figure 9. The growing urbanization of poverty in Ceara, 1981-1999 (percentage shares) Incidencia da pobreza - Cearhi Segundo situag5o do domicilio - em % 100.0 90.0 70.0 -. ~ ~ , . 60.0 .. ., 50.0 40.0 30.0 20.0 10.0 - (N sr 1) 'n N.o a) oo 0o rN m 'n N. a) o a) ON a) a) a) a) a a0 0) o CA a) Os a} 0 O C -4 -4 -4 -4 -4 -4 -4 -4 -4 -4 -4 -4 4 4 -4 -4 i * Rural 3 Urbano G RMF Source: IBGE/IPEA, based on microdata of PNAD. 27. How does the lack of change in inequality and consequently slow changes in poverty relate to the government's sectoral programs and policies? 28. In rural areas, there have been a significant number of programs trying to find solutions in adverse ecological conditions. The two signature programs oriented to small farmers are the Projeto Sao Jos6 and Reforma Agraria Solidaria. Projeto Sao Jos6 is a broad-based program that provides economic services to rural communities, linked to participatory mechanisms. The overall assessment is that it has brought modest gains in incomes, and significant gains in other dimensions of well-being (access to water, electricity) and in social capital (see Box 3). Within its own terms-of a broad-based program reaching a 16 See Narayan et al. (2000.) 15 large population with modest spending per household-it has been a success, making an important difference to the lives of many of the poor. But it is not a program that will transform levels of productivity. The Reforma Agraria Solidaria is an innovative program to provide land to small and landless farmers using participatory mechanisms and market-based land transfer processes. It does have the potential to transform livelihoods, but is likely to only be applicable to a relatively small group of farmers, especially when linked to irrigation. The PROGERIRH program is designed to get more efficient water use in the state, and-as one aspect of this-link water to good soils and encourage migration by small farmers. It has the potential to transform livelihoods for subgroup of the poor. Box 3. Projeto Sio Jos6 Projeto Sao Jose (PSJ) is a community-based rural poverty reduction program, estimated to have reached over 70% of Cearc's rural poor in the past 10 years. The project provides grant financing for basic services (electrification, water supply systems, rural roads, etc.) to communities that have rarely been reached by traditional, centralized delivery mechanisms. It also provides these services at a lower cost, as result of the following innovative characteristics of the program: (a) contribution of low-cost community labor; (b) selection of adequate lower-cost technologies; and (c) procurement and supervision of works by beneficiaries themselves. While the estimated annual benefit per household is small relative to the income gap of the rural poor in Cearc (10% of the annual household income), and thus insufficient to move households over most commonly used poverty lines, the benefit per beneficiary can still lead to significant improvements in the quality of life. One of the most important contributions of PSJ, however, may be the fostering of local social capital and links to municipal governance. The program's innovative design, espoused by the state government, is increasing participation and has the potential to enhanced local democratic processes in Ceari. However, this remains an area of considerable challenge, in light of the continuing pressures for capture by powerful groups that are embedded in local societal and govemment structures. The specific participatory processes also need to be further integrated within a broader program of improved municipal governance. 29. The major employment-focused policy in urban areas is the industrial subsidies program. This appears to have been successful in attracting industries to locate in Ceara, and to have fostered a dispersal of some of these to small towns in the interior. However, costs per job created are high and incentives to locate in the interior appeared to have fostered wide geographic dispersal rather than concentration around areas of existing potential. At the time of writing the government was reviewing its policy with a view to revising it to reduce total spending levels and to link it more closely to the growth potential of secondary cities. It is also worth emphasizing that the greatest past employment growth has been in the services area, in activities such as tourism to commerce, and these will probably continue to be the main areas of job creation. 16 30. The remaining challenge. The government correctly identified slow progress in income poverty and inequality as a major challenge for Ceara. Review of the past experience suggests the persistence of deprivation in this domain is not due to weak program design in key poverty-focused activities, such as Projeto Sao Jos6, that are bringing significant gains to many of the poor. It is rather due to the sheer difficulty in obtaining rapid, labor-intensive growth in the context of adverse agricultural conditions, slow overall growth and a rising premium to skilled relatively to unskilled worker. Despite the educational push, raising the skills of the workforce always takes time. However, the principal sectoral policy in industry had a relatively weak employment orientation and there is scope for redesign-as the government recognizes. Moreover, as will be discussed in the forward-looking section, the benefits of better growth prospects take time to reduce poverty, and there is scope for complementary redistributive transfers to the poor. VULNERABILITY 30. Insecurity is a central aspect of the experience of poverty. The risk of income losses is a pervasive concern of poor people everywhere, whether due to drought, ill health, lack of employment, or old age. But there also broader concerns over the personal insecurity of living, for example with respect to violence, vividly seen in poor urban areas, but also in rural parts.17 31. In Ceara there are two major areas of public action that have had a large impact: 32. First there is the Federal Previdencia Rural, which is a benefit for all farmers over 65. Analysis for this report finds that this is a highly successful program that is reaching virtually the entire elderly poor population in rural areas.'8 The program has an impressively strong positive effect on poverty rates among the elderly rural poor, for whom indigence has been virtually eliminated (their indigence rate now stands at 4%, compared to 42% for the rural population as a whole). Simulations indicate that the rural poverty rate would have been 14 percentage points higher than it was in 1999 without the income receipts from this program-with some 2 percentage points being due to gains amongst others living in households of pensioners. The indigence rate would have been 18 percentage points higher (some 6 percentage points due to these positive spillover effects). 17 See Narayan et al. 2000. Voices of the Poor and the associated Brazil case study. 18 See the background material to this report, World Bank 2001e, for documentation of this analysis. 17 33. Second, drought relief has been a major category of spending (from both Federal and state resources) in years in which the rains fail. This has been controversial in the past, with allegations of capture by local elites and construction companies, and of spending on projects that benefited larger landowners or were of low economic value. Here again Ceara has innovated, notably in the 1987 drought when it completely changed the mechanism of managing drought relief, put the agricultural extension service in the lead, and succeeded in having effective community engagement and reaching a much broader population group.19 In the 1993 and 1998 droughts, however, drought relief largely shifted back to being channeled through municipalities, and there have been persistent concerns over weaknesses in the program. The govemment's plans for revision in the 2001 drought are taken up in the forward- looking part of the report. Despite the concems, there is some indirect evidence that there was significant protection for the poor-while agricultural production plummeted in 1998, rural poverty did not rise (Figure 10).20 Figure 710: Agricultural production and rural poverty in Ceari PIB Agropecuirio e Pobreza Rural Serie hist6rica 1985/99 - 1990=100 140.0 Il l l l l l l l 1 3 0 .0 7 0 _ 0 _ O _ O_ C O_ * _ _ L_O I I I 1 1/\l 1 1 I I \ 120.0 _|__,___F-S --,___ I0 I - r4 I4 I4 - ,, 110.0 ~T- TX-k --t---l-rt- T \1n / , co ,n ,D _4e e n r o 0.0 -\-c co -a% (A -% - - |- -ON -- a% lN l l ON l% l ON l GI l ln a, l --Pobreza Rural --PIBAgropecu6rio 34. The continuing challenge. There are two kinds of challenge with respect to vulnerability. First, there can be improvements in the design of existing programs, in particular with respect to the drought management program. Second, there are significant gaps in coverage: pre-school children, children forced into work and those vulnerable to falling into poverty in urban areas, both of working age and the old. 19 See Tendler (1995), for an account of the 1987 drought. 20 This is, of course, affected by private coping mechanisms. Moreover, the overall poverty rate is a poor indicator of welfare change amongst the poorest, who are likely to be most affected. 18 GOVERNANCE, CULTURE AND INCLUSION 35. Weak governance and cultures that exclude can reduce progress in poverty and inequality both indirectly, via the impact on overall growth and development performance, and directly, via the lack of influence of poor on government action, notably in the area of service delivery. Problems in both areas were part of Ceara's-and Brazil's-historical heritage: as evidenced in the inefficient and bloated public employment levels, traditions of local coronelismo and the "drought industrialists" that traditionally captured gains from drought-related spending. (See Box 4 for an account of theses issues and mechanisms for dealing with them.) 36. Questions of culture can form part of the problem in two ways. First, political and social cultures of inequality typically form part of unequal societies-and are seen, for example, in the historical importance of clientelistic structures in Brazil (and other parts of Latin America). Second, the patterns of adaptation to adverse conditions by poor or excluded groups can sometimes lead to cultures with negative consequences for their own development-as in the development of cultures of drugs and violence in many parts of urban Brazil, including in Ceara.2' Of course, many aspects of the culture of poor people-whether in the rural sertao or in urban favelas has great intrinsic value and strength. 37. Actions of the govemment of Ceara. With respect to state-wide measures to tackle the history of weak economic management, the government's action in restoring fiscal order and rationalizing public employment was impressive and effective from the late 1980s. Debt service fell from 44 percent of spending in 1987 to about 10 percent in the 1995-99 period, freeing resources for real spending on services. 21 See Di6genes. 1998. Cartografias da Cultura e da Violencia. 19 Box 4. Better services for the poor via better governance There are a range of issues in getting efficient and inclusive govemment, though clientelism is of particular concem to issues of unequal access to public services by the excluded. Clientelism means the personalized provision (or promises) of services by individuals or groups in power, in return for political support-in a "troca de favores' One Brazilian political scientist characterzes clientelism as "a system of resource allocation and decision-making that many have seen to be the most stubbom political problem in Brazil".22 Cearn has at various times taken action to combat clientelism. There are two broad approaches to dealing with problems of local capture and unequal control of programs. The first involves use of a universalistic, formula-dnven policies, which by their intrinsic design are hard to capture. The second involves the development of institutional mechanisms that mobilize social, political and administrative pressure to assure effective and equitable local policy choices and implementation. This can, over time, lead to genuine changes in local political culture. Successful examples of the first approach in Brazil include the previdencia rural, that has been referred to in the text, and the supplemental funding for teachers salaries under FUNDEF. In both cases eligibility is universal and automatic for clearly defined categores, either of individuals, in the first case, or institutions in the second. There is sufficient public awareness of this 'rght" and commitment by the state to act upon it to ensure high levels of effectiveness in the transmission of the resources to eligible groups. The second approach is probably more challenging, but also of more fundamental importance in the long run. This is illustrated by Ceara's experience in a number of specific areas-see the brief accounts in the text-as well as the possibly deeper changes that have been implemented in Porto Alegre, for which participatory budgeting (orgamento participativo) has been a central instrument.23 Both these experiences show that inclusive local governance is by no means merely a question of decentralization-with-participation, but is a complex process of change in mechanisms and political culture. Strong central action (at state and city level respectively), and the development of institutional mechanisms for maintaining the support of different middle class and elite groups-as well as the poor-in the process, is important, as are vigorous campaigns on information and transparency. The complexity is illustrated by the wide range of experiences in orgamento participativo elsewhere in Brazil-some good and some bad-and by international experience. For example, one of the more radical approaches to decentralization and participation is Bolivia's 1994 Ley de Participaci6n Popular. While this has clearly brought benefits in some areas, there is evidence (from the municipality of Sucre) that where clientelism was deeply embedded, local elites and traditional practices persisted of even worsened after the reforms.24 The two approaches are complementary. Which approach is most effective depends on the type of measure. At least some measure of transparency and local democracy is necessary for success of the formula-drven approach. 38. Of equal importance has been innovative action in Ceara to begin to tackle weaknesses and clientelism at local levels. This has typically occurred within programs through proactive actions by the state to mobilize both local level workers, and involve communities and local elites in a fashion that fostered better performance by local govemment workers. The health agents program is the most famous example of this (see Box 1), but others include activities of the extension service in the late 1980s and early 1990s and of agents such as 22 See Abers 2000, Inventing Local Democracy, p. 1 I 1. 23 For in depth assessments of Ceara see Tendler (1997) and for Porto Alegre, see Abers (2000). 24 See Blackman 2001 Popular Participation in Bolivia, Rights versus a Clientelistic Political Culture? 20 SEBRAE in response to demand-driven programs for small-scale firms.25 The Projeto Sao Jos6 has also put great emphasis on local community involvement and has helped build social capital at the local level, and to link communities with municipal decision-making in areas affected by the project. 39. Finally there have been specific activities to reach socially excluded groups, such as support for poor urban adolescents to stay in school linked to cultural activities. For example the program Crianca fora da rua dentro da escola is one of the flagship social assistance programs of the government. Since its initiation in 1996 the program has helped get 9,910 children and adolescents out of work on the streets, with only a 4 percent rate of recidivism. Its strategy is to work with families in the process of re-integration children into formal schooling. This has been supported by scholarships, professional training and finance for housing and productive work.26 40. The challenge ahead. Institutional and social change is a notoriously slow- moving area, especially when it is embedded in political and social cultures. Ceara has been a leader in tackling these issues, with major successes by Brazilian and international standards since the late 1980s. But despite the innovative programs, there continue to be concerns over capture by local elites, weak capacities and lack of coordination of services at the municipal level. This is in the context of rising demands on municipalities. And problems of excluded groups being forced to adopt "anti-social" behaviors, for example in cultures of drugs and violence persists. The scale of activities targeted at social excluded groups is small, with many activities being led by civil society with sometimes weak links to govemment. OVERVIEW OF SPENDING27 41. Public spending has been an important part of the effort to tackle the various inherited deficits discussed above. This review of the past is concluded by a synthetic assessment of spending performance. 42. Overall, the fiscal consolidation of the late 1980s, combined with modest growth, allowed a significant expansion of real spending-including spending per person. As shown in Table 2, between 1990 and 2000, real spending rose by 65 percent for human capital development (the largest overall category), 56 percent for economic development, 73 percent for social insurance and assistance and 118 percent for public services and safety. 25 These are extensively documented in Tendler (1997). 26 Government of Ceard. (2001). Mensagem a Assembldia Legislativa 2001 27 More details on this analysis is provided in the background material for this report, World Bank 2001e. 21 Tabe 2: Ceara - ReaD Expenditures by Category, 1990, 1995 and 2000 (index 1990=100) Category 1990 1995 2000 Reais (mill.) index Reals (mill.) index Reais (mill.) index Legislative, Judiciary and General Admin. 498.2 100 658.1 132 612 8 123 Economic Development 411.9 100 340.9 83 643.6 156 Human Capital Development 705.5 100 960.4 136 1,162.1 165 Social Insurance and Assistance 349.6 100 396.3 113 603.4 173 Housing, Public Services and Safety 178.1 100 201.1 113 388.2 218 Interest, Amortization & Municipal Transfers 782.4 100 741.5 95 999 9 128 Total 2925.7 100 3298.3 113 4410 151 Total per capita (Reais) 472 100 492 104 612 130 Values in constant pnces (Dec. 2000) Source: Government of Ceara budget information. 43. How appropriate is the current set of poverty reduction programs operating in Ceara, vis-a-vis existing groups of poor and the exit strategies associated with their movement out of poverty? In addition to a general review of allocations, background work for the report draws on information for over 80 programs currently operated by the state of Ceara to begin answering this question, including target group, number of actual program beneficiaries in 2000, and the fiscal year 2000 budget. This is linked to a qualitative assessment of exit strategies from poverty for different poverty groups. Unfortunately it was not possible to get information on federal or municipal govemment spending in a form that would allow a consolidated analysis. The work yielded the following general results: o The government has clearly given high priority to human capital development, and especially education (that accounts for 18 percent of total spending)-in addition to Federal spending in this area. This clearly has helped underpin the major advances in educational and health indicators of well-being. However, some additional spending may be required to rapidly accelerate the human development of the poor. o It is not clear what the overall rural-urban mix of spending is, since there is not readily available information on Federal spending, and many state-wide programs do not have a rural-urban breakdown. However, a large proportion of the state's own spending on economic development is on industrial incentives and oriented toward urban areas-albeit with a growing focus on encouraging industries to locate in urban centers outside the metropolitan area; 28 28 It is important to note, however, that a major component of industrial incentives are tax breaks granted to participating companies, and thus forgone revenues as opposed to actual spending. Other incentives include infrastructure and training subsidies and are "real" spending. 22 * State personnel and administrative costs in Ceara remain high (70% of total expenditures), though much of this is in education and health services; . The state employee pension program is costly to operate (11% of total expenditures, and 28% of total state personnel expenditures) and requires a heavy state subsidy (85% of outlays). Raising the fiscal sustainability of this program would release significant funds for reallocation to activities with a more direct impact on poverty reduction; and * The volatility of the budget cycle has been reduced, but differences between planned budgets and actual spending allocations persist, and there seems to be relatively large underspending on programs of greater importance to the poor. While this may reflect the greater institutional complexity of these programs, there may be a case for added safeguards to protect the budgets and support the implementation of poverty-focused programs. 44. The program data also provide some general indicators of program effectiveness -- what proportion of the target population is actually reached by a given program in a given year? what is the cost-per-beneficiary of the program? -- in order to draw broad conclusions about program coverage among the poor, and efficiency in service delivery. In general, results indicate that Ceara's portfolio of poverty reduction spending includes a few high cost programs from the perspective of poverty reduction - with low coverage and therefore high costs per beneficiary (notably the fiscal incentives program referred to above), combined with a wide range of smaller programs - many of which are priority programs for the poorest such as Projeto Sao Jos6, C6dula da Terra, and a range of programs targeted to the urban poor. However, with the exception of Projeto Sao Jos6, these generally have low coverage rates. 45. More work will be required to assess the effectiveness of spending allocations-see the final subsection on process at the end of the report. III. POLICY OPTIONS FOR INCLUSIVE MODERNIZATION 46. The Govemment is committed to both modemization and inclusion. Despite progress in many areas, inequality has not declined in the past. The global environment is of heightened competition and increased risks especially for groups ill-equipped to reap the potential gains of integration and technical change. The report finds that Ceara's structural conditions imply there is no magic solution, but that a set of complementary policies will be needed to move the society over time on to a path of inclusive modemization. The main findings are as follows: 23 o Growth is necessary but not sufficient for achieving rapid poverty reduction in Ceara: high initial inequality levels reduce the poverty reduction impact of growth, and inequality typically shows high levels of persistence over time; o Pro-poor growth can be facilitated by support for more intensive agricultural production, where water resources allow, by tackling constraints in existing industrial and service activity with competitive potential, and by removing the constraints to micro, small and medium enterprise growth; o Continuing the state's progress in expanding access to education is key to reducing poverty reduction, but may not dramatically change inequagity in the short to medium term. The size of the education deficit is so large that expansion of education will take time to have substantive effects on poverty, and may have only modest impacts on inequality, owing to the long lags in; o Income transfers can have a more immediate impact on poverty in Cearii, and play a key complementary role for redistribution and security. The rural old-age pension program illustrate the potential of targeted income transfers in poverty reduction now; by implementing a child-targeted income transfer the state would rapidly reach and reduce poverty among a much larger vulnerable group; there is also scope for significant reforms in the drought relief program; o Institutional reform is necessary for effective service delivery and inclusion of the poor in all of these areas, especially through strengthening the participatory base and efficiency of municipalities, tackling social and cultural practices that perpetuate exclusion of some groups, and use of e-govemment as an instrument to support social vigilance and service access; and o An explicit spatial dimension of the above policies is important for both effective poverty reduction and a more balanced development process, including support for secondary city growth. 47. Just as the lack of progress in inequality in the past was primarily due the inherited cycle of disadvantage (rather than failures in state action), so the challenge for the future is to shift on to a virtuous cycle of complementary redistributive action that will tackle the reinforcing disadvantages at multiple points. This is illustrated in Figure 11. Much of the basis for this has already been laid in past policy. Box 5 and Table 3 provide an overall summary of the policy recommendations for the future, that are discussed in the remainder of this section. 24 Figure 11. Getting on to a virtuous cycle of redistributive action Better education and skills Inclusive Equitable governance growth in jobs governance and economic action opportunities Transfers for security and redistribution Box 5. Key elements of an accelerated strategy of reducing income poverty and inequality The report concludes that policy has been good in the past, but there is scope in the future for deepening redistributive elements of the development strategy, as part of the overall modernization effort. Both Ceara's past and international experience shows there is no magic solution to deep income poverty and inequality. However, actions on a number of complementary areas will substantially increase the probability of success to tackle the vicious cycle of disadvantage. In view of the uncertainty over development change it is of great importance that this be linked to an ongoing process of assessing experiments and re-shaping actions as results come in. Table 3 provides a condensed summary of this that divides options between those that might be implemented in the short term (the next 12 months) and those likely to be more relevant for the next government-but for which preparatory activity could be taken now. This takes as given the importance of policies of fiscal prudence and good governance as a necessary basis for aggregate growth. More details are provided in the policy matrices at the end of the report. Two messages on the policy mix are worth emphasizing. First, complementary action on various fronts will be necessary for success. Second, the timing of results varies: education will only bring major results in the long term; more labor-intensive growth can bring effects in the medium term; transfers can bring quicker results. 25 Table 3. Summary of key policy colncllusions Timing of potential actions Timing of Short term Medium term effects Equitable 1. Maintain core broad-based rural 1. Comprehensive measures to Medium to growth programs (including Projeto Sao improve the business tong term Jose) complemented by exploration environment of micro and small of niches of high-productivity firms (including titling, potential. streamlined licensing, access to 2. Complete re-design of industrial, technology, micro finance) agro-industry and service policy to 2. Roll out of new focus on support for existing areas industrial/service policy. of activity and potential, including in 3. Active monitoring, evaluation, secondary cities. participatory assessment and 3. Short-run actions for micro and continued redesign of both rural small firms: e.g. via measures to and urban strategies increase access of firms to public procurement Education 1. Implement current plans to raise 1. Increase funding for ECD and Long term quality especially in poor areas pre-school services, linked to 2. Implement systemr to monitor community health agents FUNDEF use 2. Expand to all municipalities 3. Evaluate adult education special support to schools contingent on better results. 3. Diversify adult education curriculum-expand adult literacy if evaluation warrants Transfers 1. Adopt a two-pronged approach to 1. Rollout, and ongoing Short to drought relief, based on a mixture of evaluation of transfer programs medium term crop insurance and a reformed crucial. approach to public works 2. Design (and implement) a state- level supplement to the Bolsa Escola. Institutions 1. Assess new demands on 1. Develop mechanisms for Medium to municipalities from Projeto Alvorada strengthened municipal long term 2. Look for immediate opportunities governance, building on existing for linking civil society activity on participatory mechanisms, e- excluded groups to govemment government, backed by civil action society based tracking and 3. Expand e-government. monitoring. 2. Deepen action on reaching the excluded and working with civil society groups. Spatial 1. Develop an explicit spatial 1. Implement, track, continuous Medium to policy dimension to policy and spending, evaluation and redesign as more long term with a combination of progressive information on spatial dynamics geographic poverty-targeting for becomes available. human capital and transfers, and an increased focus of directly growth- linked policy on areas of potential, including secondary cities. 26 THE CONTEXT: INCLUSION AND MODERNIZATION IN A GLOBALIZING WORLD 48. Both modemization and inclusion have been major themes of the govemment of Ceara in the past decade. While the past experience and above results underline the absence of any magic solution, the search for a policy mix that fosters inclusive modernization is a central issue. 49. This quest is occurring within a global context-of heightened international integration of markets, the information technology revolution and apparently rising pressures for competitiveness and flexibility.29 Mobile factors of production, especially capital and skills, will increasingly move to areas where institutions and productive opportunities are greatest. Effective positioning to make use of areas of competitive potential and the ability to attract or keep capital and skills is central to development strategy. 50. Does this context of heightened competitive pressure run counter to policies for inclusion and social cohesion? Protection for unproductive activities-for capital and workers-is an increasingly ineffective strategy (as well as an inequitable one)-but this had largely run its course by the 1980s as a central strategy in Latin America. However, there is a different path to inclusion that is complementary to competitiveness and modernization, that emphasizes broad- based skills, efficient use of transfers and risk management mechanisms, and inclusive, transparent and efficient governance structures. We expand on this here. 51. Growth is a key element of any poverty reduction strategy, given its direct effect on incomes, as well as its indirect effect on the At current capacity of the govemment to finance public services which are critical to enhancing the well-being of the poor, through rising public growt rates, revenues. Intemational examples of rapid income poverty reduction poverty in have all involved high and sustained rates of growth. However, Ceard will rapid poverty reduction has also been greatly facilitated by either continue to fall modest levels of inequality (e.g. Indonesia and Taiwan, China) or very slowly if reduced inequality (Korea and Malaysia in the 1970s and 1980s). there is no Ceara's extremely high initial levels of inequality significantly reduce reduction in the potency of growth in reducing poverty. inequality.. in 52. In Ceara, simulations indicate that at current growth rates -- 4% 2020, over one- growth in GDP per year -- poverty in Ceara will continue to fall very third of the slowly (see Table 4). If inequality is assumed to remain unchanged population will (i.e. every person's income increases by the same proportion, not remain poor the same absolute amount), the poverty rate would fall by 3 percentage points over 5 years (from 51% to 48% on the basis of a poverty line of half a minimum wage in 1999). And in 2020, over one-third (35%) of the population would still remain below this poverty line. 29 See Castells (1997) for a general discussion of the sociological dimensions of globalization. 27 Table 4: Poverty IReduction Dmpact o Econornc Growth: ProJected Poverty Rates for Ceari in 2004 and 2020 Income Growth per BEFORE AFTER 5 AFTER 21 capita per year: (1999) years (2004) years (2020) 2% (4% GDP 51.2 47.6 34.7 growth) 4% (6% GDP 51.2 44.1 23.0 growth) Notes: (1) These simulations assume distributionally-neutral growth - from the current (1999) distribution of per capita income, each person's income is projected to rise at the same percentage rate. (2) The 1999 poverty rate is based on estimates by the University of Pernambuco using PNAD data calculated for these simulations. It differs slightly from the IPLANCE estimate for the poverty rate in that year. Source: background work by Jos6 Ricardo Bezerra Nogueira and Rozane Bezerra de Siqueira (Federal University of Pernambuco) 53. Inequality is so severe in Ceara that even if the state managed D 0 to double the rate of per capita income growth to 4% per year (6% $3 N$ I f$ growth in GDP), roughly one-quarter of the population (23%) Rod1y flon Coed would still be poor in 2020, after 21 years of sustaining this growth even gl pebF rate (Table 4). The size of the poverty gap (the distance between capka M Tcome the average poor household's income from the poverty line) is grcMh doUb§ed large, such that decades of sustained income growth above today's levels will fail to move a large proportion of the poor out of b 4% L@§ Ygg' poverty. Countries that achieved rapid poverty reduction despite rMU'i417II9Y V4 of high initial inequality (Chile in the 1990s), or rising inequality le popuDidon (China and Thailand in the 1980s), did so at much higher growth wo§dg1 oflff bl rates. p>< § 2020 POLICIES FOR AN EQUITABLE GROWTH PATH 54. Achieving rapid and more equitable growth is one of the biggest challenges facing Ceara, and is especially influenced by Brazil-wide conditions-because the country's macroeconomic financial and demand conditions are key to the economy-wide context for Ceara, and because the labor market is significantly integrated with the national labor market. 55. If Brazilian growth accelerates this would almost certainly be beneficial overall, but may have some mixed effects for the course of development in Ceara. It would greatly help overall demand conditions, but it may increase the attractiveness of capital locating (or staying) in the more prosperous South East, 28 that was growing more slowly than Ceara in the 1990s. This increases the importance of increasing Ceara's competitive advantages. 56. With these qualifications, the following policy areas were identified as within state control and supporting faster and more equitable growth: 57. Overall governance, service quality and the investment environment for employment growth. International and within-Brazil evidence supports the view that better governance and better infrastructure services helps growth, inter alia through raising investment levels. As noted above, Ceara has a good reputation relative to the rest of the North East. Maintaining this reputation and continued deepening of service quality will have an overarching influence on state-level growth. A particular dimension of this concerns the labor market: the costs of creating a formal sector job are a deterrent to such job creation, and tend to encourage informalization at best, or reduced overall labor demand. However, labor policies are largely a matter of federal government policy. 58. Ongoing implementation of reforms of fiscal incentives. In the 1990s Ceara spent substantial fiscal resources in attracting industry to locate to the state, especially from the South East of Brazil. While this policy appears to have contributed to Ceara's success in attracting industries, the cost per job created was on average high and incentives to locate in the interior fostered wide geographic dispersal rather than concentration around areas of existing potential. The state is cutting back on fiscal subsidies-in line with the requirements on fiscal prudence in the Ley de Responsibilidade Fiscal and focusing remaining subsidies on market failures in areas of potential (see (c)). 59. Shaping infrastructure and other public investment around an overall spatial strategy. There is considerable interest in shifting the pattern of public investment to support existing industrial and service activities with potential for expansion, notably in secondary cities.30 It is recommended that policies to influence the pattern of growth start by identifying and supporting areas of competitive potential, especially around existing or new private sector activities, whether in agriculture, industry or services. This private-sector supporting strategy, with an explicit spatial component, represents a significant implementation challenge, and for this reason could best launched as a (major) experiment, with ongoing evaluation. To oversimplify a complex reality, two aspects are highlighted: 30 See in particular the parallel analysis on this issue, Bar-El et al. (2001). 29 60. A two-part rural strategy. The Projeto Sao Jos6 is a good instrument of broad-based support for relatively large numbers of farmers and rural communities-including low productivity areas. While it generally does not transform productive potential, benefits are significant for relatively low costs, and these are enhanced by gains in social capital. A complement to this is continued active pursuit of options for larger productive transformation of agriculture, where the underlying potential exists. A major element of this will be the more efficient matching of people, water and soils. Current rural location pattems reflect a historical pattern of production, water management and institutional development that is almost certainly underutilizing the potential. This can be significantly increased in relation to the state's new initiatives to harness and manage Ceara's water resources, under the PROGERIRH program, that includes both capital investments and use of incentives to encourage better management of water resources. Part of the program will allow irrigation of previously underutilized good soils, allowing their potential use for more intensive production forms, such as fruit production. This can directly help land-starved farmers and workers through linking with migration and agrarian reform (for example using the C6duia da Terra project) and in some cases to labor-intensive commercial farming, for example in cashew and other tropical fruits. The key lies in the process-of a continuously managed spatial diagnostic, followed by demand-driven technical or infrastructure support for farmers. Such a strategy could being substantial benefits, but probably only to minority of the states poor, farming population, because of the overall ecological constraints. Over the longer term, rural-urban migration is likely to remain the dominant mechanism for transforming living conditions for the rural poor. 61. Benchmarking, supply chains and clusters. The proposed approach to urban industrial and service production that is being developed by the state involves careful assessment of existing activities and potential-using a mixture of technical assessment and participatory engagement. The technical element- that would almost certainly involve state-level institutions, and with preferably both private and public involvement, should include benchmarking of potential against competitive activities elsewhere and focus on the supply and marketing chain of production relationships. Often this approach will lead to emergence or expansion of clusters of similar or connected activities-that is a common feature of industrial location patterns elsewhere. Public action could involve infrastructure support, demand-driven technical support or, where justified by market failures, start-up subsidies. Ongoing analysis of determinants of productivity increases in the Northeast suggests training and upgrading can play a major role in this. However, there have been cases of both success and failure in the past. Successes have typically involved linking training to existing clusters of small and medium-size firms with tailoring of needs to the demands of the firms, as in the SENAI center that has been working with the Fortaleza garment industry. Large firms can play a role in this, but experience indicates that it is of great importance that training and technical assistance-and indeed the terms of any subsidies for such firms-be designed to foster links with the existing small 30 and medium scale activities. Large input suppliers to local industries can play a valuable role, since they will have incentives to upgrade the whole local enterprise market for their goods, as evidenced in the role of YKK, a Japanese multinational, whose major Fortaleza plant has played a significant upgrading role in the local jeans industry.31 62. Cross-cutting support for micro, small and medium enterprises A proactive policy toward industries and services will often involve micro, small and medium sized firms-and links between them and larger firms. There is also important scope for improving the overall investment environment for small firms. An assessment of conditions is now underway in the Northeast-past work indicates a poor investment environment is a deterrent to job growth. Elements of a concerted strategy are as follows: 63. The regulatory environment. There is considerable potential for reducing the cost of business through reduced red tape-Brazil-wide evidence finds there are major costs to legal production. 64. Titles. Titling can play a potentially significant role in formalizing production and giving access to credit and public services. This is the major theme of the renowned work by De Soto, and is consistent with the very high levels of insecurity of title in favelas in urban Ceara.32 65. Public procurement. Public spending is a potentially important source of demand for small firms-Ceara had a successful program in the early 1990s that combined an overall reservation of spending for small and medium sized firms, with innovative ways of assuring quality.3' While the specific design of the preferences were ruled illegal in the mid-1 990s, there remains scope for reshaping spending at the state or sectoral level to increase access to the small- scale sector, through the design of tenders, for example, or through types of public spending (e.g. slum upgrading that encourages local firms to participate). 66. Demand-driven technical support. While supply-driven technical support has a poor reputation, program of public spending referred to in the previous paragraph was effectively linked to demand-driven service provision by SEBRAE just as examples of training or technical assistance involving SENAI and large firms (see above) show the potential for effective support. 31 See Tendler (2000 and 2001). These also provide examples of weak linkages, including where state support was channeled to new large firms, but linkages with existing small and medium sized firms were not effectively achieved, for example in the footwear industry in Cear. 32 See De Soto (2000) 33 See Tendler (1997). 31 67. Financial access for small and medium enterprises. The small and medium sector (as opposed to micro-enterprises) is now underserved by formal financial institutions relative to the potential: this is in part because of Brazil-wide weaknesses in the financial sector, and the high cost of credit, but action at the regional or state level can help by encouraging private banks to open small-firm credit departments, better registries and other facilitating measures. 68. Micro-credit. The Banco do Nordeste's Crediamigo program is a well- designed and fast-growing program for micro-enterprises-this, along with other broad micro-credit programs are an important complement to work on titling and improving the business environment, and their continued expansion will be a valuable complement to other measures affecting the environment of micro- enterprises. 69. Information technology. New information technology has large potential benefits for helping the small and medium firm sector get access to markets- through access to information, reducing the need for middle-men, allowing firm- to-firm dealings, and smaller individual production runs. So far there are anecdotes of success in this sector (e.g. in the Pernambuco technology cluster)-as well as in some micro-enterprises (generally involving NGOs). This is an area for experimentation: using the Federal supported roll out of IT infrastructure and encouraging small-large firm linkups and civil society engagement to train and make use of the new technologies. 70. The above approach to micro, small and medium enterprise development is based on general analyses of Brazil conditions, comparable experiences elsewhere. Once the survey of conditions for these segments of the enterprise sector in the Northeast is available it could be used for both a focused discussion-including with enterprise groups-and design of more specific measures. TACKLING CEARA'S EDUCATION DEFICIT - A MEDIUM TERM PROPOSITION 71. Education, education and education are often given as the key(s) to reducing inequality, while also underpinning growth. According to one estimate, 40 percent of overall inequality in Brazil can be ascribed to lack of education.34 With the rising premium on skills due to technological change, this seems only more important in the 21st Century, though there is some new evidence that industrial 34 Paes de Barros et al (2000) 32 firms prefer to have relatively unskilled labor that they upgrade with in-firm training (often state-subsidized.)35 72. There is no question that educational improvement for all is central both to modernization and inclusion. No country has succeeded in transforming living conditions without such an effort. As already emphasized, the inheritance of very high illiteracy and weak schooling is a huge problem. The concerted past action by the state and federal govemment has been central to the effort of reaching the poor. But more will be required in the future. A major, inclusive expansion of education has to be one of the central elements of government strategy. And this is desirable not only for education's role in expanding material opportunities for the poor (and non-poor), but because it has great intrinsic value in personal and cultural development, with important potential benefits for society at large. However, projections developed for this report suggest that education, while of fundamental importance, is no panacea. Effects on poverty take time, and effects on inequality may be quite modest over the short to medium term. This is for two reasons. Improvements in the coverage 73. First, the benefits of education take time to be reaped, and qualty of especially for gains in incomes. Even with rapid educational education expansion the rise in the average years of schooling in the services are workforce will be slow, as better educated cohorts enter the necessary for workforce and others retire. Most of those now working-and poverty with typically low levels of education-will still be working over reduction, but the next 30 years. Levels of illiteracy were 46 percent for rural higher 15-24 year-olds in 1999. For this reason, projections of the educational impact of a education expansion suggest a rise in average years eduain al of schooling from less than 5 years in 1999 to 7 years could take attainment alone up to 20 years. This could be reduced to as little as 10 years will probably under optimistic assumptions on the implementation of education have only a and adult literacy programs. Figure 12 illustrates the possible modest effect on effects of an aggressive program of educational expansion, inequality combined with success in literacy programs on various age groups for 2010 and 2020. As can be seen, both the potential major gains in educational attainment of the young, and the more modest possible advances in literacy only affect part of the workforce. 35 See Tendler (2002). 33 Figure 12. Projected years of schooling for different age groups with an aggressive program of education and literacy, 2000=2020 ProjeVio da escolaridade - Ceari Por faixa etaria - em anos de estudo 14.0 0 10.0i --- -- ----_- -- - _ _ _ _ -- -- Un 8 .0 --- -- - - - - - - - - -- - - - - - -- - - - - - - - - - -- - - - - - - - 2000 --- 2010 6.0 --------- -- 2020 2 0 C 15a 20a 25a 30a 35a 40a 45a 50a 55a 60a 19 24 29 34 39 44 49 54 59 64 Faixa etana ( em anos) Note: this assumes sharp rises in years of schooling for new entrants to the educational system (and subsequently the workforce) shown in the graph, combined with a roughly 50 percent reduction in illiteracy for 15-30 year- olds by 2010. Source: projections developed in IPLANCE on the basis of micro data of PNAD 1999. 74. Second, while large scale and broad-based expansion in education will bring significant income gains for the poor in the long term, the analysis suggest that this alone is unlikely to have a major impact on inequality. The projections explored the consequences of a large increase in education for incomes, participation in work and demographics, that is based on the results of an econometric analysis of the impact of education on family size, work choices from the past. This assumed a broad-based increase in educational attainment to seven years.36 More education tends to increase what individuals can eam, increase participation for the low-skilled and reduce the number of children that households choose to have. 75. The effects on individual incomes depends on the interaction between supply and demand for different levels of skills. It is very hard to predict what will happen to the pattern of demand, but a plausible range of results was obtained by exploring the effects of alternative assumptions for the structure of returns to education, from a significant improvement to a further worsening. As noted in the historical section, retums to education in Ceara are significantly influenced by 36 For more details see the background papers for this report, World Bank (2001e), and Ferreira and Leite (2001) that applies the techniques reported for the urban Brazil in Ferreira and Paes de Barros (1999) 34 Brazil-wide developments because of the partial integration of the state in the national labor market. The principal results of this analysis are as follows: 76. in terms of the distribution of individual earnings educational expansion has very little impact on overall inequality-under all the scenarios there is only a modest decline in the Gini coefficient of earnings from 0.54 in 1999 to a minimum of 0.51 (in the scenario of a reduction in the dispersion in the returns to skills). One of the major reasons for this minor effect is a large increase of participation of low-skilled workers, especially of women, associated with the education expansion. However, while alternative assumptions on the structure of retums to skills make a small difference to the overall index of inequality, the effect on earnings is large, especially for those with low skills. Educational expansion could lead to large gains to the very poor if returns to the low-skilled rise, but would leave them hardly better off if there continues to be a widening of differences between the high and low educated. This underlines the importance of strategies that provide work for the low-skilled, even with a major educational effort. 77. In terms of household incomes, the results are different: much more positive for poverty, but still not dramatic on inequality. Even though the substantial number of women entering the workforce due to educational expansion receive low earnings, they are contributing to household incomes amongst poor households. On top of this, the reduced household size, induced by greater education, leads to greater gains on a per capita basis-since incomes are shared among fewer household members.37 These positive effects on incomes are substantial even if the earnings for the low-skilled stagnate (under the scenario of further widening of returns to skills) as a consequence of the effects on participation and fertility. The medium-term impact on poverty ranges from a 9 percentage point decline (with a widening dispersion in the returns to skills) to a 14 percentage point decline (with a narrowing in earnings differentials across skills). However, the effect on overall inequality is again modest: the Gini coefficient for per capita household income declines from 0.62 in 1999 to 0.56- 0.57 with the medium term expansion of education to 7 years, depending on the particular assumptions. This is a non-trivial decline, but would still leave Ceara highly unequal. 78. As with any scenarios, these should be treated as only illustrative. However, they are consistent with what is known globally and historically for Ceara about the difficulty in reducing inequality. Countries as diverse as Chile, Malaysia, the United Kingdom and the United States have experienced periods of rising 37 Note, however, that the analysis did not make allowances for economies of scale of household consumption that would moderate somewhat this result. 35 eamings inequality in the past 20 years, despite expansion in education. This only underscores the centrality both of affirmative programs to assure education reaches the poorest and of complementary actions-especially on factors affected unskilled labor demand and, especially for the long transition, on transfers. 79. While education expansion may not have a major impact on inequality, it is likely to have large medium term impacts on poverty. Education policy recommendations thus focus on improving the coverage and quality of Ceara's public education services - and are designed in particular to provide greater access to these services among the poor, especially to break the inter- generational transmission of poverty. Recommendations include: (a) raising access and quality of Early Child Development (ECD) and pre-school services, by increasing state funding for these services, training service providers and partnering with NGOs as service providers; (b) improving the quality of basic education services, by expanding state funding contingent on improved municipal teacher recruitment and education standards to all municipalities in the state, and by increasing state provision of basic educational materials; (c) addressing the particularly poor performance of municipal schools in grades 5-8 by monitoring municipal use of FUNDEF (federal) funds, and by assisting municipalities in developing local school management capacity through technical assistance (particularly in financial management tools); (d) evaluating current adult education services - particularly those targeted to young (ages 15-24) entrants to the labor market, and seeking guidance from international experience on raising employment and income effects of current adult training programs, in view of the difficulties in making these programs effective; (e) raising access to secondary and higher education among poor students, by providing scholarships for low-income students (extending Bolsa Escola to secondary students), implementing a public service program placing university students as volunteers in remote secondary schools, and raising the profile of adult learning opportunities through distance-learning- renovate state TV studio (infrastructure and equipment) and diversify curriculum offerings. 80. Given Ceara's deep income poverty, which constrains educational attainment for children in poor households, the above improvement in the supply of education services will be most effective in extending educational opportunities to the poor if complemented by demand-side interventions. Such programs are currently missing from the state poverty reduction strategy, and the report recommends that the state implement an income transfer program to poor households, linked to educational attainment (see below). 36 81. While the report did not focus on population growth, it is worth noting the positive poverty effects that were associated with lower fertility and reduced dependency ratios in poor households. Intemational evidence generally finds that the primary sources of reduced fertility are higher education of mothers (an effect captured in the scenarios), rising incomes and better health. Access to contraceptive services is likely to play a facilitating role. FASTER POVERTY REDUCTION THROUGH INCOME TRANSFERS 82. Income, transfer programs are effective poverty reduction tools, that are complementary to measures affecting capabilities and opportunities. Even though they involve redistribution rather Income transfer than incorme creation, they have a key role to play in the programs do modernization process and have been a feature of almost all substantially cases of success intemationally (see Box 6). One such program - reduce poverty: Brazil's federal rural pensions program, that currently operates in Ceara -- effectively illustrates this outcome, as noted above. in Ceara, rural However, given that this is the main income transfer program poverty has targeted to the poor in Ceara, there are sizable gaps in coverage. been reduced by Most rural pension recipients have fewer than 2 dependents 14 percentage (82%), so the 'spillover' effects of these income benefits to the oi t general population are limited. There are no transfer program of poins as a significance currently operating in urban areas - where the result of the majority of the poor live. And most importantly, families with federal rural young children are not covered by any income transfer program. pension Since children ages 0-18 have higher income poverty rates than program. any other age group, and make up over half of the total poor in Ceara, this is a particularly significant omission in current poverty reduction policies. 83. Transfers linked to human development of poor children. Recent innovations in Brazil (the Bolsa Escola and PETI) and intemationally (Mexico's Progresa/Oportunidades) point to the promise of income transfers to families with children tied to attendance at school, or, in some programs, other social services. These programs are designed to accelerate the pace of poverty reduction by delivering immediate income support to poor families, reducing child labor and improving educational outcomes of today's children. They thus have an immediate impact on current poverty and also raise their chances of moving out of poverty in the medium-term (in other words, break the inter-generational transmission of poverty). Evaluation results are positive, indicating that both household consumption levels and children's educational outcomes--enrollment, repetition and dropout rates improve as a result of participation in the program. 37 Box 6. The role of redistributive transfers in the modernization process Transfers to individual and households are a key component of social policy. They play an important- and rising-role in the modernization process.' They form a major part of social policy in all industrialized countries, both in European societies, where the 'welfare state" is most developed, and in countries such as Japan and United States. While newly industrializing societies of East Asia had a reputation for avoiding transfers, in fact they have played a significant role in countries such as Korea, Malaysia and Thailand'. Even before the East Asian crisis, there was widespread concern about how to make these societies both caring and competitive, especially with the decline of informal, family-based mechanisms of risk management. And transfers are, if anything, of rising importance in a globalized world, in view of heightened concern over risk and losers from changes that flow from increased integration and technical change. A coherent policy on transfers is a complement to a market-oriented strategy. Transfers are used for two reasons: to protect people from adverse shocks; and to transfer income to particular groups in normal times. They are used both for the middle class and for the poor. Typically transfers for the middle class are a form of insurance to deal with risks, even if they are mediated by the state. A large part of old age pensions and unemployment insurance fall into this category, and form the major component of social security systems in OECD countries, Latin America and other industrializing societies. The issues for such middle class transfer or insurance systems is to design them in a way that avoids adverse incentive effects on employment or savings, and to avoid "perverse" redistributions to the middle class from general revenues. As noted in the text, within this category there is an important reform agenda for public pensions in Ceara, as in the rest of Brazil. However, the major focus of this report is on poverty reduction, and the role of redistributive transfers within this. There are powerful reasons for such poverty-focused transfers playing a central role within a comprehensive strategy, as they do in all industrialized countries. These are complementary to actions that promote opportunities for the poor, expand their capabilities and increase their influence on decision- making. o While the poor often have sophisticated mechanisms for managing the risks they face, they remain highly vulnerable to shocks and are least well equipped to manage them. This applies in particular to shocks that affect whole communities or groups, such as droughts or macroeconomic declines. o Transfers are the only way of getting quick gains in poverty reduction in the context of moderate growth rates (see text) and can help lift the very poor out of poverty traps. For highly unequal societies such as Ceara (and the rest of Brazil), the case for redistributive transfers is especially strong. This is both because the effects of distributionally-neutral growth is much weaker on poverty reduction; and because relatively modest costs in terms of average incomes can finance large increases for the poor, precisely because the gap between incomes at the bottom and average incomes is so large. Design is key. Transfer schemes ideally reach all those in need with modest administrative costs, are complementary to expansion of the capabilities of recipients-rather than creating dependency-are not susceptible to capture or abuse by those involve in managing them, and sustain broad political support. International experience suggests the need for a variety of schemes to reach the range of risks, needs and institutional capabilities of different societies. This can involve basic pensions to the elderly poor (as in Brazil's previd6ncia rural), workfare with low wages for self-targeting linked to valuable social investments (such as Trabajar in Argentina or Manos a la obra in Colombia), transfers linked to education and health of poor children (as in the Bolsa Escola or Mexico's Progresa/Oportunidades) and nutritional support for the malnourished. 38 84. A federal proposal at the time of writing involved extension of While a Federal the Bolsa Escola program to Ceara under federal financing. Bolsa Escola of Then proposed benefit levels, at R$15 per child per month and R15 would be of targeted to families below the poverty line, will make a valuable value to the poor, contribution to reducing the income gap for the poor (the extent it would be unlikely to which the poor people are living on incomes below the poverty line), but will have a small impact on the proportion of people to have a living below the poverty line. Further, coverage as currently significant impact planned, is limited to households with children between the ages on poverty of 7-14, excluding many poor households, including those with incidence in infants and children through age 6, from coverage under the C hih program. As such, simulations indicate that under this design, seara, wi ch Is income poverty will fall by less than 2 percentage points as a simulated to fall by result of the program (from 51 % to 50%), and the indigence rate less than 2 will fall by only a slightly greater amount (from 26% to 23%)- percentage points see Table 5. as a result of the program. Table 5: Options for Bolsa Escola in Ceara: Projected Poverty Impact and Costs Program BEFORE (1999) AFTER Costs Description Povertv Poverty (R$ mill) Indigence Indigence (1) Federal Bolsa Escola - $15 per 51 26 50 23 147 child, ages 7- 14 only (2) State raises benefit to R$25 per 49 21 246 child (3) State raises benefit, and expands coverage to 46 16 443 all children ages 0-14 Source: Background work by Jos6 Ricardo Bezerra Nogueira and Rozane Bezerra de Siqueira (Federal University of Pernambuco) 39 85. The impact of the Bolsa Escola program on poverty in Ceara could potentially be increased by supplemental state financing for increased benefits and/or coverage. As Table 5 illustrates, a marginally greater impact on poverty - particularly indigence - can be achieved by raising benefit values, but a significantly greater impact on both poverty and indigence can be achieved by broadening coverage to all families with children. For example, if the state were to maintain coverage rules of the federal program (only households with children ages 7-14 would be eligible), but raise benefit amounts to R$25 per child, the program would reduce indigence rates from 26% to 21% (the additional impact on poverty, however, would be marginal - dropping from 51% to 49%). The cost of this program to the state in benefits paid $ * - out is estimated at R$99 - or 2.3% of total state expenditures38. However, if the state were to raise benefit IIIIg7?pa2 levels by this amount and expand coverage to all children ages 0-14, indigence would fall to 16% and poverty to 46%. PGOf)tfly and The cost of this program is substantially higher at R$443 ilndgoence can million - requiring R$296 million in additional state funding. b, ,bs,Cto%a5<54\S4t 40000 ~_____________________45 30000 ,- 20000 10000 Souce CacuatinsbyILAN, CEl frmmir data fro PND 99 46 The work for this report did not evaluate the technical packages on offer. However, it is worth noting that earlier work on CearS emphasizes the importance of local farmers playing a leading role in deciding on their opportunities and needs, with government technical resources in a demand-driven support role. See Tendler (1997). 45 o Continue with analysis and planning for upcoming years to institute a drought management strategy that links into the rural development and poverty reduction strategies. More information is needed to develop a weather-based index insurance - especially the costs and financing structure.47 99. Given the greater uncertainty over the first, new, scheme, it will be important to continue to have significant resource allocations to the second. If the wage is set low enough (in contrast to past practice), there should not be a need to have additional restrictions on who is eligible. Most importantly, both approaches should be systematically evaluated, using quick qualitative evaluations in 2001, with community involvement and more systematic quantitative analysis in the future. This will allow the development of designs that provide a truly effective, long-term means of protecting the most vulnerable groups in Ceara. 100. SocaG insurance programs: the high cost Of subsiding state pubNic sector pensions. The state currently operates one social insurance program, which provides old-age pensions, disability benefits and health insurance for public sector employees. Total outlays for this program were R$480 million in 2000, or 11% of total state spending, and 28% of total state spending on personnel in that year. As with other Brazilian states, Ceara's pension program is not financially sustainable as currently designed, with benefits paid out vastly outpacing contribution rates48. Costs of the system have been rising over the 1990s, and in 2000 the state financed about 85% of benefits paid out by drawing from general tax revenues49. As designed, the system absorbs a significant proportion of state revenues to finance a program of subsidized income transfers which are by and large not directed to the poor. 101. Reforms to the state pension system, in particular focused on raising its fiscal sustainability, will be an important contribution to improved poverty reduction policy in Ceara - reducing costly subsidies that are not directed to the poor, in order to make room for financing of more pro-poor transfer programs. However, these reforms will have only medium and long-term payoffs (in terms of releasing state funds for other programs). Thus while pension reform is critical, it cannot be realistically viewed as a source of funding for additional programs recommended in the report, at least not in the short-term. 47 Technical assistance for a risk and vulnerability assessment and an action plan: technical approaches are available in the World Bank, and under ongoing efforts by University of Arizona and UFC on vulnerability to drought and Grupo do Israel on rural and regional development. A joint research effort to monitor and evaluate drought management could be very useful. 48 See World Bank (2000c). 49 Benefits paid out through the pension program totaled $456 million in 2000, representing 28% of all state expenditures on personnel in that year (Table 4.4). The portion of these costs subsidized from general tax revenues (85%) represents 9.2% of total state expenditures in 2000. 46 GOVERNANCE AND INCLUSION 102. A process of inclusive modernization requires an effective state, and one that is responsive to the needs and demands of the poor. There was not a specific analysis of institutions for this study, but the issues are of such fundamental importance to all the domains of public action discussed in this report that they are highlighted here. 103. As noted in Section II, past work on Brazil (and Latin America) has documented how institutions of the state have tended to work to exclude the poor and serve the interests of the elite. In Ceara, action by the government in the past decade or so has been instrumental in starting to tackle these problems, using active state interventions to increase the orientation of programs to poor groups, to motivate civil service workers and foster local participation. Examples include the innovative program of agentes de sauide, periods of effective action by the agricultural extension service, community-based rural development work under Projeto Sao Jos6 and active interactions between local government and civil society in some poor urban areas. However, the agenda is far from complete. Municipal govemance in particular remains highly variable in quality, and there continue to be concerns over capture of resources by local elites and weak coordination of services at local level. At the same time, the demands on municipalities are rising, and are likely to rise further if more resources are channeled to local government via the Projeto Alvorada that is a Federal initiative to channel resources in a number of existing and new programs to municipalities. 104. A particular dimension of this issue concerns those groups outside the mainstream of society, especially those urban youth with little education or poor prospects for decent work, who move into cultures of drugs, violence, and prostitution. 105. The report has a number of recommendations in this area, for further assessment and study, including: (a) building on existing participatory mechanisms in agentes de saude, Projeto Sao Jos6, etc. to develop a more comprehensive and coordinated approach, with municipalities over time becoming a key element; (b) developing monitoring mechanisms for tracking the relationship between decentralized decision making, local political and social conditions, resource allocations and outcomes in service delivery-backed by mechanisms to promote transparency and "social vigilance"; (c) experimenting with participatory budgeting and planning at municipal level, linked to local development of plans, and monitoring of results; 47 (d) using E-government as a mechanism for fostering open information, access to services by the poor and reducing the risk of corruption in procurement, complementing institutions that provide the basis for social accountability and redress; (e) working with civil society groups and strengthening existing state programs on key socially excluded groups-notably adolescents at risk-to bring them back into schools and work. SPATIAL IMPUCATIONS OF A RENEWED ATTACK ON POVERTY AND INEQUALITY 106. Each of the above areas of economy-wide or sectoral policy has a spatial dimension. CearA displays two marked features of the spatial distribution of the population and well-being: first, a large minority of the population, and almost half the poor, live in dispersed patterns in the semi-arid and drought-prone serldo- often in conditions of extreme poverty; second, a high concentration of urban population live in one single metropolitan (Fortaleza) region.50 107. The difficult agro-ecological conditions greatly limits the scope for a broad- based rural productivity transformation that was so marked a feature of equitable growth in all cases of East Asian success, except for the city-states of Hong Kong and Singapore. Migration from rural to urban areas will be a major source of poverty reduction in the long term. Urban job opportunities, greater rural education and an improved communications infrastructure will all support this migration process. However, intemational evidence shows that the dynamic transition to a predominantly urban location takes decades, with only a slow decline in the numbers of people living in rural areas-even when these areas are resource-poor. In Ceara only 12 out of 184 municipios experienced negative population growth in the 1996-2000 period. Moreover, it is usually the case that rural areas continue to have the deepest poverty and worst social indicators. 108. The design of spatial dimensions of the policy needs to balance supporting the most dynamic areas of job growth with redistributive efforts on human capital, current incomes and the factors affecting the quality of life. And it needs to do this recognizing that the transition to an predominantly urban society will take decades. This implies relatively strong support for poor rural areas in terms of human capital, transfers and core infrastructure, but greater concentration of productivity-enhancing support in areas of productive potential. Current plans to make secondary cities a special focus of attention makes sense in principle-but this is best driven by assessment of areas of real potential, combined with strong local participation. It is important that any focus on urban development in the 50This subsection draws in part on the parallel work of the Israeli team, see Bar-el 2001. 48 interior of the state be integrated within an overall poverty reduction and development strategy. Moreover, the poverty analysis found that over a fifth of the poor live in metropolitan Fortaleza (especially in periurban areas), and that the share of Fortaleza in total poverty has been on the rise. It is important that this group not be left out of any poverty strategy. 109. The spatial "mapping" of the recommendations are summarized in Table 7. Table 7: Spatial dimensions of policy recommendations Area of strategy Spatial dimension Education Strongly equalizing across space: additional affirmative action needed to provide quality basic education in all rural areas (to include most deprived groups), backed by high quality upper secondary and tertiary education in secondary cities Transfers Equalizing across space: priority to indigent groups will lead to disproportionate rural focus, with also relatively high focus on periurban areas in secondary cities and Fortaleza. Growth Basic support for some core infrastructure should be modestly equalizing; focus on productive opportunities, driven by local conditions likely to lead to concentration of production both in secondary towns and continued growth in Fortaleza. This would be complemented by special efforts to reach poorer locations/favelas in Fortaleza. Institutions Municfpios will be a key layer. Institutional design and effort needs to balance access and responsiveness to all-especially dealing with inefficiencies and clientelist practices in rural areas-while paying attention to the efficient scope, size and reach of institutions (as already evidenced in some excessively small municfpios); the extent of decentralization will generally be service- specific. In addition, effective spatial policy will also require state- wide diagnostic and design work, and demand-driven actions of state-level institutions that are already located throughout the state, with scope for synergies where there already exist clusters I of decentralized institutions (such as SEBRAE) 110. Implementing a more fully articulated spatial strategy is best undertaken by an assessment of areas of current dynamism and activity, molding institutional, service and infrastructural responses in a demand-driven fashion. This would have to be underpinned by more effort on the overall data and diagnostic base for the spatial distribution of public effort-work that is already underway by the government. While the geographic information base is very rich in some areas, there is only limited information on the spatial or rural-urban distribution of sectoral spending. 49 FINANCING AND THE LEGACY OF GOOD FISCAL GOVERNANCE 111. Many of the proposed options for deepening the attack on poverty and inequality would not involve increased spending. For example, improving the business environment for small and medium-scale enterprises is about policy, not spending, and shifts in the spatial focus of program would largely involve reallocations. The major areas implying increases in spending could concern any supplement to a Bolsa Escola, and to a lesser extent some of the proposals in education. 112. Ceara has deservedly earned a reputation as a model of D §g o good policy-making at the state level - particularly in terms of fiscal management. Growth has been accompanied by public good f$sca§ expenditure containment, which has fluctuated fairly predictably gOvernanCe D$ between 15% and 18% of GDP. At the same time, expenditures MV @dasiy igie have been reallocated into poverty reduction efforts - notably are no obWous investments in human capital development (education and health), sgaay §$ while reducing the share of total expenditures on debt service, the O 9g % of legislature and administration. Wn7 empendDigure go 113. The legacy of good fiscal governance is that today there are gDf Assesses existing job growth in Fortaleza limited job creation. activities in industry and Fortaleza and Incentives for location in services secondary cities interior have been too > Supports concentration dispersed and in secondary cities discouraged possibly where this makes productive clusters. economic sense Relatively little attention to > Uses benchmarking existing production against comparable activities until recently. activities in other Tourism has been growing states/countries to fast; is employment- assure potential intensive. competitiveness > Identify infrastructure and market bottlenecks > Uses public-private existing institutions or new partnerships for diagnosis support > Phases down of industrial subsidies; where used link to local activities with high job content Medium term Active monitoring, evaluation, participatory assessment and continued redesign of strategy 55 Table 8 (continued) Informal and self- Low productivity High regulatory burden, Short term employed workers of micro, small including six weeks for and medium formalization and Reintroduce mechanisms for Exit strategies: enterprises registration. public purchases from small Productivity growth Limited markets and and medium enterprises of informal firms market information. within the legal ruling. (formalization of Expensive short-term Roll out of Crediamigo and some) working capital; limited other micro-credit long-term capital; in part Develop program for radical linked to weak titling. improvement of business SEBRAE has broad environment for the small coverage of services, but and medium sector. unclear how big an impact Emerging technology Miedium term sector, but fixed line > Titling phone density low. > Streamlining of licensing Telephone uncertain )> Demand-driven training markets and service provision- e.g. with vouchers > Encourage small/medium lending departments in banks. > Possible sectoral policies e.g. in tropical fruit processing and export > IndustriaVservice policies to encourage links with large firms and use of IT 56 Table 9: Education - Recommendations Population Problem Diagnosis Policy Recommendations Group/Exit Strategy Poor children ages 1. Low enrollment Low quality services: 39% Medium term: 0-6 rates and poor of pre-school teachers 1. Broaden programs to train quality of Early have not completed pre-school teachers Exit Strategy: Child secondary school. 2. Train community health Attend ECD and Development Municipalities are agents as ECD service pre-school (ECD) and responsible for providing providers preschool these services; service 3. Use NGOs as service programs coverage and quality is providers worst in the poorest 4. Increase state funding for (11% of children municipalities. municipal ECD and pre- ages 0-3 attend school services ECD programs; 72% of children ages 4-6 attend preschool) Poor children and 2. Low quality of Low level of teacher Medium term youth ages 6-18 basic education qualification and skills. 5. Expand special support services Lack of operational for poor municipalities to all Exit Strategy: standards that apply municipalities (funding human capital (46% of students uniformly across schools. contingent on improved acquisition (attend complete grade 8 Lack of educational recruitment and educational primary and on time, state test materials in schools standards) secondary school) scores below 6. Increase state provision of national average) education materials Poor rural school- 3. Pocket of Municipal teacher Short term age children severe low quality recruitment is not merit- 7. Implement system to in municipal based. monitor use of FUNDEF Exit strategy: schools, (esp. Municipal schools lack (federal funds for municipal same as above grades 5-8); rural access to teacher training education service provision). children fail to and monitoring systems advance beyond that currently operate in Medium term: lower secondary state schools 8. Provide technical Poor management assistance to municipalities (indicator?) capacity, limited access to on educational financial financial management management. .__________________ system s Poor new entrants 4. Large Low quality basic Short term to labor force population of education services. 9. Evaluate adult education (ages 15-24) under-educated Adult education and programs for impact on young adults training have limited employment and incomes of Exit strategy: entering labor coverage and beneficiaries human capital force effectiveness is unknown. acquisition (training Medium term: programs), (60% of urban 10. Diversify adult education migration population ages curriculum; offer "certificacao 15-24 has less de competencias" than 8 years of schooling) 57 Table 9 (Continued) Secondary and 5. Low Distribution of secondary iWedium term: higher education- enrollments and schools urban-focused; 11. Provide scholarships for aged students poor quality in disparities in access and low-income students in secondary quality between income secondary school (Bolsa Exit Strategy: education groups. Escola) human capital Curricula unmatched to 12. Implement program for acquisition -obtain (30% of population labor market demands. college students as teachers higher education ages 15-19 17% of secondary school in remote areas. enrolled in teachers failed to meet 13. Renovate state TV secondary school) state standards. channel (TVC) facilities; widen TV curriculum I offerings 58 Table 10: Income Transfers - Recommendations Population Problem/lndicator Diagnosis Policy Recommendations Group/Exit Strategy__ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ Poor families with 1. Income poverty Current rates of growth Short term children ages 0-14 rates fall slowly projected to reduce 1. Supplement Federal despite economic poverty rate by less than 3 Bolsa Escola program Exit Strategy: growth; poverty percentage points in 5 (increase benefit amount Income transfers for concentrated years). from R$15 to R$25 per poor families, + among the young Existing income transfer month, with a maximum human capital programs exclude poor benefit of R$75, and expand investment (school (Poverty rates families with children. coverage to all children ages attendance) for stagnant in Ceara' Children from poor 0-14) children 1995-99; 44% of families have low Impact: poverty rate the poor are under educational attainment, reduced by 5 percentage age 15; two-thirds high grade repetition and points, indigence rate by 10 of young people dropout rates, and enter percentage points. live in poor labor force early at below- Cost: R$296 million households.) poverty wages. (benefits only), 6.9% of state expenditures. 2. Proposed Medium term: Federal Bolsa The program is projected 2. Expand coverage of Bolsa Escola Program to reduce poverty by less Escola to secondary schools will have an than one percentage (i.e. Mexico's Progresa important but point, and indigence by model) modest impact on two percentage points Introduce smart cards and income poverty (benefit levels of R$15 per other ICT options to increase child are low, and efficiency and transparency coverage restricted to of benefit delivery systems. I children age 7-14.) 59 TabWe 10 (Continued) Families of farm 3. Groups with Concerns that existing Short term: and non-farm highest level of programs do not reach all Introduce a menu of workers affected structural poverty in need and insufficiently programs for areas affected by drought. at great risk in linked to longer term by drought (according to drought years, development. objective rainfall data) Exit strategies: including 2001. including: transfers linked to In designing new 1. Crop insurance, with insecurity and risk programs important to potentially two windows: of poverty as a experiment with (a) basic universal coverage complement to alternatives that reach for all small farmers medium to long different groups: small (b) additional coverage, with term strategies of farmers and landless farm a subsidized premium for migration and and non-farm workers farmers participating irrigation affected by the drought. farmers, linked to new technical packages 2. Reformed workfare with wages set at wage for low- paid rural casual workers in a normal year, and linked to ongoing programs and community projects (e.g. adding wage payments to Projeto Sao Jos6) 3. Quantitative and participatory mechanisms for tracking and assessing effectiveness of elements of the menu. Medium term Ongoing evaluation and redesign. 60 REFERENCES Abers, Rebecca Neaera. 2000. Inventing Local Democracy - Grassroots Politics in Brazil. Lynne Rienner Publishers, Boulder Colorado Arretche, Marta. 2000. Estado Federativo e Polfticas Sociais: Determinantes da Descentraliza,cao. Editora Revan. Rio de Janeiro. Baker, Judy L. 2000: Evaluating the Impact of Development Projects on Poverty. The World Bank. Washington, D.C. Banco do Nordeste. 2000. Projeto Novo Modelo do Irriga,ao. Estado do Ceara. Fortaleza. Bar-el, Raphael, Arie Shachar, Dafna Schwartz and David Bentolila. 2001. Promoting Economic Development in the Interior of Ceara. Jerusalem. Report prepared for the Government of Ceara. Blackburn, James. 2001. Popular Participation in Bolivia, Rights versus a Clientelistic Political Culture? Mimeo, Lima Camargo, Jose Marcio, and Francisco H.G. Ferreira. 2001. 0 Beneficio Social unico: Uma Proposta de Reforma da Polftica Social no Brasil. Departamento de Economia; Pontificia Universidade Cat6lica do Rio de Janeiro. Rio de Janeiro. Case, Anne and Angus Deaton. 1998. "Large Cash Transfers to the Elderly in South Africa". Economic Joumal 108 (Sept): 1330-1361. Castells, Manuel. 1997. The Power of Identify. Blackwell Publishers. Chu, Ke-Young, Davoodi, H. and Sajeev Gupta. 2000. "Income Distribution and Tax and Government Social Spending Policies in Developing Countries". IMF Working Paper 0062. Dar, Amit and Zafiris Tzannatos. 1999. Active Labor Market Programs: A Review of the Evidence from Evaluations. SP Discussion Paper No. 9901. The World Bank. De Ferranti, Guillermo Perry, Indermit Gill and Luis Serven. 2000. Securing Our Future in a Global Economy. The World Bank. Washington, DC. De Soto, Hernando. 1989. The Other Path. Harper Collins. 61 De Soto, Hemando. 2000. The Mystery of Capital. Why Capitalism Triumphs in the West and Fails Everywhere Else. Basic Books. Di6genes, G16ria. 1998. Cartografias da Cultura e da Violencia - Gangues, Galeras e o Movimento Hip Hop. Annablume editora, Sao Paulo. Dreze, Jean and Amartya Sen, 1989, Hunger and Public Action. Clarendon Press, Oxford. Evans, Alison. 2000. " Working Paper to Address Spatial Considerations in National Poverty Reduction Processes". World Bank/UK DFID Collaborative Program for Rural Development. December. Ferreira Alves, Francisco. 2001. Possfveis explicag6es para o crescimento econ6mico diferenciado do Ceard, Frente ao Brasil, IPLANCE, discussion paper Ferreira,.F.H.G. and R. Paes de Barros. 1999. "The Slippery Slope: Explaining the Increase in Extreme Poverty in Urban Brazil; 1976-1996". Revista de Econometria, 19 (2), pp.211-296. Ferreira, Francisco H.G. and Phillippe George Leite. 2001 "Effects of Expanding Education on the Distribution of Income in Ceara: A Micro-simulation." WIDER Discussion Paper #2001/88. Government of Bolivia, Vice Presidency for Popular ParticipationNVorld Bank 1998. Estudio de Productividad Rural y Manejo de Recursos Naturales. La Paz. Govemment of Ceara. 1995. Todos pela Educa,ao de Qualidade para Todos. Secretaria do Educacao (SEEC). Fortaleza. Government of Ceara. 1999. Informa,ces Educacionais Estado do Ceara. Reuniao dos Secretarios Estaduais de Educa,cao. I NEP. Government of Ceara. 2000a. Consolidando o Novo Ceara: Plano de Desenvolvimento Sustentavel, 1999-2002. Secretaria do Planejamento e Coordenacao (SEPLAN). Fortaleza. Government of Ceara. 2000b. A Escola do Novo Milenio. SEEC. Government of Ceara. 2001 Mensagem a Assembl6ia Legislativa Grosh, Margaret. 1994. Administering Targeted Social Programs in Latin America: From Platitudes to Practice. World Bank. Washington DC. Juhn, C. et al. 1993. "Wage Inequality and the Rise in Returns to Skill". Journal of Political Economy. January. pp. 410-442 62 Kenny, Charles et al. 2001. Information and Communication Technologies and Poverty. Draft - February 2001. World Bank. Washington, DC. Maia, Maria, et al. 1999. Qualidade e Equidade no Ceara: Dimensionando Disparidades Educacionais. Banco Mundial/Governo do Estado do Ceara. Projeto de Qualificacao Basica (1993-2000). Setembro. Fortaleza. Menezes, Naercio and Mark Roland Thomas, 2001. "The Distributive Quality of Growth in Brazil". Mimeo. World Bank. Washington, DC. Morduch, Jonathan and Manohar Sharma. 2000. "Safety Nets, Non-Formal Insurance, and Microfinance". Safety-Nets Primer, Module 11. World Bank. Washington, DC. Paes de Barros, Ricardo, & Rosane Mendonga. 1997. Costs and Benefits of Preschool Education in Brazil. Rio de Janeiro: IPEA. Narayan, Deepa, Robert Chambers, Meera Shah and Patti Petesch. 2000. Voices of the Poor: Crying out for Change. World Bank. Washington, DC Reimers, Femando, 2000. Unequal Schools, Unequal Chances. Harvard University. Cambridge. Romeo de Vasconcelos, Jose, et. al. 1999. Ceara: Economia, Financas Publicas e InvestimentosnoAnosde 1986 a 1996. February. Mimeo. Brasilia. Sen, Amartya. 1982. Poverty and famines: An essay on entitlement and deprivation. Oxford: Oxford University Press, 1981. Sen, Amartya. 1999. Development as Freedom. Alfred Knopf. New York. Shonberger, Steven. 2000. Microfinance Prospects in Brazil. World Bank. Washington, DC. Mimeo. Sisnando, Pedro Leite. 2000. Estrategia Para 0 Desenvolvimento Economico Com Menos Pobreza Rural Do Ceara. Novembro. Secretaria de Desenvolvimento Rural, Estado do Ceara. Fortaleza. Tabor, Steven. 2000. "Direct Cash Transfers". Social Safety Nets Primer. World Bank. December. Tanzi, Vito. 1997. "The Changing Role of the State in the Economy: A Historical Perspective". IMF Working Paper WP/97/114. 63 Tendler, Judith. 1997. Good Government in the Tropics. Johns Hopkins University Press. Baltimore. Tendler, Judith. 1999. The Rise of Social Funds: What Are They a Model Of? Department of Urban Studies and Planning, Massachusetts Institute of Technology. Boston. Mimeo. Tendler, Judith. 2000. The Economic Wars Between the States. Department of Urban Studies and Planning, Massachusetts Institute of Technology. Boston. Mimeo. Tendler. 2001. Transforming Local Economies: Lessons from the Northeast Brazilian Experience. Department of Urban Studies and Planning, Massachusetts Institute of Technology. Boston. Mimeo. Tendler. 2002. The Fear of Education. Department of Urban Studies and Planning, Massachusetts Institute of Technology. Boston. Mimeo. Vasconcelos, Jos6 Romeu de. 1999. Ceara, Pernambuco, Parana e Rio Grandedo Norte Economia, Finan,as Puiblicas e Investimentos nos Anos de 1986 a 1996. IPEA, Discussion paper #626. Brasilia. Vasconcelos, Jos6 Romeu de, Manoel Bosco de Almeida and Almir Bittencourt da Silva. 1999. Ceara: Economia, Finan,as Publicas e Investimentos nos Anos de 1986 a 1996. IPEA, Discussion paper #627. Brasilia. World Bank 1993. The EastAsian Miracle. Washington, DC World Bank 1995. World Development Report: Workers in an Integrating World. Washington, DC World Bank. 1997. Project Appraisal Document: Secondary School Education Project - El Salvador. Washington, DC. World Bank. 2000a. Brazil: Poverty Reduction, Growth and Fiscal Stability in State of Ceara. A State Economic Memorandum. Brazil Country Management Unit. Brasilia. World Bank. 2000b. Attacking Brazil's Poverty: A Poverty Report with a Focus on Urban Poverty Reduction Policies. Brazil Country Management Unit/PREM Sector Management Unit. Washington, DC. World Bank. 2000c. Brazil: Critical Issues in Social Security. Washington DC. World Bank. 2000d. Brazil: Broadening the Base for Growth: A Report on the State of Bahia. Washington DC. 64 World Bank. 2000e;. World Development Indicators World Bank. 2000/01. World Development Report: Attacking Poverty. Washington, DC. World Bank. 2001 a. Brazil: Eradicating Child Labor in Brazil. Human Development Sector Management Unit & Brazil Country Management Unit. Washington, DC. World Bank. 2001 b. Brazil: An Assessment of the Bolsa Escola Program. Human Development Sector Management Unit & Brazil Country Management Unit. Washington, DC. World Bank. 2001 c. Ceara Basic Education Quality Improvement Project. Project Appraisal Document. Washington, DC. World Bank. 2001 d. Brazil: Teacher Development and Incentives: A Strategic Framework. Washington, DC. World Bank 2001 e. Brazil - Rural poverty alleviation in Brazil: towards an integrated strategy Washington, DC. World Bank. 2003. Strategies for Poverty Reduction in Ceara. Background Material. Mimeo. Washington, DC. 65