Report No. 20488-NI Nicaragua Poverty Assessment Challenges and Opportunities for Poverty Reduction Volume l: Main Report February 21, 2001 Poverty Reduction and Economic Management Sector Unit Latin America and the Caribbean Region Document of the World Bank CURRENCY EQUIVALENTS US$1 = 12.60 C6rdobas (2000) US$1 = 10.58 C6rdobas (1998) FISCAL YEAR January 1 to December 31 ASDI Swedish Agency for International Development BCN Central Bank of Nicaragua CORNAP Corporaci6n Nacional del Sector Puiblico DANIDA Danish International Development Agency DHS Demographic and Health Survey ENACAL Water and Sanitation Public Company ERP Effective Rate of Protection ESDENIC Encuesta Socio Demografica de Nicaragua FISE Emergency Social Investment Fund GDP Gross Domestic Product GON Government of Nicaragua HIPC Highly Indebted Poor Countries IDB Inter-american Development Bank IDF World Bank Institutional Development Fund IDG International Development Goals INIFOM Municipal Development Institute INSS Social Security Institute LSMS Living Standards Measurement Survey MAGFOR Ministry of Agriculture MECOVI Programa de Mejoramiento de Encuestas de Condiciones de Vida MIFAM Ministry of the Family MINSA Ministry of Health MITRAB Ministry of Labor NORAD Norway Agency for Development NQPE Nicaragua Qualitative Poverty and Exclusion Study NRP Net Rate of Protection PAD Project Appraisal Document PROTIERRA Project for Rural Municipalities PRSP Poverty Reduction Strategy Paper SAS Social Action Secretariat SETEC Technical Secretariat to the Presidency SNIP National System for Public Investment UNAM Autonomous University of Nicaragua UNFPA United Nations Population Fund UNDP United Nations Development Programme Vice President: David de Ferranti Country Director: Donna Dowsett-Coirolo Sector manager: Norman Hicks Lead Economist: Ian Bannon Task Manager: Florencia T. Castro-Leal NICARAGUA POVERTY ASSESSMENT: CHALLENGES AND OPPORTUNITIES FOR POVERTY REDUCTION TABLE OF CONTENTS VOLUME 1: MAIN REPORT EXECUTIVE SUMMARY i INTRODUCTION I CHAPTER I. MACROECONOMIC AND STRUCTURAL TRANSITION IN NICARAGUA 2 A. NICARAGUA'S ECONOMIC RECOVERY IN THE 1990s 3 B. SIZE AND ROLE OF GOVERNMENT 4 C. MACROECONOMIC STABILITY AND PERFORMANCE 4 D. THE EXTERNAL SECTOR 6 E. THE EXTERNAL DEBT 7 CHAPTER II. THE NATURE OF POVERTY IN NICARAGUA 10 A. THE EVOLUTION OF POVERTY 1993-1998 10 B. THE POVERTY PROFILE TODAY 16 C. THE QUALITATIVE "FEEL" OF POVERTY 24 CHAPTER III. PUBLIC EXPENDITURES ON SOCIAL SERVICES 28 A. TOTAL SOCIAL SPENDING BY PROGRAMS 30 B. SUSTAINABILITY OF ADVANCES IN SOCIAL SERVICE PROVISION AND SOCIAL OUTCOMES 37 CHAPTER IV. AGRICULTURE AND RURAL POVERTY 42 CHAPTER V. KEY ELEMENTS OF A POVERTY REDUCTION STRATEGY 54 A. PROSPECTS FOR REDUCING POVERTY 54 B. SUSTAINING THE POVERTY REDUCTION MOMENTUM 55 REFERENCES OF THE MAIN REPORT TEXT TABLES: 1.1 SOCIAL INDICATORS, 1977-1993 3 2.1 NICARAGUA - POVERTY TRENDS, 1993-1998 11 2.2 NICARAGUA - POVERTY EVOLUTION BY REGION, 1993-1998 12 2.3 NICARAGUA - FEMALE OPPORTUNITIES BY SES QUINTILE 17 2.4 NICARAGUA - PERCENT OF CHILDREN UNDER FIVE SUFFERING MALNUTRITION, 1993 AND 1998 20 2.5 INTERNATIONAL COMPARISONS FOR HEALTH AND EDUCATION 20 2.6 WELL-BEING IN AN ASSET-BASED FRAMEWORK AS DEFINED BY POOR PEOPLE 24 2.7 PRIORITY NEEDS AS DEFINED BY THE POOR 26 3.1 NICARAGUA - SECTOR DISTRIBUTION OF CENTRAL GOVERNMENT SPENDING 28 3.2 NICARAGUA - SPENDING IN HEALTH AND EDUCATION 29 3.3 NICARAGUA - TOTAL SOCIAL SPENDING BY PROGRAMS, 2000 30 3.4 NICARAGUA - EDUCATION SPENDING BY PROGRAMS, 2000 31 4.1 DISTRIBUTION OF EMPLOYMENT BY SECTOR, 1998 43 4.2 AGRICULTURAL PRODUCTION: TOTAL GROWTH 1993-98 44 4.3 AGRICULTURAL PRODUCTION - OUTPUT GROWTH BY REGION 44 4.4 AVERAGE SHARE OF HOUSEHOLD MEMBERS THAT ARE EMPLOYED 46 4.5 NICARAGUA: THE EVOULTION OF SOURCES OF INCOME 1993-98 46 4.6 EMPLOYED PERSONS IN THE HOUSEHOLD WORKING IN AGRICULTURE 47 4.7 PROTECTION INDICATORS 48 4.8 PRODUCTIVITY IN AGRICULTURE 50 5.1 POVERTY PROJECTIONS WITH ALTERNATIVE ANNUAL GROWTH OF PER CAPITA CONSUMPTION 54 TEXT BOXES: 1.1 A CHRONOLOGY OF MAJOR POLITICAL EVENTS AND NATURAL DISASTERS 3 1.2 ECONOMIC PERFORMANCE, 1993-98 5 2.1 MEASURING AND COMPARING POVERTY 10 2.2 THE IMPACT OF HURRICANE MITCH 12 2.3 THE ATLANTIC REGION 14 2.4 SEGREGATED GENDER ROLES CAN HAMPER THE ATTAINMENT OF HIGHER WELFARE LEVELS 18 2.5 RURAL BOYS' EDUCATIONAL ATTAINMENT IS SEVERELY HINDERED BY CHILD LABOR 21 2.6 ADOQUIN ROADS CAN BE TARGETED TO THE RURAL POOR WITH SIGNIFICANT SPILL-OVERS 23 3.1 DETERMINANTS OF INFANT AND CHILD MORTALITY 32 3.2 NINDIRI CASE STUDY 35 3.3 COST OF MEDICINES LIMIT USE OF PUBLIC HEALTH SERVICES 39 4.1 DECOMPOSITION ANALYSIS OF CHANGES IN POVERTY 45 4.2 INCREASED LABOR PATICIPATION WITH FALLING AGRICULTURAL WAGES 47 4.3 REDISTRIBUTION EFFECTS OF AGRICULTURAL INCENTIVE POLICIES 49 4.4 DIRECT AND INDIRECT EFFECTS THAT AUGMENT THE NEGATIVE IMPACT OF TOT SHOCKS 51 TEXT FIGURES: 1.1 INCOME & CONSUMPTION: 1965-1999 2 1.2 NICARAGUA TOTAL ABSORPTION 4 1.3 CONSUMPTION AND INVESTMENT 5 1.4 EXTERNAL SECTOR 6 1.5 CURRENT ACCOUNT DEFICIT 6 1.6 TOTAL EXTERNAL DEBT 8 1.7 EXTERNAL DEBT SERVICE 9 2.1 URBAN/RURAL HEADCOUNT INDEX (LSMS 1993 & 1998) 11 2.2 PRENATAL CARE 15 2.3 INSTITUTIONAL BIRTHS 15 2.4 BIRTHS ATTENDED BY DOCTOR 15 2.5 CHILDREN AGED 7-12 NOT ATTENDING SCHOOL 15 2.6 CHILDREN AGED 4-6 ATTENDING SCHOOL 15 2.7 GLOBAL FERTILITY BY QUINTILE 17 2.8 REPRODUCTIVE PREFERENCES 17 2.9 STUNTING BY AGE AND POVERTY GROUP 19 3.1 INCIDENCE OF HEALTH SPENDING, 1993 38 3.2 INCIDENCE OF HEALTH SPENDING, 1998 38 3.3 INCIDENCE OF EDUCATION SPENDING, 1993 38 3.4 INCIDENCE OF EDUCATION SPENDING, 1998 38 3.5 SHARE OF PER CAPITA HOUSEHOLD EXPEDITURES DEVOTED TO EDUCATION 40 3.6 SHARE OF PER CAPITA HOUSEHOLD EXPENDITURES DEVOTED TO HEALTH 40 4.1 EXPORT CROPS: AREA AND YIELD 42 4.2 BASIC GRAINS: AREA AND YIELD, 1990-98 42 4.3 AGRICULTURAL PRICES - EXPORT GOODS 43 4.4 AGRICULTURAL PRICES - BASIC GRAINS 43 4.5 COFFE PRICES, 1993-1999 51 VOLUME 2: ANNEXES (available on request) I. THE CONSUMPTION AGGREGATE, by Carlos Sobrado 2. MEASURING AND COMPARING POVERTY, PRE AND POST-MITCH, AND FUTURE POVERTY PROJECTIONS, by Florencia T. Castro-Leal and Carlos Sobrado 3. THE INCOME AGGREGATE, by Carlos Sobrado 4. SUMMARY STATISTICAL APPENDIX 5. STATISTICAL APPENDIX 6. ANALYSIS OF POVERTY COMPARISONS: CONFIDENCE INTERVALS SENSIVITY ANALYSIS, PRICE CHANGES AND DOMINANCE CONDITIONS, by Florencia T. Castro-Leal and Carlos Sobrado 7. INEQUALITY COMPARISONS, by Florencia T. Castro-Leal and Carlos Sobrado 8. COMPARING POVERTY FOR THE YOUNG AND THE OLD, by Carlos Sobrado 9. A SHORT GUIDE TO POVERTY AND INEQUALITY MEASURES, by Gabriel Demombynes 10. THE DETERMINANTS OF POVERTY IN NICARAGUA, by Carlos Sobrado 11. HEADSHIP, GENDER AND POVERTY IN NICARAGUA, by Nadeem Ilahi 12. A GROWTH ACCOUNTING APPROACH, by Ulrich Lachler 13. EDUCATION AND POVERTY IN NICARAGUA, by Gustavo Arcia 14. RURAL EDUCATION EXPERIENCES, by Gustavo Arcia and Vanessa Castro 15. RATES OF RETURN TO EDUCATION, by Diana Kruger 16. DETERMINANTS OF PRIMARY SCHOOL ATTENDANCE, by Diana Kruger 17. DEMAND FOR HEALTH CARE, by Mukesh Chawla 18. MALNUTRITION AMONG PRE-SCHOOL CHILDREN, by Mukesh Chawla 19. NICARAGUA POVERTY MAP FOR TARGETING THE EXTREME POOR, by Florencia T. Castro- Leal, Carlos Sobrado, Peter Lanjouw, Berk Ozler, Gabriel Demombynes, Carlos Lacayo, Juan Rocha, Dulce Maria Mayorga, and Luis Alaniz 20. NICARAGUA QUALITATIVE POVERTY AND EXCLUSION STUDY, by Mary Lisbeth Gonzalez 21. PUBLIC SOCIAL SPENDING AND PROGRAMS BY SECTOR IN 2000, by Ema Budinich 22. THE INCIDENCE OF PUBLIC EXPENDITURES ON HEALTH AND EDUCATION, by Julia Dayton. 23. BETWEEN POVERTY AND PROSPERITY: RURAL HOUSEHOLDS IN NICARAGUA, by Benjamin Davis and Rinku Murgai 24. REDISTRIBUTION EFFECTS OF AGRICULTURAL INCENTIVE POLICIES, by Diana Kruger 25. LABOR MARKETS AND POVERTY REDUCTION, by Nadeem Ilahi 26. FINANCIAL MARKETS, by Susana SAnchez 27. DETERMINANTS OF INFANT AND CHILD MORTALITY, by Maria Femanda Muffiz 28. DO PUBLIC HEALTH AND WATER SERVICES IMPROVE CHILD HEALTH?, by Julia Dayton The Poverty Assessment Team: Acknowledgements This report is the product of intense collaboration between the Government of Nicaragua, the World Bank, and the Nicaragua MECOVI program to contribute to the Nicaragua Poverty Reduction Strategy Paper (PRSP). The World Bank's task team included: Norman Hicks (Poverty Reduction Sector Manager), Ana-Maria Arriagada (Social Protection Sector Manager), Helena Ribe (Sector Leader), Ulrich Lachler (Nicaragua Resident Representative), Florencia T. Castro-Leal (Task Manager), Mukesh Chawla, Nancy Gillespie, Nadeem Ilahi, Kathy Lindert, Kalpana Mehra, Berk Ozler, Laura Rawlings, Susana Sanchez, Carlos Sobrado, and Diane Steele. Gustavo Arcia, Ema Budinich, Vanessa Castro, Benjamin Davis, Julia Dayton, Gabriel Demombynes, Mary Lisbeth Gonzalez, Diana Kruger, Maria Fernanda Mufiiz, and Rinku Murgai provided valuable background papers and notes. Diana Kruger, Kalpana Mehra and Carlos Sobrado provided excellent research assistance. Kathy Bain, Sara Calvo, Joaquin Caraballo, Miguel Angel Castell6n, Klaus Deininger, Alejandro Izquierdo, Emmanuel James, Isabel Lavadenz, Maurizio Guadagni, Judith McGuire, Ana Julia Moreno, Norman Piccioni, Dagoberto Rivera, Andrea Verrnehren, and Tom Wiens also contributed. Aida Alvarado provided excellent administrative assistance and coordinated the production of the report. Peter Lanjouw, Giovanna Prennushi and Alberto Valdes are Peer Reviewers. Maria-Valeria Junho Pena is Peer Reviewer of the Nicaragua Qualitative Poverty and Exclusion Study. The team also benefited from the guidance of Ian Bannon (Lead Economist). The task team from the Government of Nicaragua included: Mario De Franco (Advisor to the President), Luis Duran (SETEC Secretary), members of the SETEC team (Luis Alaniz, Mario Arana, Francisco Diaz, Carmen Largaespada, Matilde Neret and Carlos Sevilla), actual and former members of MAGFOR (Carlos Arce, Roberto Bendania, Oscar Neira, Horacio Rose and Diana Saavedra), members of BCN (Mario Flores, General Director; Jose de Jes(is Rojas and Luis Rivas), members of FISE (Carlos Noguera, President; Carlos Lacayo, Technical Director; Erasmo Vargas, Operations Director; and Transito G6mez and Joaquin Murillo, Poverty Map. The National Institute of Statistics and MECOVI-Nicaragua task team included: Luis Benavides (INEC General Director), Gonzalo Cunqueiro (MECOVI-Coordinator), Margel Beteta (INEC Director Surveys and Census), Martha Vargas (National Coordinator LSMS98 & 99, and Poverty Map), Melva Bernales (International Coordinator LSMS98 & 99), Irene Alvarez (Methodology), Javier Argehal (Data Processing), Rutilio Moreno (Field Operation), Elsa Maria Gutierrez (Administrative Support). Francisco Arag6n, Denis Canizales, Silvia Chamorro, Thomas Gutierrez, Carol Herrera, Praxedis Hurtado, Elisa Lugo, Mayra Martinez, Arcides Menjivar, Yanett Narvaez, Rene Navarro, Javier Perez, Rolando Silva, Mario Solis, Juana Urroz y Cristina Zuiiiiga provided invaluable contributions. Juan Rocha and Dulce Maria Mayorga (Poverty Map). Edmundo Berumen and Alfredo Aliaga (Sampling). Vicente Merino, Alberto Valle, and Domingo Primante also contributed. Funding for this report, including the LSMS98 & 99 and the MECOVI-Nicaragua, was generously provided by the Government of Nicaragua, the World Bank (including funding from FISE, the IDF and the Strategic Compact), ASDI, DANIDA, IDB, NORAD, UNFPA and UNDP. Nicaragua Poverty Assessment Executive Summary The State of Poverty 1. In the decade of the 1990s, Nicaragua made extensive efforts to reduce civil violence, establish macroeconomic stability, and restore growth. Despite laudable successes on many fronts, poverty in Nicaragua continues to be pervasive and acute. About half of the population lives in poverty, and about 17 percent live in extreme poverty. While positive growth rates since 1993 have helped reduce the rate of poverty, a continued rapid growth of population has undercut these gains, and there are actually more people living in poverty today than in 1993. Thus, Nicaragua remains one of the poorest countries in Latin America. 2. In analyzing poverty in Nicaragua, a number of factors stand out: * Fertility rates are twice the Latin American average. As a result, Nicaragua's population is young-50 percent of the population is under 17 years of age. Furthermore, adolescent fertility rates are the highest in Latin America and represent an increasing share of all births. * Nicaragua exhibits a high incidence of domestic violence, which raises broad concerns about women's status and a lack of social cohesion. Almost one-third of Nicaraguan women report having been physically abused, frequently when their children are present and often during pregnancy. The shame of violence prevents most women from seeking help. * Poverty remains largely a rural phenomenon. About 70 percent of the rural population is poor, compared to about 30 percent for the urban population. Most of the poor, whether urban or rural, are heavily involved in agricultural employment. * Malnutrition is widespread and constrains child health, welfare and opportunities. Approximately 20 percent of children under 5 are chronically malnourished, or stunted. This share rises to 36 percent among the extreme poor. * Nicaragua suffers from elevated levels of infant mortality and maternal mortality, and a high prevalence of infectious and parasitic diseases. Although infant mortality has declined significantly since 1985 it still remains above the Central American average. Infant and maternal mortality account for the majority of all premature deaths and are due to infectious diseases, malnutrition and respiratory infections. * Economic growth has been largely financed from foreign capital inflows. While these have permitted high rates of investment with low levels of domestic savings, they have also raised the level of the exchange rate and discouraged the development of exports. ii 3. However, there are also a number of positive factors, reflecting both the recovery of the economy since 1993, and the Government's increased efforts to address social issues: * Nicaragua exhibits a low illiteracy rate compared to other low-income countries, but few years of schooling and great disparities in education attainment levels. Illiteracy is 19 percent among the total population over 10 years old, but 41 percent among the rural extreme poor. Urban/Rural Headcournt Index: % below poverty line (LSMS 1993 & 1998) * The share of Nicaragua's population living v7. below the poverty line declined, from 50 percent 80 in 1993 to 48 percent in 1998, and the share living 70 below the extreme poverty line declined from 19 60 68,6 percent to 17 percent. Although modest, these declines are statistically significant. 50 40 * Most poverty is found in rural areas, but most of 30 30.5 the gains in poverty reduction have also occurred in 20 rural areas as a result of a recovery in agriculture Urban Rural (see graph). Urban poverty has actually increased 13 199 in the Pacific and Atlantic regions. * Social outcome indicators have generally improved during the 1990s; both infant and child mortality rates improved by 20 percent between 1993 and 1998, and the coverage of primary schooling increased by 9 percent overall. * Coverage of basic social services also has improved. In all sectors, the Government of Nicaragua undertook a considerable effort to prioritize basic services for the poor. Significant gains are evident in access to prenatal care (14 percent), institutional births (23 percent) and physician attention during childbirth (29 percent). In all cases, the gains were driven by improvements in rural areas. * Gains in primary school coverage were almost exclusively rural, and predominantly benefited the poor, with a 26 percent gain in coverage among the extreme poor (as measured by the share of 7-12 year olds attending school). * Fertility declined significantly, from 2.9 births per woman in 1993 to 2.5 births in 1998. This decline continues the downward trend observed since 1985. * Access to other infrastructure services by the poor improved modestly. Investments in infrastructure mainly appear to have kept pace with population growth, resulting in modest gains in coverage of the population with access to these services. One sector where there have been major gains in infrastructure development is transport, which benefited from a significant expansion in the road network. * Hurricane Mitch did not significantly change Nicaragua's overall poverty profile. Hurricane Mitch had a terrible human impact as 3,000 persons died and 870,000 were displaced at least temporarily. Nevertheless, post-Mitch poverty levels broadly remained the same at the national, urban and rural levels. iii Perceptions of the Poor 4. In spite of the measured gains made in reducing poverty during 1993-98, the qualitative poverty work undertaken for this report finds that the poor associate the decade of the 1990s with a decline in their well being. What can explain this finding? First, the task of rebuilding Nicaraguan families and society, damaged during a decade of civil war, insecurity and violence, is just beginning. The prevalence and persistence of domestic and civil violence hint at the extraordinary social stresses that continue to affect Nicaraguan families. Keeping in mind that incomes in Nicaragua fell by over 50 percent from 1978 to the early 1990s, a 7 percent income improvement (as happened in 1993-98) is likely to pass unperceived by many segments of the population. 5. Second, being poor means living at risk. The risks faced by Nicaragua's poor may have increased during the 1990s. The replacement of a socialist state by a private market economy may have opened greater prospects for growth and economic opportunity, but it has left the poorest with virtually no formal safety nets to protect them. And after the extensive social and economic upheavals of recent history, Nicaragua's informal safety nets may also be fractured. 6. Third, the poor see improvements in infrastructure, but they do not perceive that their access to the services of that infrastructure has improved. The poor report that the main reasons for not using social services are cost and distance. The private costs of social services have risen fastest for the extreme poor, so that even when they do gain access, they face costs that may be becoming prohibitive. While infrastructure construction should have improved physical access, the impact of these investments has been less than anticipated. In many cases, rehabilitated facilities lack key elements for their operation, e.g., teachers, trained health staff, textbooks or medicines. Elements of a Strategy 7. The Government is in the process of defining a poverty reduction strategy. The purpose of this Poverty Assessment is to provide background material and analysis that will assist the Government in designing this strategy, but it is not itself designed to formulate that strategy. However, some key priorities emerge from the analysis that should be considered in any strategy for Nicaragua. These include: * Broad-based growth is essential to reduce poverty. Collapse of the economy during the 1980s, when consumption expenditures were reduced by 50 percent, was the single most important cause of growing poverty. Therefore, maintaining macroeconomic stability and providing an environment for private sector development remain key elements for assuring continued and accelerating growth of output, and thereby a reducing poverty. * Poverty reduction is limited by population growth. Nicaragua's high fertility rates represent a major constraint on poverty reduction. Although the demographic transition has begun, fertility declines are concentrated among the non-poor. Expanded access to family planning services, integrated with family and child health, are a high priority, but access alone is not enough. Efforts must be made to change family behavior patterns, provide education and employment opportunities for young women, and provide reproductive health counseling. * Sustaining rural income growth. Nicaragua's exceptionally high growth of agricultural output during the mid-1990s contributed importantly to the decline of extreme poverty, iv especially in the central region of the country. Many of the factors helping to produce this growth are not likely to be repeated, especially pacification in the countryside and favorable terms of trade. For the future, the two obvious areas of opportunity for boosting agricultural sector growth in Nicaragua would be accelerating technological advance and reducing the remaining anti-export bias in the sector. * Improvements are needed in the public sector to build strong programs that reach the poor. The overall amount of resources available from HIPC and from new official aid flows is likely to remain limited. Part of the HIPC resources will serve to prevent a rise in debt service, and while some reallocation of spending to the social sectors will be possible, given the high degree of poverty and overall needs, it is critical that government programs be well chosen and effective. * Foreign capital inflows distort expenditures. In general, the high reliance on official aid flows biases government spending unduly in the direction of new investments, and leads to a serious underfunding of maintenance and operations expenditures. Donors need to examine the allocation between infrastructure and recurrent investments and seek out a better balance. * Donors share the responsibility with the Government of Nicaragua to reconsider priorities in order to ensure greater impact of development assistance on poverty reduction. The PRSP is a Government-led effort to integrate all development agents including civil society in establishing a strategy for poverty reduction which will provide the basis to redefine public sector and donor funding priorities. Current social sector allocations, including external aid, are at more than half of overall public spending but consistently demonstrate low impact, with nutrition projects even demonstrating important negative impacts. To be consistent with the PRSP, it is crucial that donors coordinate together with the Government to choose results-oriented effective programs and decide to close down those that are clearly not attaining intended goals. Donors as well as the Government need to recognize the need for more specific, better targeted projects and reorient their assistance accordingly in response to the new poverty strategy. * Sustaining the provision of basic social services and building human capital. With resources scarce and poverty high, it is extremely important that public social services be extremely well focused and efficient. In general the social services suffer from a variety of problems, including low salaries for teachers and health workers, high levels of administrative costs, and the absence of a system of program evaluation. - In the health sector, there are systemic problems that prevent effective public sector delivery of health services outside urban areas. Two areas with much room for improvement are the management of health personnel, which is heavily weighted toward urban, curative, hospital-based care. Not enough emphasis is placed on the more cost effective model of rural, preventive, community-based care. Nicaragua's health system also needs to move from an input-driven system to a results-oriented system that places households at the center of the health strategy. > In the education sector, past gains may not be sustainable. Substantial gains in school attendance during the mid-1990s reflect the considerable demand by families for schooling. In order to sustain these advances, consideration should be given to developing targeted demand subsidies to the poorest, expansion of coverage in rural areas, programs to improve v retention, and the introduction of alternative delivery mechanisms, such as distance learning to reach remote areas more cost effectively. > Nutrition programs are extensive, but they do not target their actions sufficiently to reach the neediest. The problem of malnutrition in Nicaragua is a function of behavioral patterns as well as a matter of food availability or income. Existing programs do not deliver a service that will promote a sustainable change in nutritional status, do not define impact goals, do not define or use monitoring and evaluation mechanisms, and do not coordinate among themselves. Future efforts should focus on a strategy to address chronic and acute malnutrition, and to reduce micronutrient deficiencies. At the national level, program reformulation should emphasize better entry and exit criteria based on consistent targeting criteria across programs, use of evaluations of cost effectiveness and sustainability of different programmatic approaches to guide investment decisions, systematic impact monitoring and evaluation, and improved program coordination. At the local level, programs should focus on nutrition education, to help families improve their own nutritional status * Establishing effective social protection mechanisms for the poor. Given that many measures designed to achieve faster growth may also increase an economy's vulnerability to shocks, the creation of effective social protection mechanisms is needed to complement a growth-based poverty reduction strategy. Regardless of the source of a crisis, the poor typically develop coping strategies that are designed to minimize their vulnerability to crises at the local or household level, such as crop failure, catastrophic illness, marital abandonment and bouts of unemployment. While minimizing risk, these coping strategies also have the effect of perpetuating poverty. Effective safety nets are needed under these circumstances to provide sufficient insurance against such detrimental behaviors. Some guiding elements for the design of effective social protection mechanisms in Nicaragua should include: a detailed classification of the most vulnerable and the nature of their vulnerabilities; a clearly defined benefit tailored to address the vulnerability of a particular group, which may entail direct transfers in some cases and access to insurance markets in others; and targeting mechanisms that direct programs to the intended beneficiaries. Nicaragua Poverty Assessment: Challenges and Opportunities for Poverty Reduction INTRODUCTION 1. Nicaragua experienced a major economic and political transition during the 1990s. The civil war that raged during the 1980s came to an end, authoritarian political systems gave way to increasing democratic participation, macroeconomic stability was established and major policy reforms were implemented to transform a centrally planned, socialist economy into a private sector-led, market economy. Living standards had been spiralling down since 1978, and this transition began when Nicaraguans increasingly banked their hopes on fundamental changes as the only way to reverse the country's economic and social decline. This report seeks to find out to what extent these hopes were borne out by examining how the country's living standards have evolved during the 1990s. A key objective is to identify segments of the poor whose living standards have deteriorated and to look for warning signals with respect to the most vulnerable members of the population. This should help in identifying possible corrective actions or compensatory interventions. Positive experiences should provide useful lessons for policies and strategies to be maintained or deepened in future poverty reducing efforts. 2. This report is based largely on the findings from the 1998 Living Standards Measurement Survey (LSMS) and its comparison with the findings of a comparable LSMS that was carried out in 1993. Overall, Nicaragua's poverty profile in 1998 turns out to be very similar to the poverty profile in 1993, which was analyzed in the Bank's 1995 Poverty Assessment (World Bank, 1995). This is to be expected, given the slow speed with which poverty indicators generally tend to evolve. The most interesting aspect of the present Poverty Assessment is the light cast on the evolution of poverty and its determinants, which has only become possible now with the existence of two poverty snapshots that are separated in time.' 3. The main intended audience of this report is the Government of Nicaraguan, which is preparing a Poverty Reduction Strategy Paper (PRSP) in collaboration with civil society. Nicaragua is one of four Latin American countries that are eligible for debt relief under the HIPC initiative. Under this initiative, and to underpin future IDA support, Nicaragua must prepare a results-oriented PRSP based on a sound understanding of the nature and determinants of poverty, and of the public actions that can help reduce it. This report is intended as a contribution toward that effort, helping to diagnose Nicaragua's poverty problems and suggesting options for the future, while leaving the formulation of the PRSP itself to the Government. 4. Chapter I looks at the macro situation in historical perspective, including structural reform during the 1990s, and the current debt and balance of payments situation. Chapter II examines poverty in Nicaragua today and during 1993-98, and includes the results of a qualitative assessment of poverty. Chapter III looks at public expenditures in the social sectors and their impact on poverty. Chapter IV looks at rural poverty and agriculture, while Chapter V lays out key issues that would be important to address in a poverty reduction strategy. ' The two survey years provide a good setting for comparisons: both are well after the conflict stopped (1989), the change of political regime (1990) and macro stabilization (1991), while key structural reforms were well underway by 1993. There has been no significant break in policy regime during 1993-98, nor significant external shocks; Hurricane Mitch (end-October 1998) struck after the survey was collected. CHAPTER I. MACROECONOMIC AND STRUCTURAL TRANSITION IN NICARAGUA 1.1 Nicaragua has made visible inroads in poverty reduction during the second half of the 1990s. However, it is still recovering from a traumatic social and political upheaval during the past two decades that destroyed infrastructure, reduced productivity and substantially increased the level of poverty. Real per-capita GDP fell by 40 percent in 1978-79, and then contracted at an annual rate of -3.0 percent over 1980-94. In spite of the eventual recovery of per capita growth, per-capita output at the end of the 1990s is still only around half of the per capita output produced in the 1960s and 1970s (Figure 1). Likewise, per capita consumption in 1994 was 47 percent of the level of 1978, and between 1994 and 1999 has increased only by 5 percent. While many other developments contributed to the evolution of poverty during this period, it is doubtful that any has been more decisive than this macroeconomic phenomenon. Figure 1.1 Income & Consumption: 1965-1999 12000 X ADP/capita 10000 o 8000 _____ _ _ X Cons6) ption/cap. oo 6000 z 4000 - 0 m2000 0 0 0 ) O- I 1-;Ijjj,~0 10 II0 Ii;;;g; I: '11 I _ _ ~ ~~~~_ _ _ _ _0 _0 _0 _ _ _ I~~~~t 00 UJ 00£t) 0£ 1.2 What was responsible for this major decline in income and welfare, which affected all Nicaragua's citizens, but particularly the poor? The major factors behind this collapse were: * The economic and social upheaval that accompanied the revolution in 1978-79 and the civil war that followed through much of the 1980s had a major impact on output. In addition to the visible destruction of human and physical capital, the unsettled conditions in this period kept away private investors and forced the reallocation of public resources toward unproductive activities, especially defense expenditures (see Box 1.1 for a chronology of political and natural events). * The adoption of a centralized, state-managed development strategy by the new regime in 1979 resulted in a bloated public sector that squeezed out private initiative. * Fiscal and monetary discipline was progressively lost, leading to macroeconomic destabilization that culminated in several hyperinflationary episodes. * The closure of the economy in the 1980s, both through restrictive trade policies meant to promote import substitution industrialization and the imposition of the US embargo, greatly reduced the opportunities for growth through specialization. * As macroeconomic imbalances deepened in the 1980s, the country's creditworthiness became impaired and a huge debt overhang developed that discouraged investors. 3 An increasing dependence on foreign savings partly helped to cushion the poverty impact of negative growth, but also contributed to an appreciation of the real exchange rate that rendered many domestic industries less competitive. 1.3 Despite the economic collapse, welfare as measured by social indicators actually improved over most of the period, even as private consumption was falling.2 As shown in Table 1.1, infant mortality was reduced from 90 (per 1,000 births) in 1978 to 48 in 1992, and primary net enrollment rose from 70 percent in 1980 to 79 percent in 1992. In part, this reflects a redirection of Government spending, especially during the early Sandinista years, toward the social sectors, which tended to protect real social spending even while the economy was shrinking rapidly. Also, as observed in other Latin American countries during the "lost decade" of the 1980s, the continued improvement in social indicators during this period of economic decline is likely to be reflecting the lagged impacts of earlier social investments. Table 1.1: Social Indicators, 1977-93 Indicator 1977 1980 1987 1990 1992 Life expectancy (years) 58 59 62 64 66 Infant mortality (per 1,000) 90 84 65 55 48 Primary school enrollment, gross (%) 87 94 94 94 100 Primary school enrollments, net (%) - 70 72 72 79 Source: World Bank, SIMA data base 1.4 Looking at the composition of growth during these years, we find that there was little or no growth in capital, but continued growth in the quality of labor inputs (education). However, poor economic policies and mismanagement, plus the civil war, resulted in negative growth in total factor productivity. In the years since 1993, the resumption of growth appears to have continued to depend on the improvements in education, as well as positive growth in physical capital. Total factor productivity growth has been positive, but small. A. NICARAGUA'S ECONOMIC Box 1.1: A Chronology of Major Political RECOVERY IN THE 1990S Events and Natural Disasters Civil Strife and Public Expenditures Somoza Regime: 1930-79 Managua Earthquake 1972 1.5 With the end of the civil war and the Sandinista Revolution: 1979 adoption of a new economic policy regime in US Embargo: 1983-90 1990, the most obvious obstacle to growth Civil War: 1983-90 was eliminated. Social turmoil and armed Hurricane Joan 1988 violence in rural areas continued during the Election of Pres. Chamorro: 1990 .. . . . . .......... . Tidal Wave: 1992 demobilization and reinsertion in civil society Volcaner 1992 of former combatants in the early 1990s, but El Ninro i 1992,1994 both factors declined notably by the mid- HurricaneMitch: 1998 1990s (Box 1.1). The onset of peace has permitted the Government to reduce the size of its armed forces and to substantially reorient its spending away from "unproductive" defense expenditures. 2 Because of an absence of comparable household surveys during this period, it is impossible to make estimates of poverty levels prior to 1993. 4 1.6 This "peace dividend" of roughly 25 percent of GDP was mainly spent on normalizing financial relations with international creditors (raising debt service from about I percent of GDP in the mid-1980s to 8-9 percent in the late 1990s) and restoring macroeconomic stability by reducing the fiscal deficit (mainly through reducing total public expenditures from an average of 50 percent of GDP in the mid-1980s to 33 percent in the late 1990s). Social sector spending was broadly maintained, but declined in real per capita terms until 1997. Spending for infrastructure expanded somewhat during the period. B. SIZE AND ROLE OF GOVERNMENT 1.7 Nicaragua's public sector shrank considerably in the 1990s as two successive post- Sandinista governments adopted a private sector-oriented development framework. The rising tendency in private sector absorption (investment and consumption spending) after the early l 990s was accompanied by a declining GDP-share of public absorption (Figure 1.2). This pattern reflects both a contraction in Government spending, as well as the divestiture of most state-owned commercial enterprises. The most important structural measures were: * The state managed 351 commercial enterprises in 1990, accounting for almost 30 percent of GDP, under the state holding company, CORNAP. All but 6 of these enterprises were either closed or privatized by 1995. * Whereas the banking sector was fully state-owned in 1990, less than 3 percent of banking sector assets and liabilities continued under state-management in 1999. * Public sector employment was reduced from 285,000 persons in 1990 (or 24 percent of the labor force) to around 100,000 persons in 1998 (5.3 percent of the labor force). Figure 1.2 Nicaragua Total Absorption 0 35000 30< 25000 - /\ Total Absorption 25000 - .n 20000 3 X60 15000 Privatc - 10000 p > : 3 oooo -i5 .2 5000 Pubi C. MACROECONOMIC STABILITY AND PERFORMANCE 1.8 With the rapid expansion of the public sector in the 1980s, fiscal control became more difficult. At the same time, the exigencies of war in this period often demoted the objectives of fiscal balance to a secondary order of priority. This resulted in increasingly higher inflation rates that culminated in hyperinflation during the later part of the 1980s. At its peak, inflation reached over 14,000 percent inI988, and over 7,000 percent in 1990. In the 1990s, fiscal discipline was progressively restored and inflation was brought down to under 10 percent per annum. This 5 change in policy has contributed greatly to a more predictable macroeconomic environment and improved investor confidence. As a result, after years of declining output, Nicaragua has been able to have a sustained growth rate of 4.3 percent, during the 1993-98 period, with an estimated growth of 7 percent in 1999. 1.9 Although the Government succeeded in slimming down and cutting back its fiscal deficits, Nicaragua as a whole has been highly dependent on foreign capital inflows, especially aid, that are now much higher than in the 1970s and early 1980s. The drawback is that this inflow has tended to appreciate the real exchange rate, inhibiting a faster export growth and the development of more competitive import-substituting industries. However, it has also permitted a higher rate of investment without suppressing domestic consumption, i.e., permitted a higher level of domestic absorption (Figure 1.3). Figure 1.3 Consumption and Investment 160% s 140% _ _ _ 120% -Total Absorption A 100% - -- = 80% - _ r Consu-ption 60% __ 40% Total Investment ow_ private Invcst. ¢ 20% - u~0% 0. 0' --j 0 O 00 00 '. 'O I~~~~~~~u - U. \o w -.a£W I~ ~ ~~ . :J - .J . 1.10 During the period 1993-98, domestic investment averaged about 27 percent of GDPI, while foreign capital flows averaged about 34 percent, and thus they financed over 100 percent of domestic investment and part of domestic consumption (Box 1.2). Without these inflows, domestic savings would have had to be raised and private consumption reduced in order to Box 1.2: Economic Performance,1993-98 maintain the same investment rate. Furthermore, the efficiency of investment GDP growth (per annum) 4.3% rA Consumption/GDP 107.0% seems very low. An investment rate of 27 Investment/GDP 27.0% percent with an average growth during the Exports/GDP 31.5% period of 4.3 percent, implies an incremental Imports/GDP 65.6% capital output ratio of 6.3, which is high by Net inflow (M-X) 34.1% developing country standards. Given the fact ICOR 6.3 that the economy had the capacity at one time Source: Central Bank data to produce a larger GDP, and present investments in some part are restoring past capacity, it is surprising that growth has not been stronger. 3 Calculated from Central Bank data using nominal price national accounts data. Data for 1999 are not used because of the distortion caused by heavy investment after Hurricane Mitch. 4 The apparent low productivity of investment during the 1 990s could be reflecting allocative inefficiencies in recent investment decisions due to errors caused by rapidly changing conditions during the transition from a socialist to a market economy, as well as a rapid obsolescence of capital accumulated during the 6 D. THE EXTERNAL SECTOR 1.11 The 1990s saw a progressive opening up of the economy. The exchange rate was unified, quantitative import barriers and export taxes were removed, all state export and import monopolies were eliminated. The ceiling on import tariffs was progressively reduced from over 60 percent in 1990 to 10 percent in 1999. Also, all domestic price controls were removed, except for public utility services, some petroleum products, milk and basic medicines. These measures have caused the GDP-share of exports and imports to double, from a combined share averaging 55 percent in 1980-90 to over 100 percent in 1996-99 (Figure 1.4). Figure 1.4 External Sector 80% X6 % Imports 6 40% /o - -- 40% 20% -- - 'S 0% -- -20%/ :: i. t ~~ -40% ~ Resource Balance V740%/ .. 1hesond tO i u vo in 1.12 Export growth responded favorably to the trade liberalizing measures introduced in the first half of the 1990s, but with a lag, while imports initially responded faster, widening the trade deficit. This gap was gradually decreasing during 1993 through 1997, in part aided by improving terms of trade, until another exogenous event, Hurricane Mitch, reversed this trend in 1998, at least temporarily. Sandinista regime. Similarly, the high incremental capital output ratio could be reflecting measurement problems. One common source of error is the misclassification of current expenditures as capital expenditures, especially when these are financed by foreign aid. It is also likely that the national income accounts understate GDP, and by extension, GDP growth. For instance, private consumption per capita in the national accounts in 1998 was C$4,344 while the LSMS found an average of C$6,445. This would imply that the national accounts understates private consumption by about 33 percent. In most countries, the underreporting of income results in the opposite effect: consumption data in surveys is usually smaller than consumption data in national accounts. Hence, the national accounts may be seriously understating consumption, and by inference, total GDP. For an estimate of the underreporting found in typical surveys, see Wodon, et. al. (2000). 7 Figure 1.5 Current Account Deficit (+Reserve Adjustments) 1200 a1000 -- - '600 - c400_ d M 200 - 1.13 In spite of the improved export performance in 1994-97, it is important to keep in perspective that the current US Dollar value of merchandise exports in 1997 is just nearing the levels reached in the late 1970s, before the economy began its long downward slide. The US Dollar declined in value over this period, so the deterioration in exports was even more pronounced in real terms. In contrast, the current US Dollar value of merchandise imports reached about twice the value in the late 1970s, and about two and a half times if one adds the Mitch-induced reconstruction-related imports. This uneven evolution of exports and imports is related to the fact that substantial foreign inflows, both private and official, are available to finance the current account deficit (Figure 1.5). 1.14 Sustainability of Growth with High Capital Inflows. Poverty reduction will depend heavily on continued growth, and growth in turn on future investments. The large size of the current account deficit implies that Nicaragua now relies almost exclusively on foreign savings to finance its investment needs. Capital inflows have been dominated in the past by official flows, both loans and grants, but private flows are playing an increasingly important role, as private capital has been attracted by the reforms undertaken. 1.15 Recent crises in Latin America and East Asia have illustrated the dangers of sudden private capital flow reversals. However, this does not appear to be the case for Nicaragua, where private capital flows do not appear to be as responsive to short term yield differentials, perhaps because of the relative underdevelopment of capital markets. For instance, during the East Asia crisis, exchange rates and interest rates showed no perceptible pattern of movement. On the other hand, there is now a greater danger of capital flow volatility than in earlier periods arising from the official flows side, particularly as donors increasingly link their flows to policy conditionality and governance issues within the country. This link underscores the importance of strengthening governance, both to improve overall development effectiveness but also to reduce volatility in aid flows. Separately, there is also the problem of "aid fatigue" and the gradual reduction observed worldwide in the supply of official assistance. This raises the need over the longer term to continue efforts to increase domestic savings (especially by strengthening the public sector's revenue base) and exports, so as to reduce the country's dependence on foreign savings and capital inflows. E. THE EXTERNAL DEBT 1.16 Nicaragua's total external debt ballooned in the 1980s and peaked in the early 1990s at US$12 billion, which was roughly 6 times the value of GDP. Much of this debt consisted of 8 arrears, which accumulated rapidly with the suspension of debt servicing in the mid-1980s (Figure 1.6) Nicaragua was able to reduce its overall debt through various renegotiations and debt-reduction deals, so that by 1995 its overall external debt fell to about US$6 billion. However, with a net present value of about 540 percent of exports after traditional debt relief mechanisms have been exhausted, Nicaragua's debt remains unsustainably large. By comparison, the objective of the enhanced HIPC program is to reduce the net present value of debt of eligible countries to a ratio of no more than 150 percent of exports. Figure 1.6 Total External Debt 14 120 - - a 8 | iw Arrears 2_ 0 '0 '0 \ 00 00 00 '0 SD O 'i c ' t'I bJ CA - 0 1.17 The direct benefits of the HIPC Initiative for Nicaragua are significant: the Initiative is expected to provide about US$4 billion in debt service savings on a scheduled basis over 20 years, or US$215 million annually on average. This represents roughly two-thirds of the annual combined budgets of the health and education ministries. However, it would be misleading to equate this entire amount of debt service relief to an enlargement of the public sector's total resource envelope. Although the projected debt service relief is very large compared to what Nicaragua would otherwise have to pay, it is relatively modest compared to what it had been paying in recent years (Figure 1.7). Because of debt service deferrals provided by the Paris Club and other bilateral creditors in the aftermath of Hurricane Mitch, Nicaragua's actual debt service flows declined significantly during 1998-2000. These deferrals are coming to an end in 2001, so that the scheduled debt service in the absence of HIPC would increase significantly after 2000. Initially, most of the assistance to be provided under the HIPC initiative will prevent an increase in debt service payment, rather than achieve major reductions in debt service below the already low levels of the late 1990s.5 This means that most ministries cannot initially expect to receive significant increases in additional resources with which to implement new programs. However, it also means that the Government would not have to institute major expenditure reductions and cut back public services to accommodate rising debt service, or raise new taxes. Reducing the threat of future tax hikes, in turn, should boost private investor confidence and thereby contribute to the expansion of Nicaragua's overall resource envelope. 5 For example, in 1999-2000, the Nicaraguan Government paid around 6.3 percent of GDP in servicing its external debt. This debt service was scheduled to increase to 7.1 percent in 2001, but with the relief to be provided under the HIPC Initiative, this service declines to 4.4 percent, representing a savings of 2.7 percent of GDP. This lower debt service level, however, is only 1.9 percent of GDP lower than what the Government had been paying in 2000, so that the overall gain in resources to be expected by the Government in 2001 is less than the total amount of debt service savings. 9 Figure 1.7 External Debt Service 1800 _ 1600 -- - - a tutal projected c1400 __A I X 1400 )Debt ServiceDue X - ~1200 : 800 a_ter _ LC X 600 _ Paid priorizadoi 4 400 - - - 2000 0, - -a 00 00 \t 'D 'IO O O o .Al 00 too a9 0) o 00 tQ 0 1.18 In December 2000, Nicaragua reached the Decision Point under the HIPC Initiative, which triggered certain interim debt relief assistance from several donors. To conclude the definitive write-off of its debt however, Nicaragua must reach the Completion Point under the Initiative. To achieve this, the Government needs to demonstrate that it has a viable program for reducing poverty. Donors want to be assured that incremental resources made available under the debt relief program will actually benefit the poor. To build such a strategy, there is a need to have a better understanding of the current state of poverty in Nicaragua, who the poor are and why are they poor. CHAPTER II. THE NATURE OF POVERTY IN NICARAGUA A. THE EVOLUTION OF POVERTY 1993-98' 2.1 Since 1993, Nicaragua's economy has experienced a recovery, which has had a beneficial effect on poverty. GDP averaged annual growth of 4.3 percent during 1994-98, and per capita consumption 1.4 percent. As a result, the share of Nicaragua's population living below the poverty line declined slightly, from 50 to 48 percent, and the share living below the extreme poverty line declined from 19 percent to 17 percent (Box 2.1)2 Although modest, these reductions in both overall and extreme poverty are statistically significant (see Annex 6). The improvement is largely due to the decrease of poverty in rural areas, where the incidence of extreme poverty fell by 7.4 percentage points and the incidence of poverty fell by 7.6 percentage points (Table 2.1). Box 2.1: Measuring and Comparing Poverty This report measures poverty with reference to poverty lines constructed on the basis of consumption.4 A key objective of the current Poverty Assessment is to compare poverty over 1993-98. In the 1993 Poverty Assessment,3 the general and extreme poverty lines were determined by first, estimating the relationship between caloric intake and total/food expenditures5 using the regression: LnCj = a + 3 LnXj + ej where LnCj is the natural log of per capita (adult equivalent) caloric intake for the jth person, and LnXj is the natural log of per capita total/food expenditures for the jth person, and second, deriving the poverty lines values from the estimated regression by replacing the caloric requirement of 2,2266 calories/day for Cj, and the estimated a and 3 values. In this report, the 1998 general and extreme poverty lines, comparable with 1993 are obtained by first, constructing four consumption aggregates: two comprehensive for each year (1993 and 1998), using all of the available information for each year, and two common for each year (1993 and 1998), including only the information common to the two years.7 Second, the 1998 general and extreme poverty lines are inferred for the comprehensive 1998 aggregate by making use of the two common consumption aggregates. These poverty lines render consistent poverty comparisons between 1993 and 1998, and they are at once consistent with the 1993 Poverty Assessment but take into account the more comprehensive nature of the 1998 consumption aggregate.8 The results of the above-described process are the general and extreme poverty lines for Nicaragua in 1998. The general poverty line in 1998 is at C$4,259 per capita per year (or US$403, slightly more than US$1 per day). A person is considered poor if his/her total per capita annual consumption is below this amount. The extreme poverty line is at C$2,246 (or US$212). A person consuming less than this amount per capita per annum is considered extremely poor and she/he cannot meet her/his minimum daily caloric requirements even when the entire consumption is devoted to food. (For reference purposes, GDP per capita in 1998 was C$4,719, or US$446) 'Analysis of poverty conditions and trends prior to 1993 is not possible due to the absence of good survey information. 2 Poverty results in Nicaragua are based on a representative sample of households, and are therefore an approximation of the true values with a margin of error. Taking these into account, the incidence of poverty can be expressed within a range of minimum and maximum values with a 95 percent confidence level. With this confidence level, the incidence of poverty in Nicaragua can be depicted as follows: Year Point estimate 95% confidence interval 1993 50.3% 48.8 to 51.8% 1998 47.9% 46.3 to 49.3% 11 Table 2.1: Nicaragua: Poverty Trends, 1993-1998 (percent) Extreme Poor I All Poora 1993 1998 Change 1993 1998 Change Incidence: Headcount Indexb National 19.4 17.3 -2.1 50.3 47.9 -2.4 Urban 7.3 7.6 +0.3 31.9 30.5 -1.4 Rural 36.3 28.9 -7.4 76.1 68.5 -7.6 Depth: Poverty Gap Indexc National 5.9 4.8 -1.1 21.8 18.3 -3.5 Urban 1.6 1.9 +0.3 10.9 9.9 -1.0 Rural 11.8 8.3 -3.5 37.1 28.3 -8.8 Severity: Foster-Greer-Thorbecke Indexd National 2.6 2.0 -0.6 12.1 9.3 -2.9 Urban 0.6 0.7 +0.1 5.1 4.5 -0.6 Rural 5.5 3.5 -2.0 21.9 14.9 -7.0 Source: Nicaragua LSMS 1998. (a) "All poor" throughout this report includes the extreme poor; thus, the extremely poor are a sub- set of the poor. (b) The Headcount Index (Po) is the share of the population whose total consumption falls below the poverty line. (c) The Poverty Gap Index (PI) represents the amount needed to bring all poor individuals up to the poverty line (full or extreme), expressed as a percent of the poverty line taking into account the share of the poor population in the national population. (d) The Foster-Greer-Thorbecke Index (P2) is a derivation of PI that takes into account the distribution of total consumption among the poor, thus, it measures the degree of inequality among the population below the poverty line. 2.2 In spite of these declines the absolute number of persons in poverty and extreme poverty increased, as the modest poverty rate declines Figure 2.1 were swamped by an increase in the overall U H c Inde population. With an overall population growth Urban/Rural Headcount Index: rate of 2.6 percent, total population between l 1993 and 1998 grew by about 600,000, and the 80 1;7_6 numbers in poverty increased by about 200,000 (from 2.1 to 2.3 million). The numbers in 68,5 extreme poverty rose by about 20,000. 60 2.3 Poverty and extreme poverty remain 30 overwhelmingly rural (Figure 2.1). The rural 303 population represents 46 percent of the national 20 population. About 69 percent of the rural Et993 0t998 3 World Bank (1995), Volume II: Annexes, pp 2-5, 28-30. 4The measure of consumption in both 1993 and 1998, as in most poverty studies, does not include implicit in-kind transfers through government programs in the social sectors (see Annex 1). Thus, poverty estimates most probably overstate the poverty level by a few percentage points, particularly since the poor more often consume public services than the non-poor. 5Total consumption is used for the general poverty line and only food consumption for the extreme poverty line. Neither of them are adjusted for economies of scale or use adult equivalences. 6 The minimum caloric requirement of 66,780 calories per month is substituted into the equation (which is equivalent to 2,226 calories per day): X = e ** [(In 66,780 - ri)/l] 7The 93 common aggregate includes 75 percent of the original (comprehensive) consumption aggregate, and the 98 common aggregate includes 70 percent of the final (comprehensive) consumption aggregate value. 8 See Annex 2 for the complete methodology to measure and compare poverty developed on the basis of Lanjouw and Lanjouw (1997). The methodology applied to Nicaragua relies on the assumptions of comparable consumption pattems over 1993-98 and on the ability of the Consumer Price Index to capture local currency purchasing power changes. 9 The total population of Nicaragua was 4.175 million in 1993 and 4.803 million in 1998. 12 population was living in poverty in 1998, compared to 31 for the urban population. Not only are there more poor in rural areas, the depth of their poverty is greater. In general, the rural poor are almost three times further from the poverty line than the urban poor. However, the improvements in poverty during 1993-98 were larger in rural areas; rural poverty declined by 7.6 percentage points but the decline was only 1.4 percentage points for urban dwellers. Among rural areas, rural poverty is substantially higher in the Pacific than in the Atlantic regions. Table 2.2: Nicaragua: Poverty Evolution by Region, 1993-1998 (percent) Incidence of and Change in Incidence of and Change in Extreme Poverty Poverty Contribution to the Headcount Index Headcount Index change in Poverty Extreme Region 1993 1998 Change 1993 1998 Change Poverty Poverty National 19.4 17.3 -2.1 50.3 47.9 -2.4 Urban 7.3 7.6 +0.3 31.9 30.5 -1.4 Rural 36.3 28.9 -7.4 76.1 68.5 -7.6 Managua 5.1 3.1 -2.0 29.9 18.5 -11.4 -12.9 -45.5 Pacific Urban 6.4 9.8 +3.5 28.1 39.6 +11.5 +38.2 +79.2 Rural 31.6 24.1 -7.5 70.7 67.1 -3.6 -30.2 -7.8 Central Urban 15.3 12.2 -3.1 49.2 39.4 -9.8 -2.8 -13.8 Rural 47.6 32.7 -14.8 84.7 74.0 -10.7 -54.1 -29.4 Atlantic Urban 7.9 17.0 +9.0 35.5 44.4 +8.9 +29.1 +20.8 Rural 30.3 41.4 +11.1 83.6 79.3 -4.3 +32.7 -3.5 Source: Nicaragua LSMS 1998 and World Bank (1995) for 1993 data. (a)Departments by region: Managua includes Managua Urban and Managua Rural; Pacific includes Chinandega, Leon, Masaya, Granada, Carazo and Rivas; Central includes Nueva Segovia, Jinotega, Madriz, Esteli, Matagalpa, Boaco and Chontales; and Atlantic includes RAAN (North Autonomous Atlantic Region), RAAS (South Autonomous Atlantic Region) and Rio San Juan. (b) Nicaragua's population share by region in 1998 is: Managua 27.7%, Pacific Urban 19.7%, Pacific Rural 14.1%, Central Urban 10.9%, Central Rural 18.6%, Atlantic Urban 4.3% and Atlantic Rural 4.7%. The Pacific (including Managua) has 61.8%, the Central has 29.5% and the Atlantic 9%. 2.4 Poverty changes varied greatly by region (Table 2.2). The most notable reduction in the incidence of poverty between 1993 and 1998 was in Managua and the Central regions. These two regions contributed almost 90 percent to the reduction in poverty. Managua experienced a decline of 11 percentage points while the decrease in the Central regions was about 10 percentage points. By contrast, poverty increased sharply in the urban areas of the Pacific and Atlantic, while at the same time it was decreasing in the rural areas of these regions. While in most regions extreme poverty followed the trend in total poverty, in rural Atlantic there was a worrisome reverse trend; extreme poverty increased while total poverty fell, suggesting that the poorest were not benefiting from growth and recovery. Hurricane Mitch did not significantly change Nicaragua's overall poverty profile (Box 2.2 and Annex 2) Box 2.2: The Impact of Hurricane Mitch Hurricane Mitch did not significantly change Nicaragua's overall poverty profile. Post-Mitch poverty levels broadly remained the same at the national, urban and rural levels. Although about 3,000 people died and about 870,000 were displaced, the Pacific rural area experienced a slight improvement in the overall and extreme incidence of poverty, while the Central rural area experienced a slight deterioration. These changes, though significant, did not reverse the overall declining trend in poverty observed between 1993 and 1998, nor the regional rankings in terms of poverty levels. The slight improvement in Pacific rural is most probably related to the lesser impact of the Hurricane compared to Central rural and to the flow of international aid and the rebuilding process it brought about. In contrast, the slight worsening in the Central rural region can be explained by the Hurricane's impact on agriculture, by the region's more difficult accessibility and by its higher vulnerability as compared to the Pacific rural. 13 2.5 Managua concentrates over one-fourth of the total population in the country, it has the lowest poverty rates, is where poverty declined the most and its contribution to the decline in overall poverty between 1993 and 1998 was close to 50 percent. Managua experienced an average wage growth almost twice as large as overall Nicaragua, with wages outpacing inflation in agriculture, construction and financial services. Although few agricultural products are grown in the Managua region, it concentrates 22 percent of sugar cane production in the country. Economic growth in Managua during the period is mainly related to increased industrial, construction and financial activities. The export processing zone (Zona Franca) generated 18,400 direct and 1,700 indirect jobs during the period. The evolution of real wages in construction in Managua outpaced growth in all other sectors. The financial sector, although small, expanded rapidly and experienced high growth of wages and number of jobs. Employment by poverty groups shows that while agriculture favored mainly the extreme poor,° employment in manufacturing favored all groups, in construction it favored mainly the poor, and employment in the emerging financial sector favored poor and non-poor groups. 2.6 The Pacific region experienced negative growth rates of export crops and all basic grains, and switching to new crops, with a slight decline in rural poverty but a substantial increase in urban poverty."1 The production of cotton in the Pacific region-which historically brought most revenue to the region-disappeared by 1998. Real wages declined in all sectors. In Pacific Urban, manufacturing and public sector employment, sectors which used to be the largest employers, shrank drastically with wages declining even in nominal terms for the former. Employment in commerce and agriculture increased, but with falling real wages, reflecting more a coping strategy by low growth in labor demand in other sectors which comes about when preferred jobs are not available. Surprisingly for an emerging financial sector, the share of employment also declined with falling nominal wages. 2.7 The Central region experienced rapidly growing revenues from coffee production, with urban and rural poverty declining sharply. The Central region grows 95 percent of total coffee production in the country, which grew at an average of 9.5 percent per year in 1993-98. Furthermore, the international price of coffee grew at an average annual rate of 22 percent during the same period, providing the Central region with rapidly growing revenues. In the Central region, where the largest decline in rural poverty occurred, the share of employment in agriculture declined slightly (from 82 to 77 percent), with high paying sectors increasing sharply in urban areas such as manufacturing, construction, transport and services. 2.8 The Atlantic region has the worst poverty and social conditions in the country, and they have shown little progress or worsened since 1993. The Atlantic region covers about half of the country, but is sparsely populated, with only 10 percent of Nicaragua's population. The Atlantic Rural was the second poorest region in 1993 after Central Rural, but it is now the poorest region with nearly 80 percent of its population living in poverty. It has the highest extreme poverty rates, 17 percent urban and 41.4 percent rural. Its social indicators are not only the worst in the country, but they show very little progress since 1993. Births per woman are the highest at 3.7, compared to a national average of 2.5. Education and health indicators show almost no progress-two-fifths of 7-12 year-old children are not attending school, unchanged from 1993; 10 Managua is the only region in the country that exhibits an improvement in real wages in agriculture. l The Pacific region concentrates about one-third of the total population. This region plus Managua comprises almost two-thirds (62 percent) of the Nicaraguan population. The evolution of poverty in Managua contrasts drastically to that in the Pacific region, which might explain to a certain extent the high prevalence of opposing perceptions regarding changes in poverty over time in Nicaragua. 14 close to 80 percent of births remain non-institutional; and the incidence of diarrhea in children under 6 increased slightly to 29 percent in 1998. Housing and infrastructure conditions are also the worst in the country, with overcrowding, and access to safe water, latrines and electricity showing a deterioration since 1993 (Annex 4, Summary Statistical Appendix). Box 2.3: The Atlantic Region Occupying half of the territory of Nicaragua, the Atlantic Region has the highest extreme poverty rates- 17 percent urban and 41.4 percent rural-but only 10 percent of the country's roads. In many other ways, the Atlantic region stands apart from the rest of Nicaragua. The Atlantic region's political and economic development has differed from the Pacific and Central regions. As part of the United Kingdom, the region was headed by a Miskito King named by the British Crown. With the independence of the Atlantic region and the arrival of fruit exporters from the United States, the Atlantic coast was transfonned into an "enclave" economy, in which Creole migration was encouraged and Miskito peoples relegated to lower status under a slave labor system. The Atlantic coast has undergone changes in political and economic regimes that have fostered social stratification. Both the British and enclave economy isolated the region as a political and economic strategy. The British Monarchy protected a political regime, a territory, and natural resources. The enclave economy protected the economic interests of a private company (closely related to U.S. financial interests), a contract labor force, and political beliefs. After the overthrow of the Miskito-British regime, the Atlantic region was incorporated into the Republic of Nicaragua. However, the mestizo regime of the Pacific and Central regions continued to isolate the Atlantic coast, an isolation that the Atlantic population itself accepted. Development in the Atlantic region rests on many institutional reforms now in process: * The land titling process * Indigenous peoples land demarcation process * Operational execution manual of the Law of Autonomy for the Autonomous Regions of the Atlantic Coast (otherwise, the law will not be effective) * Definition of natural resources and protected areas policy in the region Source: Nicaragua Qualitative Poverty and Exclusion Study (2000). 2.9 All agricultural products in the Atlantic region are basic grains (rice, beans and corn). While production increased, prices remained stagnant. The only export crop grown in the Atlantic region is sugar cane and it represents less than 1 percent of national production. 2.10 Although there are no reliable census data on Nicaragua's indigenous and Creole populations, most are concentrated in the Atlantic Region.'2 Official estimates indicate that out of a total population of about 390,000 in the Atlantic Region, about 50 percent are indigenous (Miskito, Mayagna or Sumu, Garifona and Rama), about 8 percent Creoles, and the remainder mestizos.13 While survey data does not disaggregate between indigenous/Creole and mestizos, qualitative information suggests that indigenous populations are especially disadvantaged (World Bank, 1997, Annex 8). 12 The LSMS does not ask questions on race or ethnic origin, although it does include questions on language spoken. Using language as a proxy (those speaking a language other than Spanish) would suggest that only about one-quarter of the population in the Atlantic Region would be classified as indigenous and Creole, which is more than half of official estimates. 13 World Bank. 1997. Nicaragua: Atlantic Biological Corridor Project, Annex 5. Global Environment Facility. Washington, D.C. 15 2.11 Social outcome indicators and the coverage of basic social services have generally improved. With respect to the International Development Goals, both infant and child mortality rates improved by 20 percent over the period. Maternal mortality shows an increasing trend (with wide year to year variations) over the period (59 percent), although this increase most probably is due to improved reporting. The overall level remains high, and a priority for attention. Coverage of primary schooling increased by 9 percent overall. Gender equality in primary education is not relevant for Nicaragua, since female enrollment rates are already slightly higher than for males. 2.12 In health, significant gains are evident in access to prenatal care (14 percent, Figure 2.2), institutional births (23 percent) and physician attention during childbirth (29 percent, Figure 2.3). In all cases, the gains were driven by improvements in rural areas. For all income groups, rural households are more likely to consult when ill, whereas urban households are slightly less likely to consult. Although the incidence of diarrhea increased for the poor (possibly due to seasonal differences between surveys), when diarrhea occurs, the rural poor are more likely to seek care than in 1993 (Figure 2.4). Figure 2.2 Figure 2.3 Figure 2.4 PFenatd CGm _ 31998 lsittuonai s Brths attenfdlbyDocor ES 199399 19SI1r9--- 100 40 6 20 4f 40r k 0 0 20 Rural Urban Rural Urban Rural Urban Rural 2.13 Gains in primary school coverage were almost exclusively rural, aDd predominantly benefited the poor, with a 26 percent gain in coverage among the extreme poor (as measured by Figure 2.5 Figure 2.6 Children Aged 7-12 not attending school Children aged4-6 attencing school 50.0 70 30 40.0 30.0S 40°g° <0 t e9 _do y ¢2 e 20.0 4'4~~~~~0 ~~~~ ZQ~~~~~S2 L. _ ___ __ 16 the share of 7-12 year olds attending school, Figure 2.5). Among the 4-6 year old group, gains in school attendance are equally dramatic (Figure 2.6). Overall, this group shows a 43 percent increase (from 31 to 45 percent) in school attendance. Again, the greatest gains were in rural areas (from 13 to 34 percent) and among the extreme poor (from 8 to 25 percent). At the secondary school level, there was virtually no change in coverage. Illiteracy declined by 17 percent nationally, and by 24 percent among the rural poor. Mean years of schooling improved for all age and poverty groups except the urban poor, with the greatest gain (26 percent) among the rural extreme poor. Nevertheless, large gaps in access and attainments remain, and future strategies will need to focus more strongly on quality, for which we still lack adequate measures. B. THE POVERTY PROFILE TODAY 2.14 Given these trends, both good and bad, what can we say about the nature of poverty today in Nicaragua? Given its 1999 per capita income of US$430, Nicaragua remains one of the poorest countries in the Latin America and the Caribbean region. As such, it is not surprising to find high poverty rates and weak social indicators reflecting high and extensive poverty levels. Nevertheless, certain aspects stand out. 2.15 Children, especially those under age five, are priority population groups for poverty reduction efforts. 4 Children zero to five are the most likely to be in poverty and extreme poverty as compared to the rest of the population, associated to higher rates of population growth in Nicaragua among poor and extremely poor families. Six out of ten young children (zero to five years old) are in poverty and two out of these six are extremely poor. These youngest children represent 13.7 percent of the Nicaraguan population but they account for 18.2 and 17.1 percent of the extreme and overall poor, respectively. 2.16 Zero to seventeen year-olds are also more likely to be in poverty and extreme poverty than the rest of the population, but less so than zero to five year olds. This older cohort represents 50.1 percent of the Nicaraguan population but they account for 61.1 and 57.3 percent of the extreme and overall poor, respectively. Households headed by females are not poorer on average than those headed by males. 2.17 Nicaragua's fertility rates are twice the Latin American average. As a result, Nicaragua's population is young-50 percent of the population is under 17 years of age. Furthermore, adolescent fertility rates are the highest in Latin America and represent an increasing share of all births. By age 19, half of all women have experienced at least one pregnancy. The poorest households are characterized by more than three times as many children as the richest households, due in part to the failure of poor women to meet their reproductive preferences (Figures 2.7 and 2.8). However, in recent years fertility declined significantly, from 2.9 births per woman in 1993 to 2.5 births in 1998.15 This decline continues the downward trend observed since 1985. The overall dependency ratio also fell modestly over this period, from 1 percent to 0.9 percent. A major concern is the uneven and widening distribution of fertility declines: while overall rural fertility fell from 3.7 to 2.9 births per woman, the decline was only 14 The elderly (61 years and older) are not at a higher risk of being in poverty than all Nicaraguans, as their incidence of extreme and overall poverty is lower than for all Nicaraguans. Likewise, female-headed households are not poorer than those headed by males. 1 These figures are from LSMS93 and LSMS98, and are sharply lower than those reported by ESF93 and DHS98. The reason is that the LSMS calculation is the average number of children born to all women with children, without adjusting for the additional children that young women may bear in the future. The objective of this paragraph is to illustrate trends among poverty groups. 17 from 4.0 to 3.4 births among the poor and from 4.4 to 4.0 births among the extreme poor. Similarly, adolescent fertility declined more slowly than for other age groups. This means that poor households, and especially adolescents in those households, account for a growing share of all births. Figure 2.7 Figure 2.8 Global fertility by quintile Reproductive preferences 7 65 3Qii i ,|n 4t- I etil7 4 2 - . l. children desired desired and I V~~~~~~~ (women 20-24) attained SES quintiles 2.18 Poor women are less likely to be in school, and they are less likely to work outside the home. Differences in fertility by poverty group reflect in part the substantial differences in opportunities available to poor women (Table 2.3). Yet when they do work, they are more likely to leave their younger children in the hands of their elder daughters, thereby further reducing opportunities for those daughters. As a result, unless these patterns change, the existirg disparities in reproductive outcomes will be repeated in coming generations. Table 2.3: Nicaragua: Female Opportunities by SES Quintile (percent) SES Quintiles Indicators Total 1 2 3 4 5 % female school attendance, 15-19 years old 48.8 15.2 28.1 44.4 64.0 76.5 % women of reproductive age currently working 39.3 23.0 35.7 41.6 46.1 51.1 % children under 15 of women who work, cared for by daughters 13.0 28.1 15.5 14.6 7.5 6.5 Source: UNFPA Technical Assistance Team-Mexico. This table is based on data from DHS-98, and uses an unsatisfied basic needs index to group households by SES (socio-economic status) quintiles. 2.19 Nicaragua exhibits a high incidence of domestic violence, which raises broad concerns about women's status and a lack of social cohesion. Almost one-third of Nicaraguan women report having been physically abused, frequently when their children are present and often during pregnancy. The shame of violence prevents most women from seeking help. A micro-study in the city of Leon found that women suffering violence have 4 times the probability of giving birth to low-weight babies.16 When subjected to both physical and sexual violence, their babies are 9 times more likely to die in infancy, and their children are more likely to suffer behavioral and learning problems. Violent households tend to have more children, but since the episodes of violence typically begin during the first two years of marriage, the chain of causality seems to run from violence to having more children, as the result of women's reduced freedom to seek family 16 Reproductive and Child Health Project, UNAN-Le6n. 18 planning and reproductive health services without permission.'7 Adolescents in violent homes have twice the fertility rates of peers in non-violent homes, perhaps a desire to have "something of one's own" in a disempowered environment.18 This pattern tends to lock adolescent mothers into an ongoing dependency with limited opportunities for themselves. In Leon, 33 percent of under-5 deaths, 16 percent of low-weight births and 15 percent of adolescent pregnancies are attributable to living in an environment of domestic violence. Box 2.4: Segregated Gender Roles can Hamper the Attainment of Higher Welfare Levels Quantitative and qualitative evidence from the LSMS and the Poverty and Exclusion Study indicates that segregated gender roles prevail in Nicaragua. Segregation in gender roles -in particular, with women solely responsible for domestic and reproductive work and men understood as natural heads of households with corresponding expectations that they will provide the primary source of household income- have shown to have a significant impact on the acquisition of human capital and abilities of men and women to participate fully in social life. Moreover, when gender roles are too segregated but the opportunities to fill their expectations are not available, family and community are disrupted, with frustration and violence easily erupting. In Nicaragua, the following aspects characterize the segregation of gender roles: . Men are always considered heads of household, even when women are the main breadwinners. Only widows or single mothers living alone consider themselves as such. In consequence, women feel they should request permission to undertake certain activities, i.e. women often do not use available health services unless authorized by their husbands. . Women show a dismaying unconsciousness regarding the links between multiple pregnancies, poverty and their poor health status. To be either pregnant or nursing is very often taken as women's destiny. . Pregnant teens more likely live in poor households. However, more than others, their children will face higher risks of low birth weight, neglect and will be more likely to drop out from schools and to live in poverty. . The risk of maternal mortality increases with the number of pregnancies. Together with infant mortality, maternal mortality accounts for the majority of all premature deaths. . Responsibilities for housework and childcare constrain women's work outside their homes. Among the poor, married women generally exhibit low labor force participation rates. On the other hand, female-headed households are not poorer than conjugal households; overall, they are less poor. Female ownership of agricultural assets, their participation in agricultural activities, and their use of credit are all quite low, but higher in female headed households than in conjugal or single male headed households. Simultaneously, children in female-headed households are not less likely to be in school than children in conjugal households. . Among the poor, boys are expected to work since very early in life, severely hindering their educational attainment and future income. * Substance abuse and alcoholism, particularly among men, are important worries for their families. This behavior promotes violence among youngsters and domestic violence against women and children, and in turn becomes an obstacle to acquisition of assets (such as human capital) and to take advantage of income generating opportunities. 2.20 Malnutrition is widespread and continues to constrain child health, welfare and opportunities. Approximately 20 percent of children under five are chronically malnourished, or stunted. This share rises to 36 percent among the extreme poor. Children never fully recover from the damage wrought by early malnutrition (during the first two years of life) and the damage 17 Nearly one-quarter of women report that they cannot seek health care without permission, usually from their partner (Nicaragua DHS, 1998). '8 Zelaya (1999) and FNUAP-INIM (1999). 19 is cumulative-nearly 50 percent of all extremely poor five-year olds are stunted (Figure 2.9)'19 Micronutrient deficiencies also are a dramatic problem: 67 percent of children 1-5 years old suffered either moderate or mild vitamin A deficiency in 1993 and 55 percent of children aged 12-23 months suffered anemia. For women of childbearing age, 34 percent were also anemic. In 1998, 14 percent of all households still do not consume iodized salt. Figure 2.9 Stunting by age and poverty group 60 50. X 30 81 | ;+~~~~~~~~~~-,..Ext. Poor 40 30 ~~~~~~~~~Poor a20 i . 1 Non Poor 10 0 __ __ __ __1!_ 0-6 6-12 12-18 18-24 24-36 36-48 48-60 Age (months) 2.21 Furthermore, the trends in child malnutrition are worrisome. Child malnutrition measured by stunting (height for age) declined 16 percent nationally between 1993 and 1998 (Table 2.4).2° This trend is driven primarily by declines among the non poor and in rural areas; but it masks increases in urban malnutrition among the poor (a 45 percent increase, from 18 to 26 percent) and the extreme poor (a 30 percent increase, from 28 to 36 percent).21 Malnutrition is mainly caused by food insecurity (mainly related to adequate income to buy or access a basic diet), disease, and inadequate caring practices. In Nicaragua, food security (income) and the incidence of disease have improved since 1993, but household behavior has not changed and remains the critical bottleneck. 2.22 Consequently, little change is observed over the last 30 years in stunting, despite some gains in poverty reduction and the highest rates continue to be concentrated in the north and central parts of Nicaragua-affecting as many as 50 percent of children in extreme poverty. In addition, while the prevalence of underweight children (weight for age) is relatively unchanged, 19 Stunting (height for age) is associated with poor overall economic conditions, chronic or repeated infections, as well as abnormal nutrient intake. Stunting is related to long-term chronic malnutrition, and has also been associated with diminished intellectual capacity and impaired work performance later in life. Wasting or acute malnutrition (weight for height), on the other hand, is an indication of acute food loss due to a state of emergency, environmental disaster, or in situations where the family food supply is limited, making the child too thin for a given height. In consequence, seasonality tends to affect wasting. Underweight (weight for age) is used to look at the combination of stunting 4nd wasting. 20 Nicaragua LSMS 1998 and LSMS 1993. 21 Observed increases in urban malnutrition are most probably related to appalling sanitation conditions in densely populated slums concomitant to poor hygiene practices, mothers having to leave their children in the care of older children, lack of availability of self-produced food items from "patios", etc. In Pacific Urban, where poverty and malnutrition increased between 1993 and 1998, high incidence of malnutrition prevails with high incidence of diarrhea for children under six as compared to Pacific Rural where poverty is higher but these indicators are lower. 20 the prevalence of acutely malnourished children or wasting (weight for height) has increased across all income groups. Increasing trends in acute malnutrition as well as urban stunting for the extremely poor and the poor are cause for concern and a priority for immediate action (Figure 2.9). Table 2.4: Children under Five Suffering Malnutrition, 1998 and 1993 (percent) Extreme Poor Poor Non-Poor Indicator Urban Rural Total Urban Rural Total Urban Rural Total Total 1998 Underweight (WFA) 17.28 19.72 19.17 13.52 14.46 14.13 6.58 5.64 6.29 10.95 Chronic (HFA) 36.46 35.56 35.77 26.18 27.61 27.09 8.56 11.51 9.48 19.94 Acute(WFH) 2.15 4.79 4.19 3.54 4.37 4.07 2.04 3.34 2.45 3.32 1993 Underweight (WFA) 18.6 18.6 - 10.4 15.4 - 5.7 8.8 - 11.9 Chronic (HFA) 28.0 41.1 - 18.1 31.0 - 12.7 19.4 - 23.7 Acute (WFH) 4.6 1.8 - 1.7 2.3 - 1.1 1.0 - 1.9 Total (any 34.5 44.4 - 22.0 35.1 - 14.3 22.1 - 27.2 measure) Source: Nicaragua LSMS 1998 and World Bank (1995) for 1993 data. 2.23 Nicaragua has high levels of infant mortality and maternal mortality, a high prevalence of infectious and parasitic diseases, and pervasive malnutrition. Although infant mortality has declined significantly since 1985 (from 83/1,0001.b. to 40/1,000 L.b. in 1998) it still remains above the Central American average (Table 2.5). Infant and maternal mortality account for the majority of all premature deaths and are due to infectious diseases, malnutrition and reproduction. Acute respiratory infections are the principal cause of illness and the second cause of death among children under five, while diarrheal disease is another important cause of child deaths. Malnutrition underlies over half of under five mortality and 20 percent of maternal mortality. Table 2.5: International Comparisons for Health and Education PRSP Countries in Latin America Other Central America NIC HON BOL GUY All CR ES GU PA All CA GNP per capita (Atlas method, 1998), US$ 410 730 1,000 770 762 2,780 1,850 1,640 3,080 1,567 GDP per capita (PPP, 1998), US$ 1,960 2,300 2,910 3,210 2,502 6,650 2,940 4,160 7,220 3,785 Health Indicators Life expectancy at birth (years) 68 69 62 64 66 77 69 64 74 68 Infant mortality (/1,000 live births) 40 36 67 58 51 12 32 37 21 33 Child malnutrition (% children < 5) 20 18 9 18 13 5 11 27 6 17 Access to safe water (% of population) 84 77 60 83 67 100 53 67 84 70 Education Indicators Illiteracy (% of population> 15) 19a 29 17 2 25 5 23 33 9 26 Gross primary enrollment 107 111 91 95 100 103 93 88 104 98 Gross secondary enrollment 57 32 37 75 39 47 34 26 68 37 Source: The World Bank. aThe illiteracy rate for Nicaragua refers to share of population over 10 years of age. 2.24 Access to health care is characterized by large and persistent differences between the poor and non-poor. Extremely poor children report illness with 50 percent greater frequency than non-poor children, and when sick, the non-poor consult 50 percent more. Household practices also contribute to the impact of the disease. For example, DHS98 reports that among mothers of children with diarrhea, those who have the least schooling, are the youngest and live in rural areas (good proxies for poverty), and are the most likely to withhold liquids and food from the affected child. One third of extremely poor women receive no prenatal 21 care, one half deliver their babies without benefit of institutional facilities, and are only half as likely as non-poor women to be attended by doctors when giving birth. To obtain health care, the extreme poor must travel three times the distance, and spend three times as much to reach health facilities, as their non-poor counterparts. Rural-urban differences are also important. Rural women receive less prenatal care and are three times more likely to give birth without institutional support than urban women. Box 2.5: Rural Boys' Educational Attainment is Severely Hindered by Child Labor Child labor is pervasive among rural boys in Nicaragua, it severely hinders educational attainment and it is strongly associated with long-term poverty (as captured by non-monetary measures). Even though the overall incidence of child labor (10 to 14 age group) in Nicaragua is low at 11 percent compared to other countries in the region, child labor is pervasive in rural areas. In Nicaragua, close to 75,000 children work, the vast majority of these children are boys (17 percent compared to 5 percent of girls), and tend to live in rural areas (16 percent versus 6 percent for their urban counterparts). Almost one-third of rural boys are engaged in child labor, defined as all work done in economic activities including unpaid family labor (both on the family farm/enterprise as well as outside work), and are therefore at high risk. Child labor is highest in farm households, with almost half of all boys in mid-size farms of five to 20 manzanas working. There is a stark difference in rural child labor between farm and non-farm households, the latter with less than one in ten working. The intensity of children's work is also high- working boys average 31 hours per week and therefore not surprisingly child labor severely retards educational attainment. Given that child labor is particularly high in mid-size farms, poverty targeted cash grant programs for education cannot both increase enrollment and reduce child labor. Thus, the opportunity cost of a large number of working children is likely to be high. This suggests that poverty targeted cash transfers (such as Mexico's Progresa or Brazil's Bolsa Escola) may be effective in increasing enrollment but not in lowering child labor. 2.25 While Nicaragua exhibits a low illiteracy rate by low-income country standards, the population averages only a few years of schooling and there are great disparities in education attainment levels. Illiteracy is 19 percent among the total population over 10 years old, but 41 percent among the rural extreme poor. Also, the rural poor over 10 years have only completed 2 years of school, compared to a national average of 4.9 years. Low years of schooling reflect deep seated problems in access and quality of education services. Low technical quality of teachers, poor conditions in over half of all primary classrooms and a chronic shortage of teaching materials partly explain the high rates of repetition and desertion before 4th grade, especially in rural areas, where these problems are most acute. Poor teaching quality stems from: (i) poor quality of pre-service teacher training; (ii) inadequate in-service training; (iii) approximately one-third of primary level teachers are not accredited; and (iv) low rates of retention of good teachers due to very low salaries Y Poverty also contributes to low schooling outcomes. More than half of all children who abandon school do so for economic reasons (Box 2.3). This combination of factors has the consequence that in rural areas, one-third of primary school aged children in extreme poverty do not attend school, compared to only 9 percent of their non-poor counterparts. 2.26 In urban areas, the supply of schooling does not appear to be the principal barrier to school attendance, as nearly 84 percent of extremely poor children not attending school cited 22 World Bank (1999). 22 economic problems (demand side) as their reason for not attending. In contrast, while in rural areas demand factors are still the most important (48 percent of extremely poor children cite economic problems as their reason for not attending and another 6 percent declare that rural activities prevent them from attending), supply factors are also important (distance, insufficient teachers, and level not offered, account for roughly 30 percent of non attendance). Lack of interest, as a reason for not attending school, seems to be a predominantly urban phenomenon across all poverty groups. 2.27 Access to other infrastructure services by the poor only improved modestly, if at all. Investments in infrastructure mainly appear to have kept pace with population growth, resulting in modest gains in coverage of the population with access to these services. According to the LSMS results, access to water and sanitation services improved by 4 and 1 percentage points at the national level (for poor and extreme poor), and 5 and 5 percentage points for the rural poor, while sanitation services improved by 1 percentage point in both cases. Access to electricity even appears to have deteriorated somewhat, by 1 and 2 percentage points. All these changes, however, are not statistically significant. 2.28 One sector where there have been major gains in infrastructure development is transport, which exhibits a significant expansion in the road network. The total road network expanded by 18 percent between 1991 and 1998, with the paved network expanding from 1,565 km to 1,717 km and the unpaved network expanding from 13,010 km to 15,429 km. It is also important to mention, however, that a significant share of the road network (74 percent) is classified as in bad condition. This is far higher than the share observed in other Latin American countries (which averages 40 percent) and may be reflecting too much investment in new roads and insufficient maintenance of existing roads. 2.29 The poor perceive minimal improvement since 1993 in their access to main roads. Overall, 17 percent of LSMS households reported that their access has improved while 11 percent reported a worsening in access; the remainder either reported unchanged conditions or that they had recently migrated. This difference disguises substantial differences by income groups. At one extreme, 18 percent of the non-poor in urban areas report improved access versus only 8 percent who report a worsening. At the other extreme, 13 percent of the extreme poor in rural areas report improved access versus 11 percent who report a worsening. Adoquinado has proven to be a cost-effective investment to improve rural accessibility and help reduce poverty in rural areas, by lowering costs, providing jobs, and assigning funds to local technology and materials, with significant spillovers such as improving access to health and educational facilities, and promoting local development and institution building with projects that can be completed in a short time, with little or no environmental impact (Box 2.4). 2.30 The costs of reducing poverty in Nicaragua are considerable. They are even more so as compared to the size of Nicaragua's GDP and total public spending. Under perfect targeting the total poverty gap is US$353 million annually, which represents almost one-fifth of GDP and close to one-half of total public spending. For extreme poverty, the total dollar amount is US$49 million annually, which is a substantially smaller amount than total foreign aid coming into Nicaragua. However, this figure represents perfect transfers to the extreme poor with no administrative costs. 23 Box 2.6: Adoquin Roads can be Targeted to the Rural Poor with Significant Spill-overs Adoquinado was used immediatey after Hurricane Mitch to restore 60 km of secondary and tertiary roads in a cost-effective manner with a US$4.6 million pilot project under the Hurricane Emergency Road Stabilization program using funds from an ongoing World Bank transport project (3085-NI). Adoquinado was targeted the rural poor and proved to be more than a cost-effective solution to improve rural accessibility and reduce vehicle operating costs with little or no environmental impact. A beneficiary assessment proved the significant spill-over benefits of adoquin roads, such as increasing utilization of health and educational facilities, improving health status (by eliminating heavy dust clouds in the dry season and slippery/muddy surfaces in the rainy season) and promoting job creation. The adoquinado pilot also promoted local development by using local materials, labor and technology, and improved community and local involvement in planning and decision-making in subproject selection and design, potentially increasing sustainability in the long-run. Adoquines provided these benefits at much less than half the cost of asphalt paving; a typical 6 meter wide adoquin road costs US$72,000 per km (or even US$50,000 per km for a four meter) compared to US$250,000 per km of a regular asphalt road. Adoquinado increases the possibilities for providing all-weather roads for smaller, poorer population centers that would not normally have enough vehicular traffic to qualify for the more costly asphalt road. The most important economic advantages of adoquines are: (i) costs are lower than asphalt; (ii) roads require little maintenance; (iii) no major equipment is required to either build or maintain them; (iv) the adoquin roads can resist floods better than asphalt; (v) technology is simple and the skills developed can be applied to house building and/or community projects; (vi) supervision costs are about one-half of those for an asphalt or gravel paved road; and (vii) adoquin roads are built faster than asphalt. Children walking to school Adoquinado Shell Palacagtlina - Pueblo Nuevo Desvio Le6n Viej'o - Puerto Le6n Viejo Beneficiary Assessment Responses: Guanacaste, Tepeyac Adoquin Road - I had to take off my shoes to walk not to damage them Children - Now we can walk safely to school, and it takes less time now - I used to get sick, now there is no dust - I do not need to be washing clothes and cleaning shoes all day Women - Now I can save on soap, clothes, shoes, medicines - My husband gets home earlier - I can ride my bike to work now, it takes me less time now Men - I have worked on the road, it was good to have that job - I grow vegetables that don't get ruined by the dust and I can sell them better Micro-enterprises - It was good to have the experience, now we can do other contracts - We were able to complete this project without proper education/experience Nuns running - Pregnant women hesitated to come due to bad road conditions health center - People used to get sick more with lung diseases, fungus, ear infections - This type of contract is simple, straight forward, and quick Contractors - We have employed more than 65 percent of the people here - We have em lo ed women for cookin and washin 24 C. THE QUALITATIVE " FEEL" OF POVERTY 2.31 Statistical discussions of poverty all too often lose the "feel" of what it is like to be poor, the risks and the pains suffered by the poor, and the limitations and hopelessness of their situation. The Nicaragua Qualitative Poverty and Exclusion Study (NQPEY3 findings show that poverty as defined by the poor goes beyond low levels of income and low food intake and calories. People have a broad and integral view of poverty. Poverty is a social phenomenon that is the result of economic, social, political and cultural interactions. To be poor means to be affected by disparities in opportunities. It means to be excluded from benefits of society. Moreover it means to feel excluded and to live in high levels of ill-being (es estar mal). For rural and urban people, to live in a state of well-being-or not being poor-is to be safe. Safety embodies an integral perspective. It means to have a regular income to feed the family, to be protected by the system and to have access to material and social assets that make life comfortable (Table 2.6). Vulnerability, risks, lack of safety, insecurity and exclusion are all concepts closely related to people's perceptions of poverty. Table 2.6: Well-being in an Asset-based Framework as Defined by Poor People Natural Capital Human Capital Financial Capital Material Capital Social Capital * Access to * No food deficit. * Capital and * Access to * Domestic good land for a Access to good financial equipment, and agriculture and and private clinics solvency. machinery and other international animal husbandry. and hospitals * Access to means of production. network for * Large (urban). financial markets. . Access to financial landholdings. * Good education * Access to production inputs. support. centers. well-paid jobs. * Access to * Political * Children attend * Access to drinking and and social school and have no social security and irrigation water- connections. need to work. pension plans. supply systems. * Access to a Access to * Diversified and sewage, sanitation, paid labor force. specialized farm. and garbage . Strong and well- collection. built houses. * Access to paid * Adequate technical assistance. furniture and * Favorable appliances. dependency ratio. . Access to means of transportation. Source: Nicaragua Qualitative Poverty and Exclusion Study, 2000 2.32 People associate poverty with assets, with productivity levels, with the degree of vulnerability, and with feeling trapped in a circle of poverty. Thus, an asset-based framework for 23 The Nicaragua Qualitative Poverty and Exclusion Study 2000 (NQPE), utilized four methodological instruments of qualitative research (community's physical map, focal groups, case studies and semi- structured interviews). It builds on an innovative and cross-disciplinary approach used in Poverty in Rural India (Kozel and Parker, 1999) to make available qualitative findings to enrich quantitative methods. The NQPE used four qualitative exercises: wealth ranking, social mapping, trend analysis, and women's roles/gender issues. Topics included are: well-being, risk and vulnerability, social and economic mobility, social exclusion, power, coping strategies, gender, social capital, technical assistance, migration, basic services, access to physical assets, access to credit, access to human capital and access to employment. The NQPE visited 22 communities in three regions (Pacific, Central and Atlantic, in urban and rural areas), spent five days in each community and completed four qualitative exercises with focal groups, three case studies and six semi-structured interviews per community. 25 poverty, as defined by poor people, assumes that households have the means to generate a stable income to ensure consumption and to cope with shocks that could lead to lower quality and quantity of assets. In rural areas, this means that people associate land, other production inputs, favorable dependency ratios, information and knowledge to increases in productivity. In urban areas, it means to have employment or an enterprise in order to be in a state of well-being. 2.33 In both urban and rural areas, people strongly correlate low human capital development with high levels of social vulnerability. The poor are aware of important gaps in health services between the poor and the non-poor. They are aware that such gaps have an incidence on health outcomes such as mortality and birth rates. Awareness also exists about the strong relationship between low education and high levels of poverty, while at the same time people correlate disparities in educational services with lower quality of education for the poor. Malnutrition, small landholdings and limited access to material capital along with income inequality reflect for the poor disparities in opportunities. 2.34 Improvements in the poverty indicators are often not reflected in public perceptions. In spite of the gains made in reducing poverty during 1993-98, the NQPE finds that the poor associate the decade of the 1990s with a decline in their well being. What can explain this finding? * First, the task of rebuilding Nicaraguan families and society, damaged during a decade of civil war, insecurity and violence, is just beginning. The prevalence and persistence of domestic and civil violence hint at the extraordinary social stresses that continue to affect Nicaraguan families. Keeping in mind that incomes in Nicaragua fell by over 50 percent from 1978 to the early 1990s, a 7 percent income improvement (as happened in 1993-98) is likely to pass unperceived by many segments of the population. * Second, being poor means living at risk. The risks faced by Nicaragua's poor may have increased during the 1990s. The replacement of a socialist state by a private market economy may have opened greater prospects for growth and economic opportunity, but it has left the poorest with virtually no formal safety nets to protect them. After the extensive social and economic upheavals of recent history, Nicaragua's informal safety nets may also be fractured. Moreover, the coping strategies employed by the poor to confront risks and crises-risk-averse behavior, such as planting traditional subsistence crops instead of introducing unfamiliar higher-yielding crops, avoiding credit for fear of risking collateral, removing children from school, having many children to smooth incomes in later years-are precisely the sorts of things that tend to perpetuate their poverty into the coming generations, by reducing the participation of the poor in insurance markets, and by reducing the human capital accumulation of family members. * Third, the poor see improvements in infrastructure, but they do not perceive that their access to the services of that infrastructure has improved. The poor report that the main reasons for not using social services are cost and distance. The costs of social services have risen fastest for the extreme poor, so that even when they do gain access, they face costs that may be becoming prohibitive. While infrastructure construction should have improved physical access, the impact of these investments has been less than anticipated. In many cases, rehabilitated facilities lack key elements for their operation, e.g., teachers, trained health staff, textbooks or medicines. 26 2.35 These findings identify the need to place greater emphasis on mitigating the risks and vulnerabilities faced by the poor, while ensuring that program delivery mechanisms reduce rather than exacerbate the social exclusion of the poor. Vulnerabilities and Coping Strategies, as Defined by the Poor 2.36 In times of crisis, people reallocate their assets to respond to heightened risk. In a state of extreme poverty, people reallocate their minimum assets to survive. People in Nicaragua are aware that their low levels of assets and low productivity place them in a state of ill-being. Under these circumstances, the first priority for Nicaraguans is "to eat." "How many times can we eat today?" and "Must someone in the family go through a hungry day in order to feed the children?", become daily concerns. 2.37 In all urban and rural sites, eating is mentioned as the first priority (Table 2.7). Most rural people mention access to land as the second priority while few name housing as a second priority. In urban areas, employment or secure sources of income appears as the second priority. In urban and rural areas, health is the third priority. Table 2.7: Priority Needs as Defined by the Poor Priority Rural Urban I IT To eat To eat 2nd Land and housing Employment and housing 3rd Health Health Source: Nicaragua Qualitative Poverty and Exclusion Study, 2000. 2.38 In all sites, people perceive the most vulnerable (extreme poor) as those who: * Have great difficulties in accessing food, * Lack basic financial liquidity to afford minimum food intake, * Have houses built with waste or scavenged materials, and * Live in overcrowded conditions. 2.39 The most vulnerable (extreme poor) have to: * Go through hungry periods to feed the children; * Reduce their food intake up to two meals per day. Limit their meals to one or two staples (rice and beans, or beans and corn tortilla, or rice and corn tortilla)24 * Exchange food for food. For instance, if they have extra sugar they exchange it for beans, rice or corn tortillas. This is a common practice among neighbors in rural areas; * Borrow money to buy food. Most borrow money from private money lenders. Interest rates vary across communities; and * Ask for store credit or borrow food. People ask for credit at local grocery stores such as the pulperia (small local convenience store) and Tienda Campesina (cooperatives that fulfill the function of a hardware store, seed distributor, market, credit provider, and a department store). This latter practice prevails in most rural areas. 24 In the Nicaraguan diet there is an ingredient called bastimento. This is a term that refers to a staple that is always present in a meal. For instance, corn tortillas, plantains or corn. Depending on the region, they always eat one of these to which they add rice and beans, and other ingredients. When describing their daily diet most peasants take bastimento as a given and mention other ingredients as additions to daily food intake. 27 2.40 Other coping strategies followed by poor households: Within the legal framework: * Communal support. In urban and rural sites, poor people support each other by taking care of children and family members, taking care of animals and agricultural plots, and lending money and exchanging food for cash or labor. * Selling assets. Animals and grains are in-kind forms of savings, which are sold during critical times. Selling lands and houses is the ultimate coping strategy for rural and urban households. * Labor. Poor households use women's and children's labor as a source of family labor and additional cash income. Women work in agriculture as family labor and wage earners, in commerce selling food and handicrafts, and in services as domestic employees. Children work in agriculture mostly as family labor and in the informal sector mostly in urban areas for cash income. Miskitos have survived by alternating subsistence activities with wage labor, often in foreign enterprises. * Migration as a source of remittances. Migration is an important survival strategy for poor rural and urban households. Domestic migration is to Managua, urban cities or to rural areas for seasonal harvesting. International migration goes most often to Costa Rica, Honduras and El Salvador; mostly following harvest seasons. In the Atlantic, people prefer to migrate to the Caribbean islands and the United States. Poor people perceive migration as both positive and negative; it increases household earnings but it separates families. * Unsustainable environmental practices: Most rural and semi-urban communities cut trees to use and sell as firewood. Outside the legalframework.- - Drugs. Selling drugs is an easy way to rapidly increase earnings. This "strategy" prevails in urban areas of Managua and semi-urban areas of Esteli. Drug problems also exist in border towns and in most cities of the Atlantic region. * Joining Gangs. Lack of opportunities drives young people to join gangs as a way of obtaining cash income. * Illegal Use of Basic Services. In poor urban and some rural sites, people make illicit connections to the electricity grid (El Calvario, Bluefields and Managua) and water system (El Guanacaste). * Prostitution. Female prostitution was mentioned as a survival strategy mostly in urban cities. Male, homosexual and child prostitution was not mentioned in any site. 2.41 The extreme poor in Nicaragua are at a severe state of deprivation and are excluded from most development programs. Crises affect their daily means of livelihood. Thus, concerted and effective social protection mechanisms, targeted to the extreme poor and the most vulnerable among the poor can greatly aid in minimizing transient and long-term poverty effects of crisis. Poor households in Nicaragua have an impressive ability to adapt their coping strategies or rebusque to crisis situations. However, poor people's coping strategies are ineffective in breaking the cycle of poverty because of the limited assets they own and can use in times of distress. Sending children to work may increase immediate household consumption but it increases vulnerability over the long run. Risk management strategies by the public sector are needed to complement limited private strategies and assets. In Nicaragua, poverty is exacerbated by the inability of state institutions to aid households in coping with economic fluctuations, natural disasters and legal and civil insecurity. CHAPTER III. PUBLIC EXPENDITURES ON SOCIAL SERVICES 3.1 Implementation of the reform program in the early 1990s, and the onset of peace permitted a major restructuring of public expenditures. The Government was able to reduce the size of its armed forces and to substantially reorient public spending. Defense spending dropped from 40 percent of the budget (19 percent of GDP) in 1990, to 6 percent in 1999 (Table 3.1). This "peace dividend" was mainly spent on: first, nornalizing financial relations with international creditors (raising debt service from about 1 percent of GDP in the mid-1980s to 8-9 percent in the late 1990s and, second, restoring macroeconomic stability by reducing the fiscal deficit. This was done largely through reducing total public expenditures from an average of 50 percent of GDP in the mid-1980s to 33 percent in the late 1990s. After these two major spending items are accounted for, however, little was left over for the public sector to devote toward restoring infrastructure or expanding social public expenditures. As a consequence, social spending remained constant at about 9-10 percent of GDP during the 1990s, while infrastructure *spending expanded from a low of 5 percent to about 8 percent. Table 3.1: Nicaragua: Sector Distribution of Central Government Spending 1970-75 1985 1990 1992 1994 1995 1996 1997 1998 1999 2000w As a share of total (°/) Social services 35.1 23.8 31.4 29.5 34.2 33.8 32.6 31.5 28.8 34.9 42.0 o/w education 16.2 10.4 14.6 13.3 13.7 12.7 12.8 14.7 14.2 14.8 17.4 health 10.8 8.8 14.3 16.0 13.6 13.4 12.8 11.7 10.8 12.5 15.9 Infrastructure & prod. 27.0 17.5 10.2 20.2 14.0 18.0 17.6 20.6 23.4 -- 23.8 Defense and police 10.8 29.4 40.0 15.3 11.0 9.7 10.3 8.4 7.9 6.2 6.5 Debt service 4.7 4.9 0.0 10.3 27.7 25.1 21.2 26.5 25.0 17.1 14.2 Rest 22.3 24.4 18.4 24.8 13.1 13.2 18.4 12.3 14.4 -- 13.5 Total 100 100 100 100 100 100 100 100 100 100 100 As a share of GDP f°/) Social services 5.2 13.1 14.7 8.4 11.8 11.4 10.9 10.8 9.8 14.4 15.2 o/w education 2.4 5.5 7.1 3.7 4.7 4.3 4.3 5.0 4.8 6.1 6.3 health 1.6 4.9 6.6 4.5 4.7 4.5 4.3 4.0 3.7 5.1 5.7 Infrastructure & prod. 4.0 9.8 4.7 5.6 4.8 6.1 5.9 7.0 8.0 -- 8.6 Defense and police 1.6 15.9 18.9 4.2 3.8 3.3 3.5 2.9 2.7 2.6 2.3 Debt service 0.7 2.7 0.0 2.8 9.6 8.6 7.1 9.0 8.5 7.0 5.2 Rest 3.3 13.1 8.5 7.0 4.5 4.5 6.2 4.4 4.9 -- 4.9 Total 14.8 54.7 46.8 28.1 34.5 33.8 33.4 34.1 34.0 41.2 36.2 Memo Item: In constant (1980) US dollars GDP per capita 1,052 670 485 457 443 448 455 464 469 488P 503P Source: Figures for 1970-1990 are from World Bank (1992). Figures for 1992 are from World Bank (1997). Figures for 1994 onwards are based on data provided by MHCP. Figures for 2000 refer to the fiscal budget. 3.2 The relatively small resource base results in low aggregate levels of health and education spending. Although Nicaragua devotes a higher share of GDP to health and education spending than other countries in the region, this translates into one of the lowest absolute levels of per capita and per student spending (Table 3.2).1 Total per capita recurrent public health ' Health data includes recurrent and capital spending. Even allowing for an estimated 50 percent underreporting of GDP would still leave Nicaragua's social spending as a share of GDP near the regional average. 29 spending averaged less than US$16 per year during the 1990s. Including all public private and donor expenditures only increases per capita spending to approximately US$37.2 Table 3.2: Nicaragua: Spending in Health and Education Health Education Public and Private Sources Public Sources Total as Primary as Countries US Dollars Per % of US Dollars Per Primary % of GNP Capita GDP Student Per Capita Costa Rica 215 8.5 438 13.4 El Salvador 86 5.9 140 7.0 Guatemala 41 3.2 88 6.1 Honduras 33 5.6 209 9.0 Nicaragua 37 8.6 58 11.4 Panama 199 6.7 374 Low Income Countries 41 4.7 Source: World Bank (1999b) and (1999a). For education at 1996 US$ (except Nicaragua), IDB (1996). 3.3 Improving trends in basic service coverage appear to have occurred in large part due to donor efforts, compensating declining domestic financing. Government policy documents throughout the 1990s placed priority on improving the delivery of basic services for the poor. Between 1993 and 1998, however, per capita spending of domestic resources declined sharply in real terms in both health and education.3 Due largely to donor support, a greater share of the total subsidy (Government plus donors) in these sectors was targeted to the poor. 3.4 Under HIPC, the total resource envelope for public expenditures is projected to expand only modestly in the short and medium term. Therefore, social spending cannot expect to receive major increases in resources, which means that improving outcomes for the poorest initially will have to rely largely on targeting and efficiency gains. The direct benefits of HIPC mean that the Government would not have to undertake budgetary reductions or raise new taxes to accommodate debt servicing. It is unlikely that public spending in total GDP will rise in the future, or that the share devoted to social sectors within public spending could be expected to increase greatly. Likewise, donor financing will at best be constant. Thus, the rate of growth of social spending is likely to equal the growth rate of GDP. Therefore, the strategy for the social sectors has to focus on improving the efficiency in the use of limited resources, better targeting of the poorest, better selectivity between programs, and the introduction of more cost- effective delivery mechanisms. 2 Estimates for the mid-1990s placed this figure at US$30-32, Arredondo and Chamorro (1996). 3 Despite steady nominal growth in recurrent finance, real per capita recurrent financing of MOH from domestic sources declined 20.7 percent between 1993 and 1998 (from US$17.97 to US$14.25). This also reflects a decline in MOH's share of the total budget from 25.2 percent in 1995 to 21.7 percent in 1998. In education, real per student recurrent financing of primary and secondary education from national sources declined between 1994 and 1997 (from US$57.64 to US$41.19, a drop of 28.5 percent) and recovered partially in 1998 to US$45.71 (a net drop over the period of 20.7 percent), due to a 30 percent increase in teacher salaries. Over the same period, however, total recurrent finance per primary student increased by 50 percent, due to substantial donor financing of primary education. 30 A. TOTAL SOCIAL SPENDING BY PROGRAMS 3.5 Total social spending in Nicaragua at 54 percent of overall public spending is a significant amount that can be allocated more effectively to reach the poorest and their needs (Table 3.3). Lessons learned from a study on total (comprehensive) social spending and programs in seven sectors ' for the year 2000 indicate that there is considerable room to increase efficiency of resource use and quality of services by moving to a results-oriented system.5 Significant improvements in efficiency of expenditures could be achieved in central administration, which account for a significant share, 13 and 9 percent, in health and education. These administrative expenses are considerable given that they do not account for project administration which is charged to each project separately and accounted under programs. 3.6 In all seven social spending programs, the allocation between capital and current expenditures shows relatively abundant investment finance associated with inadequate finance for the operations and maintenance that reduces the impact of investment spending. Prioritizing expenditures within programs needs to be driven by three over-arching principles: first, focus on the highest impact and most cost-effective programs; second, finance basic services over secondary and tertiary; and, third, target the poorest and their needs. Table 3.3: Nicaragua: Total Social Spending by Programs, 2000 Total Central Government Current US$ Million: 894.7 Percent of Total Official Comprehensive Official Comprehensive Social Spending 385.7 476.7 43.1 54.3 Education 163.8 173.3a 42.5 35.6 Health 138.7 138.7 36.0 28.5 Water and Sanitation 9.9 56.3' 2.6 11.6 Rural Development 44.5c 0.0 9.2 Local Development 31.4 31 d 8.1 6.5 Social Assistance 26.5 26.5e 6.9 5.5 Housing 10.0 10.0 2.6 2.1 Other 5.5 5.5 1.4 1.1 Source: GON. Ministeno de Hacienda y Credito Publico. Proyecto de Presupuesio General de la Republica, 2000. Exchange rate= C$12.68/1 USS.(a) Includes INATEC. (b) Includes ENACAL's capital expenditures. (c) Includes IDR and selected MAGFOR programs. (d) Includes INIFOM, FISE (for local development) and municipalities' contributions. (e) Includes MIFAM, SAS and FISE. Education Spending is Heavily Skewed toward the Tertiary Level 3.7 In education there is a considerable imbalance between the basic and higher levels of education, particularly tertiary. Priorities for Nicaragua's education are in the basic levels. The opportunity cost of allocating a substantially lower proportion of total education resources to the basic versus higher levels is evident when three-quarters of the poorest children 4 to 6 and one-third of poorest children 7 to 12 do not attend school. Pre-school education has been found to greatly improve school achievement and retention at the primary level. In Nicaragua, however, pre-school spending is only one percent of total education spending, versus 24 percent allocated to universities (Table 3.4). Given that less than 6 percent of university students are among the poorest (LSMS, 1998), a 24 percentage share of total education spending is certainly 4'The official defnition of social spending is 43 percent of overall public spending, and includes education, health, housing and other [the latter includes mainly Social Action Secretariat (SAS), Ministry of Labor (MITRAB), Ministry of the Family (MIFAM), FISE, Social Security Institute (INSS), Municipal Development Institute (INIFOM), Municipalities, Water and Sanitation Public Company (ENACAL)]. 5 Budinich (2000). 31 unjustifiable on equity grounds. Expenditures per student indicate that each university student would finance 26 primary level students in a year (Table 3.4). Spending of US$5 per pre-school student can have little impact. The priorities for Nicaragua's education sector are to re-allocate enough funding through efficiency and targeting gains toward the basic levels so that: first, all children attend one year of pre-school education; second, all children attend the first cycle of primary (up to third grade); and, third, secondary education is ready to receive the poor that are reaching this level by providing them with merit scholarships. Table 3.4: Nicaragua: Education Spending by Programs, 2000 US$million ToIl Current per Toa Student (b) Total Current (%) (US$/year) Total Education 163.8 110.6 100 MECD Central Administration 15.4 7.3 9 Pre-school 1.0 1.0 1 5.3 Primary (a) 86.4 51.7 53 61.4 Secondary 8.9 6.7 5 21.7 Adult Education 0.4 0.4 0 Special Education 0.6 0.6 0 Teacher's Training 1.8 1.7 1 Nutritional Support 3.9 1.0 2 Other expenditures 0.3 0 Universities and Higher Education 38.7 37.1 24 1,611.7 Technical Training 3.9 0.8 2 Sports 2.4 2.4 1 Source: GON. Ministerio de Hacienda y Credito Publico. Proyecto de Presupuesto General de la Republica, 2000. Exchange rate= CS12.68/IUS$. (a) Includes FSS. (b) Pre-school, primary, secondary and universities/higher education total enrollment are 188, 841.5, 308.4 and 24 in thousands, respectively. 3.8 Although the introduction of school fees at the primary level under the reforn program raised needed additional recurrent finance, they now represent approximately 15 percent of the total private cost of schooling. School uniforms and materials account for most of the balance. Since the elimination of fees would result in an under-funded education system, the best way to address this problem might be through demand subsidies for the extremely poor. In other words, maintain existing cost recovery mechanisms to assure an improving quality in the supply of education services and complement this with a demand subsidy that compensates the targeted households for the additional costs of sending children to school. However, those mechanisms still have to prove their effectiveness. Health Spending is Highly Inefficient, Particularly at the Primary Level 3.9 In health the relatively equal spending allocation between primary and secondary levels conceals considerable inefficiencies in both levels. This is mainly related to the lack of a results-oriented system that could prioritize programs with the highest impact, basic services, and targeting of the poorest and their needs. The most frequent health problems in Nicaragua continue to be diarrheic, respiratory and parasitic diseases, complications associated with reproduction and malnutrition. Clearly, primary health care is the highest sectoral priority. However, external financing allocates 40 percent to the secondary level. In the long run, sustainability issues of primary health care financing would also need to be addressed, as external financing is twice as large as domestic resources at this level alone. Results-oriented priorities for Nicaragua's health system are: first, to focus on the most frequent health problems; second, to attain total coverage of priority programs such as reproductive health care and integrated child health services; and, third, the use of broad health campaigns that emphasize the relationship 32 between household behavior and their own health status, in such areas as hygiene and water use practices, household sanitation and waste disposal, and food preparation and handling practices. (Box 3.1).6 Box 3.1: Determinants of Infant and Child Mortality Estimates from the 1998 Nicaraguan Demographic Health Survey (DHS) indicate that from 1993-1998, there were approximately 40 infant deaths per one thousand births, and 50 child deaths per one thousand births, at the national level. Infant and child mortality rates were 40 and 50 in urban areas; and 51 and 63 in rural areas. Analyzing the determinants of infant and child mortality implies selecting a set of variables (z) to estimate the likelihood of survival (event) of a cross section of children born in the last five years (taken from the sample of women surveyed in the 1998 Nicaraguan DHS). Prob (event) = 1/(1 + eZ) Where z = Bo + B1X1 + B2X2 + ...BpXp The variables included in the model are a combination of individual (human capital endowments), household (assets and other proxies of income), and community level (access to adequate health and sanitation facilities) factors that are hypothesized to have a significant impact on the probabilities of survival for infants and children under 5. Results from the analysis of the determinants of infant and child mortality in Nicaragua, in both urban and rural areas, suggest that factors affecting the likelihood of infant and child survival are: (i) maternal and child health care practices are the most significant positive factor: including prenatal and natal care, breastfeeding, and knowledge of treatment of illnesses, particularly diarrhea; (ii) maternal education is the second most significant positive factor, this obvious factor is underscored in Nicaragua since extremely poor women older than ten years old have a low 2.3 average years of education; (iii) limited birth spacing and early onset of fertility have a significant and negative impact on child survival (as reflected in the mother's age or mother's age at birth) in urban, but particularly in rural areas; and (iv) access to safe drinking water is an important factor, but not one of the most significant variables, a result which is closely related to household practices, 84 percent of households have access to safe water in Nicaragua (defined as pipes inside/outside, public source and well) but only one-third have access to piped water which means that the rest have to get water from sources outside of the household and store it before consumption; therefore a policy priority is to promote household practices toward adequately treating water before human consumption so that water is potable (such as boiling and adding chlorine, which have a positive and significant impact on child survival). 3.10 Achieving better quality and lower cost health care for poor families goes beyond finance constraints alone. Nicaragua's health outcomes also reflect systemic problems that prevent effective public sector delivery of health services outside the main urban areas7 MINSA health system resources (physical infrastructure, personnel and recurrent expenditures) are biased in favor of the Pacific region relative to its population, and resource distribution continues to favor hospitals over the primary care level. In addition, simple measures suggest that system efficiency has not improved over the 1990s. Although real spending on personnel increased by 50 percent between 1993 and 1998 and the number of primary health care facilities increased by 20 percent, service production as measured by consultations at the primary level only increased 13 percent. 6 Domestic animals are not kept separate from household members in the poorest households in Nicaragua, creating unsanitary conditions particularly for young children (Nicaragua Qualitative Poverty and Exclusion Study, 2000). 7 For further discussion of sector issues, see MINSA (2000); the Report for the Inter-American Development Bank project, Nicaragua Support for MINSA Hospital Modernization, (NI-0024) (1998); and World Bank (1998). 33 3.11 MINSA's problems of resource allocation and service efficiency are largely the result of inappropriate management systems and incentives, including: weak institutional organization and management (inconsistent or absent standards, overlapping responsibilities), non-functional support systems (e.g., personnel management and pharmaceutical supply systems), the absence of accountability mechanisms (input-driven rather than results-oriented incentive structures), weak linkages between the public, private and insurance sub-sectors (leading to inappropriate planning), input imbalances (overstaffing, low productivity, poorly trained and distributed staff, persistent stock-outs of drugs and medical supplies, no equipment maintenance, and inadequate capital investment planning). All these factors impair the ability of the Government to improve health outcomes, and are particularly detrimental to the poor, who rely disproportionately on public health services. FISE Has Contributed to Poverty Reduction in Nicaragua8 3.12 FISE, the Nicaraguan social investment fund, has provided targeted social investments for poor municipalities and households, but reaching the poorest remains a challenge. Recent evaluation findings9 ,0 indicate that extremely poor municipalities received more FISE resources per capita and allocations have become more pro-poor over time. Water system investments were the most progressive (as half of investments were directed to extremely poor municipalities), education and health projects were well targeted, but social assistance (housing and early childhood development centers), environment, municipal infrastructure and latrines were regressive in targeting municipalities in extreme poverty. However, the benefit incidence analysis carried out at the household level indicates that latrine investments are very pro-poor (with one-third going to the poorest quintile), education and health investments are reasonably well targeted, water investments are neither pro-poor nor pro-rich and sewerage investments reach richer households disproportionately (with the poorest 40 percent of beneficiaries receiving less than five percent of investments). Reaching the poorest remains a challenge-only in primary education and latrines have FISE investments been progressively distributed among the 17 percent of the population in extreme poverty. 3.13 FISE investments have resulted in positive, significant changes in access to social sector infrastructure and in children's health and education status. The results of the household impact evaluation demonstrate a significant, robust impact of FISE primary education investments on education outcomes, including increased enrollment, a reduction in the education gap and children's age beginning primary school. FISE water investments point to a significant, probable impact on water supply and stunting in children under six, while other malnutrition and diarrhea variables indicate improvements, but the results are not significant health outcomes. FISE sanitation investments in latrines and sewerage significantly improved access to these services, but no measured impact was detected on health outcomes. The results for FISE health investments point to a probable, significant impact on the utilization of FISE health posts to treat children with diarrhea, but no measured impact was detected on health outcomes. 9 FISE investments are a sub-set in education, health, and water and sanitation sectors. 9 World Bank (2000b). '° Since 1992 FISE has used a poverty map to distribute its investments across extreme, high, medium and low poverty municipalities. FISE is recently introducing a participatory microplanning process through which community leaders, local government officials, line ministry officials, and FISE representatives work together to target FISE investment to the poorest members of the community. The results of this process should be assessed in future evaluation efforts. 34 3.14 The gender and poverty dimensions of household impact indicate that: (i) the impact of FISE education investments on enrollment is higher for girls; (ii) the education gap is reduced more for children from the poorer quintiles; and (iii) age at first grade is reduced slightly more for boys than for girls. 3.15 The FISE Impact Evaluation is leading to changes in FISE policies. As a result of the FISE Impact Evaluation, FISE has suspended the financing of new sewerage projects for two years and will begin financing more integrated infrastructure projects that will include components complementary to the basic physical works (e.g., access roads and living quarters for teachers in rural areas). FISE is also revising and strengthening its own project appraisal capacities and monitoring and evaluation systems. 3.16 Policy recommendations for FISE, line ministries and others responsible for the design and implementation of social sector projects are a challenge ahead. The FISE Evaluation clearly reveals that the supply of infrastructure alone is insufficient to achieve desired household level outcomes and the participation of the poorest of the poor. The results support FISE's new efforts to increase collaboration at the local level through its microplanning process that involves local stakeholders in project planning, but FISE needs to improve its coordination with other organizations, including both NGOs and line ministries, to achieve desired results. In particular, the evaluation points to a need for: (i) piloting alternative health service provision arrangements in collaboration with MINSA based on contracting out health post management to NGOs to provide needed staffing, medical supplies, medicines and outreach; (ii) exploring different water supply models for poorer areas with INAA and/or other water service providers; and (iii) supporting outreach, information and education campaigns, particularly in health and hygiene. Access to Potable Water Remains a Challenge, as Does Adequate Use of Sanitation 3.17 In water and sanitation considerable resources have been invested in improving coverage, but access to potable water remains a challenge, particularly for the poorest. Although 84 percent of all households have access to safe water, only a third have piped water inside the house which means that the rest needs to get water from outside the home and store it before consuming it. Among the poorest, water safety is even more questionable as only five percent has piped water inside the house while 95 percent get it mostly from public standpipes or rivers and have to store water. Coverage of sanitation services is extremely low, and although considerable funds have been allocated to latrines, the evaluation shows that their impact on health outcomes has been negligible." The high incidence of diarrheic diseases could be largely explained by these pattems. Priorities for water and sanitation are to focus on assuring the consumption of potable water by massive education on storing and preparing water for safe consumption.12 Rural Development Projects Tend to Have Low Impact 3.18 Rural development spending, involving programs for poor farm households to improve their living conditions and productive potential, is considerable as compared to traditional social sectors, but evaluation findings indicate very low impact. Rural " Qualitative evidence indicates that latrines are not used adequately, particularly as the poorest themselves report they lack resources to purchase lime (Nicaragua Qualitative Poverty and Exclusion Study, 2000). 12 The majority of the poorest rural and urban households allow domestic animals to have free access to water sources for human consumption (Nicaragua Qualitative Poverty and Exclusion Study, 2000). 35 development spending is close to one third of total health and education spending separately. The multiplicity of actions, including integrated rural development, productive but also social infrastructure, credit and technical assistance makes coordination essential but in reality has generated duplication, unequal coverage and low sustainability. A recent study reports that only 18 percent of investments in rural development were adequate, while almost 60 percent needed to be modified.13 Lessons learned from this as well as other studies (Box 3.2) indicate that priorities for rural development are: first, project formulation needs to incorporate the poor and less organized groups; second, emphasis should be placed on cost-effective high impact designs; third, sustainability is better achieved when local institutions and civil society organizations are involved; and, fourth, objective criteria needs to be used to identify and select beneficiaries. Local Development Projects often Lack Technical Feasibility and Profitability Criteria 3.19 Local development, mainly undertaken by municipalities supported by INIFOM and FISE, has had positive results on strengthening local capacities and ownership, but projects often lack technical feasibility and profitability criteria. Three important lessons emerge from evaluation findings: one is the need to identify gaps in technical support needed by communities for them to select the most profitable projects and ensure effective execution; two is to define technical and financial standards to be fulfilled by projects to ensure profitability; and three the need for external controls of local administrative entities to ensure proper project execution and follow-up.14 Box 3.2: Nindiri Case Study Poor farners have been helped by an integrated project of the Nicaraguan Institute of Human Promotion (INPRHIU). The project aimed to change the productive patterns of farmers who own one hectare of land, by moving them from single-crop production to a mix of fruits and vegetables. Another objective was to change nutritional patterns through this diversification and to promote a variety of products throughout the year. The project's main components are: (i) soil conservation and appropriate land use; (ii) production diversification; (iii) financing and repayment in kind (seeds and plants), which reduces the stress of indebtedness; (iv) commercialization and market identification; (v) organic production; and (vi) nutritional education. This project changed production pattems, improved land use, introduced commercialization into the production cycle, and changed nutritional habits. Among the lessons learned: - Long-term technical assistance (two years) helps sustainability, - Loans in kind are less likely to cause a debt crisis, - Having surplus as an economic objective, - Diversified production together with nutritional education can improve land use and productivity, and change nutritional habits, - The permanent presence of agricultural extension officer helps sustainability, and - Support is needed to encourage subsistence farmers to grow and sell surpluses. Source: Nicaragua Qualitative Poverty and Exclusion Study (2000). Social Assistance Programs Consistently Demonstrate Low Impact 3.20 Social assistance programs include a multiplicity of actions with a wide variety of designs for which evaluation findings consistently demonstrate very low impact. Social 3 MAGFOR (1999). 14 World Bank (1999c). 36 assistance programs are partly concentrated in the Ministry of the Family and the Social Action Secretariat, with the Education, Health and Agriculture Ministries also executing similar programs. Lack of clarity in the Government's definition of social assistance programs is at the core of their very low impact. The Government defines these programs broadly as those covering vulnerable groups, but without defining the concept of vulnerability or clearly identifying priority groups. Thus, Nicaragua lacks an effective social safety net and current efforts are disorganized and patchy, often unable to reach the right people in a timely fashion. In consequence, an important effort is needed to develop a framework for social assistance in Nicaragua that will clearly define the scope of such programs, including the types of risks to be covered (individual versus covariate), identification of effective risk management strategies currently being undertaken by households and the role of the public sector, and the selection of the most cost- effective mechanisms for high impact interventions. Successful experiences need to be identified and replicated, focusing on cost-effective designs supported by clearly defined regulatory and supervisory responsibilities. 3.21 Substantial evidence indicates that nutrition improvements for extremely poor families should be one of the highest priorities for social assistance programs'15 Priority actions should include increased food consumption but not necessarily food handouts, diet diversification, and improved nutrition practices, hygienic habits and sanitary conditions. Child malnutrition in Nicaragua is closely associated to diarrheic episodes and consequently households who adequately store and treat water before human consumption have better nourished children.'6 Similarly, evaluation findings indicate that malnourished children can recover with timely identification and if treated adequately at early stages with a results-oriented design to attain the child's recovery. 7 A permanent community-based growth promotion program for young children based on individual nutrition status and trends is an essential element. 3.22 Nutrition programs in Nicaragua amount to about US$53 million annually,'8 but are not cost-effective due to lack of a results-oriented framework and targeting criteria, there is considerable duplication and dispersion, and programs generate important negative effects. A nutrition assessment by UNICEF'9 estimates that spending about US$5 per child per year is the minimum necessary to carry out effective nutrition programs. In Nicaragua, annual nutrition spending amounts to US$65 per child under five years, which is clearly excessive as the intended results are not attained. Generally, so called food and nutrition programs in Nicaragua are actually quasi-safety net programs, and they largely fail as nutrition programs because of this. There is a general misconception that food handouts can be the solution to malnutrition and/or food security without using international best practice designs. In Nicaragua, among the trilogy of food insecurity, disease, and household behavior, the latter remains the most relevant while the others have improved over the years. In consequence, public policies and programs need to focus on changing nutritional behavior in order to improve effectiveness. 3.23 Many well-intentioned nutrition programs exist but they lack definition of goals and expected results, making it difficult to adopt sustainable changes or evaluate performance.0 5 See Annex 20, Chapter II. 16 Chawla (1999). 17 Information from USAID's current nutritional programs in Nicaragua, submitted to the World Bank for the study on "Gasto Publico Social y Programas por Sectores en el aflo 2000". 18 The total including emergency programs is US$97 million annually. The entire sub-set is included in above-referenced sectors and not only within the Social Assistance Sector. 19 UNICEF (1999) 20 Seireg et al. (1999) 37 Although mostly financed by external donors and public funds, the majority of programs lack basic organizational and financial information. Selection criteria to target priority beneficiaries is commonly not present. Evaluation findings indicate severe duplication of benefits received by less poor families while the most needy remain excluded. Programs with different approaches are often in the same area of influence, generating inconsistencies between programs with ostensibly the same goals. In addition to operational deficiencies, some nutrition programs generate negative impacts by distributing and promoting food items rarely consumed by Nicaraguan families, by generating dependency, and by making available food items without ensuring that the most vulnerable individuals will consume them (such as young children). Positive findings are that micronutrient programs appear to be making progress, including reaching the poor with fortified foods, and breastfeeding promotion has progressed well. These programs cost little but have large impact on nutrition, learning, and health. In addition, nutritional program staff are dedicated and fundamentally interested in improving the nutritional status of the vulnerable groups in their area of influence. Housing Benefits Are Not Targeted to the Poorest 3.24 Housing benefits are all captured by the owner itself, in contrast to public health or education that generates positive externalities captured by the society at large. The role of the public sector is to facilitate savings mechanisms to foster the development of housing and mortgage markets. In Nicaragua, there are practically no evaluations of experiences with housing projects. Housing spending is US$10 million a year, which is one fifth of primary level recurrent education spending and ten times as much as pre-school. Although almost 80 percent of the poorest in Nicaragua live in homes with dirt floors creating unsanitary conditions, targeting criteria of housing projects is for low income families with re-payment capacity. Thus, these resources are not targeted to the poorest. Given the opportunity cost of scarce resources, these funds would be better spent on programs directed at improving health, nutrition and education services targeted to the poor. B. SUSTAINABILITY OF ADVANCES IN SOCIAL SERVICE PROVISION AND SOCIAL OUTCOMES 3.25 Across all social sectors broadly defined 2' there are a number of concerns which call into question the sustainability of advances in social services provision and thus in social outcomes achieved so far in Nicaragua. 3.26 Social spending is not well targeted to the poorest. Although progress has been made in health and education, many of these programs do not benefit the poor, but those who are better- off. At the primary health level, the share of the total subsidy received by the poor increased from 47 percent in 1993 to 61 percent in 1998. The distribution of the total health subsidy became fairly evenly distributed between the poor and non-poor, shifting from 40/60 in 1993 to 50/50 in 1998, mainly as a result of changes at the primary level of care (Figures 3.1 and 3.2). 3.27 In education, in 1993 the primary school subsidy was equally allocated between the poor and non-poor, but the allocation to the poor increased to 57 percent in 1998, with the non-poor receiving 43 percent. However, the distribution of the total education subsidy did not improve between the poor and non-poor groups, remaining at about 30/70 in both 1993 and 1998 (Figures 3.3 and 3.4); this occurred even though targeting to the poor at the primary level became better but for higher levels of education it worsened. The positive shift in the incidence of health 21 Including education, health, water and sanitation, rural development, local development, social protection and housing. 38 spending reflects increases in coverage especially among the poorest groups, but because the poor have higher fertility rates and therefore more children than the non-poor, the share of the subsidy for the poor would need to increase even further to equal the coverage of the non-poor. In education, the public subsidy for university and higher education continued to favor the wealthiest groups. The 24 percent budgetary allocation of total education spending to the tertiary level is not justifiable on equity grounds. Figure 3.1 Figure 3.2 Incidence of Health Spending 60% 56% 60% 60% 40%40 20% ~~~~~~~~20% 0% L0 All Health 1993 All Health 1998 Figure 3.3 Figure 3.4 Incidence of Education Spending 70% 68% 60% 60 O 40% 200/6 20%20 20 /% 0|l 0° l All Education 1993 All Education 1998 3.28 Efficiency in the use of resources is low. The only option for Nicaragua to finance social sector spending is to improve efficiency in the use of the limited resources that are available. The introduction of more cost-effective delivery mechanisms with better targeting is essential. To achieve this, the roles of the Technical Secretariat to the Presidency (SETEC) and the National System of Public Investment (SNIP) in evaluating program and project impacts need to be strengthened. There have been few studies of the long term impact of various programs, which would be a basis of making decisions on trade-offs between competing demands. Nicaragua continues to spend scarce resources on labor training, for instance, despite international evidence that these programs have very low impact. Because of constitutional mandates, the Government is forced to provide free university education, which largely benefits students from non-poor families. 39 3.29 There is a high dependence on aid to finance recurrent expenditures in the social sectors. The low levels of domestic resources available to finance recurrent spending have meant that public services are highly dependent on donor aid. Looking toward the future, the tight macroeconomic situation is not expected to permit significant increases in recurrent finance beyond current levels, even factoring in the benefits of HIPC debt reduction. This problem is more acute in the social sectors, because the largest element of operating costs in health and education is salaries-56 percent in health and over 90 percent in education. To address this problem, donors have increasingly resorted to financing those elements of recurrent costs that they can justify under sustainability rules, such as performance incentives instead of salary increases. While helpful, these efforts do not resolve the core problem of low recurrent financing. 3.30 The extremely low salaries paid in the social sectors may be reducing effectiveness. The Central Government pays the second lowest salaries in the economy, next to the agricultural sector, and within the Central Government, the lowest salaries are paid in the social sectors, especially to teachers and nurses. Most education studies identify teacher quality (as measured by teacher knowledge of subject matter and pedagogical skills) as one of the key variables most strongly and consistently associated with the overall quality of education services12 In Nicaragua, the low quality of pre-service teacher training, high incidence of non-accredited teachers, and low retention rates of good teachers mitigate against teaching quality. Similarly, in health, community-based health care, linked appropriately to an institutional referral system, represents the most cost-effective intervention to provide the preventive health services needed to address the prevalence of infectious, reproductive and nutrition-related health problems that represent the greatest burden of disease and premature death in Nicaragua. Yet Nicaragua continues to employ twice as many doctors as professional nurses, more than 50 percent of health care professionals work at the secondary level, and staff are disproportionately concentrated in the Pacific region.23 At the low salaries paid, the best teachers and health workers will continue to migrate to other activities and the private sector, making it extremely difficult to improve service quality and coverage.24 Higher salaries will not automatically yield better teaching and health care. Instead, the salaries need to be raised in exchange for better performance. Initiatives Box 3.3: Cost of Medicines Limit Use of Public Health Services People complain that although basic equipment is available for curative medicine, most health posts and centers charge for syringes and bandages, increasing private costs. In addition, beneficiaries report having to pay for medicines-health posts and health centers provide prescriptions and people either pay for medicines at public facilities or more often have to purchase medicines at private pharmacies due to lack of supplies. Poor people complain that their earnings are insufficient to buy medicines: "buildings and personnel are there, but we cannot afford to purchase medicines at private drug stores." Source: Nicaragua Qualitative Poverty Study, 2000. underway to introduce performance incentives in the health and education ministries, supported by various donors including the World Bank, represent positive initial steps in this direction. 3.31 The cost to poor households of obtaining social services, as a share of total per capita household expenditures, increased dramatically during the mid-1990s, even faster 22 World Bank (1995). 23 MINSA (2000). 24 Health professionals employed in the private sector earn 70-100 percent more than their counterparts in the public sector. "Estudio del Mercado Laboral en Salud en Nicaragua" Consorcio FUNSAGLAS, Managua, Julio 1997. Proyecto MlNSA-Banco Mundial. Domestic employees in Managua earn on average twice the monthly teaching or nursing salary. 40 than the cost increases experienced by non-poor households. While to a large degree, these increased expenditures represent increased consumption of health services, and the introduction of school fees, these cost increments nevertheless throw into question the sustainability of the gains achieved in extending service coverage for the poor. If this trend continues, the poor will be increasingly unable to afford basic social services, slowing down or even reversing the progress achieved in improving health and education outcomes (Box 3.3). 3.32 Although the public subsidy has shifted toward the poor at the household level, the share of per capita expenditures devoted by poor households to health has increased three times more since 1993 than for the non-poor (Figures 3.5 and 3.6). In particular, extremely poor households doubled the share of their per capita expenditures devoted to health over the period, while the poor increased their spending by 77 percent. In education, the increase was less dramatic, but still impacted the poor the most. Reflecting the higher price elasticity of demand for health services among the poor, the use of health care facilities and providers became increasingly segmented during the 1990s. The poor increasingly sought health care from the least cost facilities (health posts), while the non-poor increasingly selected the higher cost private sector providers. Similarly, while the non-poor are more likely to consult physicians, the poor are more likely to be seen by nurses, pharmacists, community health workers or other low cost providers. In particular, larger families prefer nurses and pharmacists over physicians. As a result, the non- poor spend nearly six times more than the extreme poor per consultation. However, for all poverty groups, roughly 80 percent of total household expenditures on primary health care is devoted to medicines. Thus, even when the poor gain access to services, these services do not include medicines-evidence that public expenditures on medicines (40 percent of MINSA's recurrent budget) are failing to benefit users at the primary level. Figure 3.5 Figure 3.6 Share of per capita household expen(itures Share of per capita household expenditures devoted to Education devoted to Health 6 6 5 .,.Total oa Extreme PoPoor Poor | 2 xtPoor P O~. *w i . | s |. Non Poor 1 Non-poor 1993 1998 1993 1998 3.33 Private costs of schooling are high-66 percent of these costs are due to uniforms, school supplies and textbooks. Transport and food at school account for another 20 to 24 percent. School fees, which more than half of all families report as not being voluntary, constitute the smallest share of all education costs, at roughly 10 percent. Qualitative evidence indicates that their unpopularity is in part due to the policy change from free education during the 1980s, and also to the discretionary nature of their application. 3.34 Finally, there is consistent evidence suggesting that negative household behavior in Nicaragua is a major deterrent to improving social outcomes and there has been limited success in modifying this behavior. Improving the performance of social public expenditures will not be enough to improve social outcomes in Nicaragua. Although the Govemment has achieved significant gains in social service coverage, the impact is less than expected because 41 increased availability of services is not accompanied by efforts to enable households to use them. The majority of households do not treat or boil water before human consumption. One fourth of all women report that they are unable to seek health care for themselves or their children without permission from someone else, usually their partners. In education, distance to schools is a much greater barrier for girls than it is for boys, reflecting family concerns about their daughters' safety if obliged to travel long distances to school. Qualitative findings indicate that in the majority of the poorest households in Nicaragua domestic animals have free access to water for human consumption and are not kept separate from young children. Broad education campaigns to positively impact household habits affecting social outcomes are crucial. CHAPTER IV. AGRICULTURE AND RURAL POVERTY 4.1 Economic growth during 1993-98 was led by agriculture. In terms of value added, agriculture grew by almost two-thirds and represents over one-fourth of total GDP. Agricultural GDP averaged annual growth of 10 percent during 1993-98, or about 7 percent in per capita terms, whereas GDP growth was only 4.3 percent. The most dynamic products in terms of annual growth of area harvested were agricultural exports: coffee (4 percent), sugarcane (7 percent) and banana (3 percent), and basic grains: beans (9 percent), rice (5 percent) and corn (4 percent). 4.2 Unusually favorable export prices, particularly for coffee and banana, plus area expansion for basic grains and favorable prices for red beans, contributed to the agriculture boom (Figures 4.1 and 4.2).' These developments arguably constitute the single most important factor responsible for the more egalitarian growth pattern observed in this period. A substantial share of the poor derive their earnings from agriculture, especially in the Central region, which has the highest share of extreme poverty and where rural poverty declined the most. Figure 4.1 Figure 4.2 E.p.rt Crops: Ares *nd Yield IBasc Gni-9: Arma *nd Yield 1990-98 199991 1991 S 9 9 13 WfW I7I W 19929 l9S,ll 290-9 29099 2969 2999-9 2999lt w. -9 _nAca Har-H-led -S99D _Coffec -Sus,gC.-e B..C... n|lna _ |-1o19TA-H -..d R - Bcam - CoI a 4.3 The majority of the poor are engaged in agricultural employment. Agriculture accounted for 60 percent of the employment of the poor, and 75 percent of the employment of the extremely poor in 1998 (Table 4.1). By contrast, only 21 percent of the non-poor work in the agricultural sector, with a higher concentration in more urban activities, such as manufacturing, commerce and government. Most of the poverty reduction during the l990s was due to a revival of agriculture. 4.4 Growth in agricultural exports and basic grains was the result of two different factors. While export growth is mostly related to productivity gains, growth in basic grains can be largely explained by area expansion (Figures 4.3 and 4.4). Total area harvested in the period 1990-91 to 1998-99 stayed about the same for major agricultural exports, while it grew by almost 60 percent for basic grains. I Among the major export crops, the price of coffee grew at an average annual rate of 22 percent in 1993- 98, while sugar and banana grew by 0.6 and 5.7 percent, respectively. The price of red beans, a major non- exported basic grain, grew at an annual rate of 11 percent during the period. 43 Table 4.1: Distribution of Employment by Sector, 1998 (percent of total) Sector Extreme Poor Total Poor Non-Poor Total Agriculture 74.6 58.8 20.5 37.3 Mining 0.2 0.4 0.7 0.5 Manufacture 4.2 6.6 11.3 9.2 Electricity, gas and water 0.2 0.2 1.0 0.7 Construction 2.8 3.9 5.0 4.6 Commerce 7.7 14.0 31.4 23.8 Transportation 0.9 2.0 4.9 3.6 Financial services 0.2 0.9 3.6 2.4 Community services 9.2 13.2 21.6 17.9 Total 100.0 100.0 100.0 100.0 Source: LSMS 1998. Figure 4.3 Figure 4.4 Agricitura.I pri a - Espot Goods AgHiwultwrl Prce9 - Basic Grains (1992 - 11M)) (1992=140) :24 / 4(62lj 9 994 129 1299 199- 2991 292 2 9 2.,995 2996 99 9 S_gar -*..Mu.2 .B a _. [ _ 1 .h.*Rd B.. Whfe C- - 9.. ghu 4.5 Yields improved only for major export crops but productivity declined for all basic grains. In the period 1993-98, yields of export crops such as coffee, sugar cane and banana grew between 30 and 40 percent, as higher export prices seem to have encouraged more intensive cultivation of existing tree-crop plantations. It is possible that planted areas increased, but have not yet been harvested. For basic grains, the increase in labor supply permitted more extensive cultivation, bringing back into production lands which had fallen idle during the civil war. As a result of more extensive cultivation, yields of basic grains such as rice, beans and corn decreased between 3 and 12 percent (Table 4.2) as area harvested increased. 4.6 The Central region experienced rapidly growing incomes with the largest decline in rural poverty. This region accounts for 95 percent of total coffee production (Table 4.3). Stimulated by high international prices, coffee production grew at an average annual rate of 9.5 percent (1993-98) providing the region with rapidly rising incomes and employment. Similarly, bean production in the Central region was also an important source of incomes and employment. Output grew at 11 percent annually with prices also rising by 11 percent. In addition, this is the only region where meat processing averaged annual growth of 4 percent over the period. 4.7 Agriculture in the Central region is highly dependent on coffee and tobacco, which have strong backward and forward linkages, and hence high multiplier effects. Backward linkages come primarily from the commercialization of agricultural inputs. Forward linkages are related to coffee cleaning, packing, transport, etc. Agriculture growth thus stimulated growth in other sectors, such as commerce, transport, construction, and services. (Box 4.1). 44 Table 4.2: Agricultural Production: Total Growth 1993-98 (percent) Product Output Area Harvested Yield Exports Sesame -74.0 -54.8 -42.5 Cotton -100.0 -100.0 -100.0 Coffee 52.6 19.8 27.4 Sugar Cane 99.7 40.1 42.6 Tobacco -2.3 89.9 -48.5 Banana 61.1 15.5 39.5 Peanuts 86.7 124.2 -16.7 Soybean 199.2 187.9 3.9 Basic Grains Rice 21.6 25.2 -2.8 Beans 49.0 56.2 -4.6 Corn 7.0 20.9 -11.5 Sorghum -48.1 -22.0 -33.5 Source: Banco Central de Nicaragua. Table 4.3: Agricultural Production: Output Growth by Region Traditional Growth Rate Share of Total Basic Grains & Growth Rate Share of Total Exports 1993-98 Production Others 1993-98 Production PACIFIC Sesame -25.7 86.0 Rice -12.4 17.2 Cotton -100.0 100.0 Beans -20.2 4.2 Coffee -3.7 3.5 Corn -16.0 7.1 Sugar Cane 19.2 77.1 Sorghum -16.5 51.0 Tobacco 8.8 34.8 Peanuts 13.3 100.0 Banana 10.0 100.0 Soybean 24.5 100.0 MANAGUA Sesame 58.6 14.0 Rice - 0.0 Cotton - - Beans -16.3 0.6 Coffee 9.8 2.1 Corn 0.1 1.1 Sugar Cane 14.9 22.2 Sorghum -11.9 17.5 Tobacco - - Peanuts b - Banana - - Soybean b - CENTRAL Sesame -100.0 0.0 Rice 0.2 33.0 Cotton - - Beans 11.4 72.0 Coffee 9.5 94.3 Com -1.6 65.7 Sugar Cane - - Sorghum -1.7 0.3 Tobacco -1.2 65.2 Peanuts b- Banana - - Soybean b - ATLANTIC Sesame - - Rice 25.4 49.7 Cotton - - Beans 16.5 23.2 Coffee - - Corn 43.6 26.2 Sugar Cane 5.1 0.7 Sorghum - Tobacco - - Peanuts Banana - - Soybean Source: Banco Central de Nicaragua and MAGFOR. b In Managua and the Central region, some production of peanuts and soybean occurred sporadically. 45 Box 4.1: Decomposition Analysis of Changes in Poverty2 Comparison of the 1993 and 1998 LSMS shows that the levels of extreme and moderate rural poverty fell between 1993 and 1998. Overall, per capita household consumption in rural areas increased over five percent between 1993 and 1998. This decomposition exercise attempts to determine econometrically what may have led to this decrease in rural poverty. The surveys are not panels, so we are limited to looking at changes in average characteristics. Given these limitations, we must refrain from implying causality in analyzing these changes. The decomposition follows techniques first used by Oaxaca (1973) to determine which household characteristics are associated with the increase in welfare that occurred from 1993 to 1998. They regress total log per capita household consumption on a set of household characteristics and assets that have been divided into nine categories: log Con, =C, +f8, *X, +1 where C and the vector B are the terms to be estimated, X is the vector of the eight categories, and 1a is a random error term. The nine groups are composed of: (i) participation in agriculture and the number of heads of cattle, (ii) age and gender of the head of household, (iii) log of family size, (iv) number of male adults with different levels of education (none, less than primary, primary, secondary, and higher), (v) number of female adults with different levels of education, (vi) demographic composition (number of females and males by age category), (vii) ownership of dwelling, (viii) access to infrastructure (running water, toilet, and electricity), and (ix) geographic region. Using the estimates, we decompose the change in consumption into the share corresponding to changes in the mean level of household characteristics (the vector X in each equation), and the share corresponding to differences in returns to these characteristics, or the Betas. More specifically, ACon = (C93 - C98) + X98 * (93 - 698) + b93 *(X93 - X98) where Con is (log of) per capita consumption, C93 and C98 are the constant terms in the regression for corresponding time period, X93 and X98 are the mean characteristics of households, and b93 and b98 are the coefficient vectors. Thus the change in consumption between 1993 and 1998 can be decomposed into three components. The first is the difference in the estimated constant term. The second is the difference in coefficients, which reflect the environments in the two periods, leading to different returns to household characteristics. The third is the difference in characteristics of households, or the change in level or composition of endowments. The results are summarized below. The first column is the change in Betas, or return to endowments, and the second column is the change in levels of characteristics. Approximately 83 percent of the total change in consumption among rural households is comprised of changes in long-term indicators of well-being (increasing educational levels and decreasing family sizes). Increases in both male and female education have smaller, though positive, shares in the change in consumption over the period. The Central region, with the largest decline in rural poverty, shows the largest increase in returns to agricultural/livestock assets. Decomposition of Change in Consumption, 1993-1998 Rural Urban Explanatory Variables A B Dx A B Dx Agricultural/livestock 8 0 2 2 Age-head of household -40 0 -37 3 Gender-head of household 2 0 14 0 Family size -36 -64 -447 -300 Male education 14 -7 -91 -12 Female education 23 -8 -64 -I Demographic composition 16 0 368 23 Housing -3 1 -32 23 Infrastructure -7 -8 -20 22 Atlantic Region I -I -4 0 Central Region -30 3 -15 3 Pacific Region -6 0 60 -4 TOTAL -6 -83 141 -241 2 A detailed description of the analysis and methodology used for this decomposition is in Davis and Murgai 2000, Section VI. 46 4.8 Increased labor participation raised household incomes. Total employment, rural and urban, increased in 1993-98 beyond the rate of growth of the labor force, even in sectors where real wages fell, suggesting a rightward shift in the labor supply curve. This increase occurred both in the number of people working as well as number of hours worked. The average proportion of household members working increased from less than 30 to almost 40 percent (Table 4.4), representing an increase of 37 percent, while the number of hours worked rose enough to reduce underemployment from 40 to 30 percent. Table 4.4: Average Share of Household Members that are Employed (percent) Region 1993 1998 Change Managua 31.0 36.3 + 5.3 Pacific Urban 29.4 40.7 + 11.3 Pacific Rural 25.4 39.4 + 14.1 Central Urban 27.8 42.1 + 14.3 Central Rural 29.5 42.8 + 13.2 Atlantic Urban 26.6 36.0 + 9.3 Atlantic Rural 28.7 40.4 + 11.6 Total 28.9 39.7 + 10.8 Source: Nicaragua LSMS 1993 and 1998 4.9 Employment in agriculture increased by almost 200,000 jobs, while real agricultural wages fell. Agricultural employment averaged annual growth of about 10 percent, which was substantially higher than growth of the labor force (3-4 percent)3 The net effect of higher employment and falling real wages, however, was to increase household incomes, as the share of agricultural wages as a source of income doubled for the rural poor over 1993-98 (Table 4.5). Table 4.5: Nicaragua: The Evolution of Sources of Income, 1993-98 1993 1998 Overall poor Non-poor Overall poor Non-poor Sources of Income Urban Rural Urban Rural Total Urban Rural Urban Rural Total Labor Income 73.0 79.7 68.0 66.0 66.0 84.1 82.4 73.7 79.8 76.6 Work wage agric. 4.8 13.0 1.9 5.3 5.3 10.1 26.8 1.8 11.3 7.3 Work wage non-ag 37.7 19.8 33.8 18.4 18.4 47.6 24.7 43.4 24.8 38.4 Work self-emp. Ag (a) 4.1 33.9 3.6 24.4 24.4 4.5 22.8 1.9 23.6 8.2 Work self-emp. non-ag 26.4 13.0 28.7 17.9 17.9 22.3 7.2 26.6 20.0 22.7 Agriculture 8.9 43.9 4.5 29.7 29.7 14.6 49.6 3.7 34.9 15.5 Non-agriculture 63.1 32.8 61.5 36.3 36.3 69.9 31.9 70.0 44.8 61.1 Imputed rent 14.5 11.6 12.1 7.6 7.6 8.0 8.1 13.5 12.0 12.1 Non-labor Income 12.5 8.4 19.1 26.4 26.4 7.7 9.7 12.8 8.3 11.3 Remittances received 5.5 2.3 4.6 3.1 3.1 3.9 3.9 7.4 4.3 6.2 Gifts, charity, others (b) 1.9 2.9 7.2 15.9 15.9 2.9 5.9 1.4 3.1 2.4 Pensions, ret. Capital 5.1 3.2 7.3 7.4 7.4 0.6 0.8 4.0 0.9 2.8 Total 100 100 100 100 100 100 100 100 100 100 Source: Nicaragua LSMS 1993 and 1998 4.10 The rural poor benefited disproportionately from job creation and growth in agriculture, as they comprise almost 75 percent of total agricultural wage earners and of all self- 3Based on a population growth rate of about 3.5 percent in 1985 (ESDENIC-85). 47 employed in agriculture. The most important source of increased welfare for the rural poor, however, was wage employment as this source of income doubled between 1993 and 1998, while agriculture self-employment shrank by almost 33 percent. 4.11 Managua was the only region to experience growth in average real wages. Not surprisingly, it is also the region where household members employed as a share of total household members increased the least, by about 17 percent (Table 4.6). Table 4.6: Employed Persons in the Household Working in Agriculture (percent) Region 1993 1998 Change Managua 6.0 8.1 + 2.1 Pacific Urban 6.3 11.5 + 5.2 Pacific Rural 43.9 55.5 + 11.5 Central Urban 22.6 21.1 - 1.5 Central Rural 83.9 76.3 - 7.6 Atlantic Urban 26.4 27.2 + 0.8 Atlantic Rural 82.2 84.8 + 2.6 Total 37.1 37.4 + 0.3 Source: Nicaragua LSMS 1993 and 1998 4.12 The increase in labor participation rates in the presence of falling agricultural wages may reflect an improvement in the overall working environment, which has permitted rural enterprises to reinitiate operations and absorb workers that had previously withdrawn from the labor market (Box 4.2). Box 4.2: Increased Labor Participation with Falling Agricultural Wages Labor demand did not remain constant during the Labor Supply and Demanid Shift j period 1993-98-almost 200,000 new jobs were created in agriculture as a result of the high growth of this sector. Wag In the Figure, both the labor supply and labor demand . \ LS),n curves experience right-ward shifts. The decrease in /* Ls, ,wages can be explained by a greater increase in the labor force relative to the increase in labor demand. How is it possible that rural poverty declined in Nicaragua between 1993 and 1998 if, simultaneously, 1,3 agricultural real wages decreased? This figure illustrate Labo the increase in labor supply in the agricultural sector-in terms of total employment, percentage of household members employed, and hours worked-observed in Nicaragua. A poverty decline implies that the increases in quantity of labor supplied more than offset the negative price effect of lower real wages, enough to improve welfare levels. The Sustainability of Agricultural Growth 4.13 A large part of recent employment and output growth was due to the revival of agriculture. These gains, however, may largely represent a one-time effect from a combination of a reopening of the country-side with the restoration of peace, plus the availability of additional demobilized labor, thus permitting more extensive agricultural activities. These favorable factors were combined with favorable international commodity prices, particularly coffee. Thus, the 48 sustainability of recent agricultural growth is a major concern. In addition, Nicaragua's agriculture has a low level of diversification, and thus vulnerable to terms of trade shocks. Table 4.7: Protection Indicators, 1998 (percent)4 Product Nominal Effective Importables: Rice 1.29 1.36 Beans 1.13 1.20 Sorghum 1.37 1.53 Corn 1.38 1.50 Soybean 1.10 1.17 Milk 1.15 1.18 Exportables: Coffee 0.98 0.99 Peanut 1.01 1.03 Sesame 1.02 1.03 Sugar Cane 0.98 1.03 Meat 1.02 0.99 Source: Valdes (1999) with MAGFOR. The 'bfficial" exchange rate is the current market exchange ratee. Results are preliminary. 4.14 Reducing or eliminating the remaining anti-export bias may help sustain future agricultural growth. Despite past reforrns,5 there still remained an anti-export bias in Nicaragua's agricultural incentive framework in 1998 (Table 4.7). The effective rate of protection6 of import-competing agricultural products is positive, while that of exportable goods is negative. Removing this bias would have a positive, but admittedly small, effect on the incomes of the poor. Complete trade liberalization actually makes poor producers worse off, but this is offset by a gain on the consumption side from lower prices. The net effect is positive for the poor; their income would rise by 1.7 percent, while the income of the extreme poor would go up by 2.3 percent (Box 4.3). More generally, the anti-export bias will remain as long as there is a lack of adequate infrastructure, transport, ports, communications, etc. to support export-oriented agricultural production. 4 The effect of direct and indirect interventions on agricultural product markets can be measured by the nominal rate of protection (NRP) and the effective rate of protection (ERP) (direct intervention = tariffs, quotas, etc.; indirect intervention = market structure problems). The NRP is defined as the difference between actual domestic market price and world market price at the same trading point. The ERP is defined as the ratio of actual value added, to the value added that would have been received at market prices. Protection rates greater than one mean producers receive higher prices than without interventions and other distortions, and less than one implies price discrimination against producers (Valdes and Kray, 1999). 5 Since 1991, Nicaragua has been implementing economic policies aimed at re-activating exports. The measures have included liberalization of export prices and providing tax benefits to exporters. The Export Promotion Law (Decree No.37-91) authorized emission of tax benefit certificates (CBTs), which were tax credits granted to exports of non-traditional goods, as well as the exemption from profit taxes; this program had been gradually phased out by December 1997, and was replaced by new tax legislation that exempts exports from trade tariffs, provides producers a tax credit of 1.5 percent of export value (f.o.b.) and exempts agricultural production services and inputs from the 15 percent value-added tax. 6 The NRP measures the direct protection given to a good, and the ERP measures the amount of protection given to the domestic value-added; if the NRP or ERP is greater than one, the product receives positive protection, and if it is less than one it receives negative protection, i.e., its production is taxed by the existing incentive framework. 49 Box 4.3: Redistribution Effects of Agricultural Incentive Policies7 Methodology of Redistribution Effects of Incentive Policies The impact of tariff reforms on households' income were estimated differently for net producers and consumers, as detailed below (from Schiff and Valdes 1992): Producers: The income effect of tariff reform entailed comparing actual real income (1) at current domestic prices, and real income in the absence of price intervention (2). For producers, farm income YF = Pi * Qi s where Pi is the actual market price of good i, and Qi s = marketed surplus of the agricultural good i. Total producer income is farm plus non-farm income: Y= YF + YNF. (1) y = Y/PI , and PI is a price index of the farmers' CPI at prevailing prices. (2) y' = Y/PI' , and PI' is a price index of the farmers' CPI at prices that would prevail without intervention. The direct effect of protection on the real income of producers of agricultural goods is: (3) Ay = [y - y']l /y' Consumers: The income effect of agricultural tariff changes was measured by the implied changes in food prices, and thus the changes in food expenditures. After classifying households into poverty categories, two consumer price indexes were estimated for each poverty group to measure the effect of changing food prices on the different households: one CPI at prevailing prices, and one after removing the protection effects (CPI'). The exercise consisted in comparing actual real income (4) and income under the different tariff scenario (5). Letting NY = nominal income, we estimate real incomes: (4) Y=NY/CPI (5) Y ' =NY/ CPI' If the effect of tariff reform was to increase food prices, then (5) will be less than (4) and consumers are worse off; if tariff reform decreases food prices, the opposite is true. Net Income Effect by Poverty Group Extreme All-Poor NonPoor All Extreme All-Poor NonPoor All --Nominal Cordobas- -Percent of Total Income-- Producers -14,606 -68,131 427,312 359,180 -1.8 -1.6 1.6 1.1 Consumers 32,831 139,171 394,320 533,491 4.1 3.2 1.5 1.7 Net Effect 18,225 71,039 821,632 892,671 2.3 1.7 3.0 2.8 Source: LSMS 1998. (a) No expansion factors were used. 7 A more detailed description of the methodology, analysis and results is in Kruger (2000). 4.15 Since recovery effects and labor supply shifts in agriculture will most likely not persist into the future, increases in agricultural productivity will be needed to continue improving welfare levels of the rural poor. Nicaragua's productivity is low compared to other countries and has been stagnant during the past 25 years. Low productivity practically guarantees stagnant incomes of poor households involved in agricultural production. Three indicators (Table 4.8) shed light on the evolution of productivity in agriculture: (i) agricultural value-added per agricultural worker, (ii) fertilizer consumption/hectare, and (iii) machine (tractor) use per worker per arable hectare. All of these show a stagnant or decreasing trend in agricultural productivity. Decreasing wages in agriculture in the mid-1990s are evidence of low marginal productivity of agricultural labor.7 7 Assuming labor markets are efficient and perfectly competitive. 50 4.16 Another source of vulnerability in Nicaragua's agriculture is the low diversification of production.8 Economic performance is highly vulnerable to the behavior of agricultural exports because these represent more than one-half of total exports.9 In turn, total exports are highly vulnerable to commodity terms of trade shocks. Table 4.8: Productivity in Agriculture Agric. Labor Growth Fertilizer Agricultural Country Productivity (%) Consumption Machinery 1979-81 1995-97 1979/1995 1979-81 1995-97 1979-81 1995-97 Argentina 12,195 13,833 13.4 46 254 73 112 Australia 20,880 29,044 39.1 269 376 75 65 Brazil 2,047 3,931 92.0 915 898 139 142 Chile 2,612 5,211 99.5 321 1,131 86 119 Colombia 1,926 2,890 50.1 812 2,853 77 118 Costa Rica 3,159 4,627 46.5 2,650 3,636 210 246 Mexico 1,482 1,690 14.0 570 538 54 71 New Zealand n.a. n.a. n.a. 1,965 4,247 367 488 Nicaragua 1,334 1,407 5.5 392 147 19 11 Source: Extracted from Valdes (1999); data is from the World Bank Development Indicators. 4.17 Economic growth performance in recent years has been fueled in part by favorable terms of trade (TOT) of agricultural exports. Favorable TOT shocks (commodity booms) are associated with increased bank deposits and (potentially destabilizing) lending booms. Likewise, adverse shocks can reverse these effects and destabilize the banking system.'0 Direct and indirect effects, such as the widening of the trade and fiscal balances, can augment the impact of TOT shocks (Box 4.4). The economy is vulnerable to these types of shocks, and to their negative impact on poverty, as on average almost 70 percent of export revenues are derived from exports of agricultural commodities.12 Of these, revenues from coffee exports play a significant role, averaging 30 percent of total exports. The international price of coffee rose steadily between 1993 and 1998, and this had a significant positive impact on the economy. Coffee prices, however, dropped in 1999 from an average of US$147/quintal to US$107/quintal (Figure 4.5), and another drop in prices is expected in 2000. 8 Almost 80 percent of the primary sector's value added is generated by eight products: coffee, sugarcane, corn, rice, beans, cattle, poultry and shrimp. Of these, only three products constitute 50 percent (coffee, cattle and sugarcane with 21, 20 and 9 percent, respectively). For the agricultural sector only two products, coffee and sugarcane, account for close to 50 percent of value added. 9 In terms of exports, coffee and sugarcane constitute three-quarters of agricultural exports and half of primary sector exports. Coffee is at present the single most important agricultural export crop: coffee alone constitutes more than 60 percent of agricultural exports, more than 40 percent of primary sector exports and more than one-third of total exports. Sugarcane is a distant second, representing about 12 percent of agricultural exports and 6 percent of total exports. tO Hausmann, 1999. " Indeed, empirical estimates on intertemporal elasticities of substitution reveal that developing countries fall in the lower end of the scale-i.e., persons in developing countries are less willing to substitute current for future consumption. This implies slow adjustment in consumption patterns to a sudden drop in income. 12 Estimated from BCN (1998), Tables 6-4 and 6-5. This is the average proportion of agricultural exports in total exports over 1993-1998. 51 Box 4.4: Direct and Indirect Effects that Augment the Negative Impact of TOT Shocks The balance of trade is immediately affected by TOT shocks. A decline in the price of exports immediately decrease export revenues, but imports-especially consumption goods-are not immediately affected, since individuals are not willing to decrease consumption.11 An adverse TOT shock is likely to widen the trade deficit, which has already increased from -5.7 percent to -16.9 percent between 1993 and 1998, and -23.4 percent in 1999 in real terms. Terms of trade shocks ultimately become fiscal shocks'4 through their indirect effect on fiscal revenues. The fiscal balance in Nicaragua is vulnerable to a reversal of the "consumption boom" observed recently. Consumption growth has increased fiscal revenue via value added and other consumption taxes. If this boom ends, fiscal revenue is likely to drop almost instantly: in 1998, more than 80 percent of fiscal total revenues came from taxes related to consumption, with 26 percent coming from taxes on imports. Expenditures do not respond instantly, so that the fiscal balance is extremely vulnerable to the behavior of aggregate demand. Figure 4.5 Coffee Prices, 1993-1999 160 -140 120 iloo! =60 40] 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 4.18 Vulnerabilities in the agricultural sector can be reduced by a more diversified export base, and the elimination of existing anti-export biases in the incentive framework. 4.19 Increasing welfare levels among the rural poor are unsustainable given the declining trend in agricultural real wages. Employment generation in agriculture of around 10 percent annually in 1993-98 compensated for the drastic decline in real wages of 6 percent per year. Households responded by increasing the share of household members employed and the number of hours worked. However, households face real constraints in providing an ever increasing supply of labor in the future to compensate for declining real wages. 4.20 Boosting agricultural productivity plays a central role in sustaining agricultural growth and simultaneously reducing poverty. Since growth of the labor force in Nicaraguan agriculture is about 3-4 percent,'3 labor-intensive agriculture is critical for poverty reduction, unless productivity increases (leading to higher real wages) compensate for lower employment 13 Based on a rate of a population growth rate of about 3.5 percent in 1985 (ESDENIC-85). 52 generation. Two likely scenarios in Nicaragua's rural labor markets highlight the role of increased productivity: * If agricultural real wages continue to decline (due to increases in labor supply without improvements in productivity), employment generation as a compensatory mechanism to maintain or increase real incomes (an thus improve welfare levels) would imply further pressure on the share of household members employed and hours worked. * A more likely scenario is that agricultural growth, and thus employment generation, will slow to a rate at or below the growth of the labor force. Future increases in rural welfare would require a reversal of the declining trend in real agricultural wages. The Contribution of Non-farm Activities to the Incomes of the Poor 4.21 The most destitute households in rural Nicaragua are those that combine agricultural wage with farm activities, followed by landless agricultural wage earners. These groups are the same as those identified in 1993. Agricultural wage labor does not seem to offer a way out of poverty, but more of a coping or survival strategy, especially when combined with farm activities. By contrast, the most prosperous rural households are those that have no land, are not agricultural wage earners, are self-employed in non-agriculture and live in the Pacific Region. These households also have the highest levels of education among all rural households, but they also have better access to infrastructure and financial resources. Education is the single most important determinant of welfare for rural households, with increasing education levels, particularly among women, having a substantial impact on household welfare. Thus, increasing the access of women to education may have a big poverty payoff. Access to infrastructure, particularly electricity but also paved roads and piped water is associated with higher welfare levels. It is likely that access to infrastructure allows better access to off-farm employment and income-generating activities. Establishing causality, however, is more complex. For example, there is insufficient data to establish whether access to electricity stimulates non- farm activities, or whether connection to the grid acts as a proxy for location of the household near an urban setting, where non-farm jobs are more likely to be available. 4.22 The rural economy is clearly linked to, but extends well beyond agriculture. Recent analyzes of the sources of growth in East Asia have stressed the central role played by the non- agricultural sector in rural areas.'4 In Nicaragua, as in other Latin American countries, the non- agricultural rural sector provides a significant share of total employment and incomes. Employment in this sector is strongly associated with lower poverty, suggesting that it might offer a route out of poverty for those that can gain access to productive non-agricultural employment. 4.23 From a poverty perspective, the non-agricultural sector comprises two distinct sets of activities. On the one hand there is a set of activities which are reasonably productive, relatively well-paid, and which may be comparatively less exposed than agriculture to climatic variations and other uncertainties. On the other hand, there is a group of activities undertaken by persons who are unable to even subsist by working in the agricultural sector. Both sets of activities have a role in reducing poverty but the policy implications are different The former can be regarded as 14 Aoki, M., K. Murdock, and M. Okuno-Fujiwara. 1995. "Beyond the East Asian Miracle: Introducing the Market-Enhancing View", in M. Aoki, H. Kim and M. Okuno-Fujiwara (eds.). The Role of Government in East Asian Economic Development: Comparative InstitutionalAnalysis, manuscript, Economic Development Institute, World Bank. 53 a source of upward mobility or a route out of poverty. From a policy point of view it requires analyzing the kinds of interventions that could stimulate more productive and stable non-farm rural activities. The latter is a coping strategy which may prevent a person from falling further into destitution. From a policy point of view it requires looking at options in the social protection area, either through targeted transfers or well-designed public works programs (as in the Maharashtra Public Works Program in India). The same set of policy options should be considered for areas of high poverty but that have limited economic potential, either because of high population dispersion or poor natural conditions. 4.24 From an economic perspective, the rural economy includes activities strongly linked, through forward and backward linkages, to agricultural production. While available data makes it difficult to test the extent and strength of these linkages, international evidence suggests that other activities not directly linked to agriculture may also be important (e.g., light manufacturing, services and tourism). These activities have the advantage that they can provide sources of rural income and employment that are not affected by agricultural fluctuations. In an indirect way they may also work to encourage agricultural diversification and a move to higher value-added crops, which can also raise farm incomes. As pointed out by Murdoch (1995),'5 the production decisions of small-scale farmers are often aimed at cropping patterns which minimize the risk of harvest failures, but which tend to have lower value added. Access by some household members to non-cyclical sources of income might help to encourage increased diversification and higher- value cropping decisions. In addition, sub-contracting arrangements are present in some Latin American countries and are believed to have played an important role in raising rural incomes in East Asia.'6 Although Nicaragua has considerable experience with maquila industries, it is not known to what extent sub-contracting arrangements are used by maquila or other larger manufacturing enterprises in Nicaragua. 15 Murdoch, J. 1995. "Income Smoothing and Consumption Smoothing," Journal of Economic Perspectives, Vol. 9, No. 3. 16 For evidence on Latin America, see: Lanjouw, J.O. and P. Lanjouw. 1996. "Poverty Comparisons with Non-Compatible Data: Theory and Illustrations", mimeo, Policy Research Department, World Bank. For evidence on East Asia, see: Hayami, J. 1995. "In Serach of Rural Entrepreneurship in Asia: Concept and Approach." Paper presented at a World Bank Conference on Rural Development in Latin America, September, 1995. CHAPTER V. KEY ELEMENTS OF A POVERTY REDUCTION STRATEGY A. PROSPECTS FOR REDUCING POVERTY 5.1 Nicaragua made inroads during 1993-98 in reducing poverty. Is this performance good enough or could it be improved? Future poverty scenarios for Nicaragua can be projected under alternative assumptions of per capita consumption growth. Between 1993 and 1998, GDP averaged real annual growth of 4.3 percent, but due to fast population growth, average GDP per capita growth was only 1.2 percent. If this performance were to continue, poverty would decline by about 20 percent between 1998 and 2015. But is that enough? One criteria to judge performance is by reference to the International Development Goals (IDG) indicators. These call for a reduction in the incidence of extreme poverty by half by 2015. Extreme poverty declined from 19.4 to 17.3 percent (or 2 percentage points) in the five years between 1993 and 1998. Continued performance at this pace would reduce extreme poverty to 13.7 percent by 2015 (Table 5.1), somewhat less that the IDG target. To halve the extreme poverty rate by 2015 would require annual growth of per capita consumption closer to 2 percent. Table 5.1: Poverty Projections with Alternative Annual Growth of Per Capita Consumption Poverty Index 1.2% 2% 3% 1998 2005 2010 2015 2005 2010 2015 2005 2010 2015 Overall Poverty: Head Count 47.9 43.5 40.6 37.9 40.8 36.3 32.4 37.6 31.6 26.5 Poverty Gap 18.3 16.0 14.5 13.1 14.6 12.4 10.5 12.9 10.1 7.9 FGT2 9.3 7.8 7.0 6.2 7.0 5.7 4.7 6.0 4.4 3.3 Extreme Poverty: Head Count 17.3 15.7 14.7 13.7 13.1 10.7 8.8 11.3 8.4 6.2 Poverty Gap 4.8 4.2 3.8 3.5 3.8 3.2 2.8 3.4 2.7 2.1 FGT2 2.0 1.6 1.5 1.3 1.5 1.2 1.0 1.3 0.9 0.7 Source: Nicaragua LSMS 1998 and staff estimates. 5.2 Accelerating growth to this level may be difficult for a number of reasons: * Investment levels are already quite high, and it is difficult to see how they could be raised further. The country lacks access to non-concessional capital inflows, and is unlikely to receive higher levels of official capital flows by which it could finance higher levels of investment. Alternatively, financing a larger share of investment with domestic savings would necessarily curtail domestic consumption. * Investment efficiency appears low, as reflected in a very high capital-output ratio (over 6 in recent years), suggesting the need to focus on improving capital productivity. * Recent high growth rates may be temporary, reflecting the reactivation of the agricultural economy, and may be difficult to maintain. Likewise, the rural economy has benefited from favorable terms of trade, which is unlikely to be sustained. * The impact of high economic growth continues to be mitigated by high population growth. 5.3 Nicaragua's rapid economic growth since 1995 is an excellent performance by Latin American and country standards. Just maintaining that performance over a sustained period of time would make a huge difference in living standards and produce a substantial reduction in poverty (even if not reaching the IDG target). Seen from this perspective, the most important priority for combating poverty will be to sustain the recent performance, which will require focus on three key areas: maintaining macroeconomic stability, sustaining high income growth in rural households, and improving the provision of basic social services. Despite progress made to date, 55 the following section describes key areas where only limited advances have been made over the last decade. To avoid major gaps and imbalances in the poverty reduction effort, it is also necessary to broaden the poverty reduction strategy by additional measures (or by strengthening existing programs) that address these areas more effectively. B. SUSTAINING THE POVERTY REDUCTION MOMENTUM Ensuring Macroeconomic Sustainability 5.4 Perhaps the main challenge for sustaining the poverty reduction momentum achieved so far is to avoid a macroeconomic crisis and maintain macroeconomic stability so as to encourage private sector investment and growth. Nicaragua has advanced significantly since 1990 in establishing and consolidating macroeconomic stability. Most importantly, fiscal deficits have been reduced to levels that are compatible with single digit inflation rates. Thus, current trends would indicate that any crisis is unlikely to arise from domestic management problems. 5.5 Nicaragua's economy, however, remains very vulnerable to external and internal shocks, which could eventually lead to a crisis. It is particularly vulnerable to terms of trade shocks, especially on account of oil prices on the import side and coffee prices on the export side. Nicaragua does not produce oil and its electricity supply depends mainly on thermal generation. On the other hand, coffee accounts for about 30 percent of exports and the top three export products represent about 40 percent of total exports. While Nicaragua's export concentration is not unusually high by HIPC standards, it remains a significant source of vulnerability. 5.6 Nicaragua's continuing foreign aid dependence is another source of concern, especially in view of the expansion of aid inflows in the aftermath of Hurricane Mitch. Nicaragua already receives foreign aid equivalent to about one-quarter of GDP. In per capita terms, Nicaragua is among the top ten foreign aid recipients in the world. Under these circumstances, Nicaragua cannot expect to receive substantial increases in foreign aid that would translate into higher investment and growth. Prior to Hurricane Mitch, Nicaragua's aid dependence had declined somewhat and was replaced by an increased dependence on private capital inflows, which appear to be attracted by an improved business climate brought about by the economic reforms introduced since the early 1990s. Recent crises in Latin America and East Asia have raised awareness of the dangers of sudden private capital flow reversals in response to volatile financial market conditions. In Nicaragua, possibly because financial markets are still very underdeveloped, private capital flows do not appear to be as responsive to short term yield differentials as in other countries. (For example, when the East Asian crisis unfolded, the parallel exchange rate and interest rates exhibited no perceptible pattern, unlike in Brazil or Mexico.) On the other hand, foreign aid inflows have become increasingly driven by governance considerations which, if not fully addressed, may render them potentially more volatile. 5.7 To reduce Nicaragua's vulnerability to short term private capital inflows, policies to encourage longer maturity profiles and avoid currency mismatches in the banking sector should be considered. To reduce the threat of partial aid interruptions, it is important to advance steadily on the governance front, as well as on the overall reform program and implementation of an effective poverty strategy. On the other hand, reducing the vulnerability to terms of trade shocks may be more problematic. A more diversified export structure may reduce risk, but possibly at the expense of slower export growth, while reducing Nicaragua's dependence on oil imports faces technological constraints. In this case, policymakers may be better advised to focus on preparing a solid crisis management strategy and the creation of safety nets for the poor (see below), rather than making greater efforts, beyond basic precautions, to reduce the vulnerability to terms of 56 trade shocks. This recommendation also applies in response to Nicaragua's notable vulnerability to natural catastrophes. 5.8 Internally, the main source of vulnerability is the weak financial system, consisting of emerging private banks that did not exist a decade ago and an equally recent system of banking supervision. The best way to reduce this source of vulnerability is by strengthening the banking oversight capacity, and issuing and enforcing appropriate prudential norms. This is an extremely delicate task. Given the key role of the financial system in a market economy, heavy-handed restrictions on banking sector growth would be self-defeating. Sustaining the Provision of Basic Social Services 5.9 Nicaragua's health system needs to move from an input-driven system to a results- oriented system that places households at the center of the health strategy. * Households need to be enabled to care better for their own health-emphasizing public health and prevention messages targeted at the household level-since the provision of facility- based services alone has proven inadequate to achieve an impact on health outcomes. Most health prevention and promotion must occur at the household level-the costs of health care for the poor are such that by the time families reach the health facility, it is usually only to seek curative care. * Service delivery mechanisms need to be results-driven, rather than supply-driven. Existing incentive regimes result in the sub-optimal delivery of low-quality services. Health policies should not support further increases in inputs without effective mechanisms to transform these inputs into higher consumption of health services (prevention and promotion) by poor households. * The relatively higher price elasticity of demand for health services among the poor means that user fees should be limited to the non-poor, while targeted demand subsidies (community-based health insurance programs, health program vouchers, etc.) can have an important impact on demand by the poor. But to have an impact, demand subsidies must be matched by adev.ate supply capacity. - Health actions need to be better coordinated and targeted at the local and national levels for all health actors (MINSA, donors, NGOs, and private sector). 5.10 Attainment of these objectives will not require substantial modification of existing reform strategies; rather it requires their decisive implementation. Specific recommendations to achieve a results-oriented system that is capable of serving the poor involve reforming the input-driven model of health care organization in MINSA. MINSA is moving, under the ongoing modernization program, toward contract or performance oriented models (with rewards and sanctions) that will make providers more autonomous (concerning decisions regarding the level and mix of inputs) but also more accountable for their actions. Performance agreements and contracts that place the households at the center of health strategy-by specifying services to be delivered, negative behaviors to be addressed and populations to be covered-are an important ingredient. For example, within the contract/performance agreement model, MINSA is moving toward demand-driven supplier subsidies to target the specific needs of specific populations (i.e., high risk pregnant women residing in rural areas-FONMAT). Strengthening MINSA's normative role in the areas of quality assurance, regulation, information dissemination, coordination of health sector organizations and monitoring and evaluation is an important complement. 5.11 Key actions that can help attain these objectives over the short to medium term include: 57 * Extend the coverage of the poorest with basic health services (prevention, promotion plus primary level treatment) at the community level via private, NGO, or MINSA-contracted services in 12 priority primary health care catchment areas. An attainable goal for this effort would be to increase coverage of maternal and child services from an estimated 40-60 percent today to at least 80 percent. * To improve the health system's capacity to respond to local needs and conditions, put in place a fully defined and approved decentralization framework, in which the roles of MINSA, the SILAIS and hospitals are clearly defined and differentiated. This framework should be applied in at least one case. * To improve service quality and impact at the household level, implement performance incentive systems, with corresponding incentive pay schemes, in at least 10 primary health care catchment areas, 6 hospitals and the central level of MINSA. The first 6 hospitals, 3 MINSA-central directorates and 3 PHC catchment areas could adopt this new system during 2000. To achieve their goals, performance incentives should be results-based (improved household behaviors and health outcomes), rather than input-driven. * To support the extension of coverage and the improvements in service quality and impact, increase real per capita recurrent finance of primary health care services by approximately 25 percent per annum. This incremental finance should be directed to the poorest areas of the country, and should be used to contract NGO, private, or MINSA services under clearly specified performance agreements. * To improve MINSA's accountability and ability to respond to local circumstances, MINSA and MHCP should reach an agreement under which MHCP shifts to MfNSA the responsibility to plan, prepare, manage and execute its entire budget. * To create the legal foundation for a more efficient and modern health system, with better prospects to reach out and extend services to all communities, present and approve the proposed new General Health Law (Ley General de Salud). 5.12 In the education sector, addressing the problem of sustainability may require a different solution. The substantial gains in school attendance during the mid- 1990s reflect the considerable demand by families for schooling. To address concerns over the sustainability of these gains in the face of rising costs and long distances to primary schools, it is important to: - Target demand subsidies to the poorest to expand coverage and improve retention; * Selectively increase the provision of infrastructure and introduce alternative delivery mechanisms, such as distance learning to reach remote areas more cost effectively; and * Review the autonomous school model to ensure that the poor can implement and benefit from this approach, without unreasonably taxing their financial and managerial capacity. * Review the poverty implications of the National Education Plan to avoid reversing the gains achieved for the poorest by reducing the relative emphasis on primary and pre-primary education. 5.13 A fundamental policy implication for Nicaragua's education system is the need to target the poorest zones and poor rural villages, where unqualified teachers, inadequate supply of teaching materials, and poor infrastructure are considerably more severe. Thus, disproportionately greater support needs to be given to schools where the poorest families may send their children. Clearly, raising opportunities for learning in these schools would provide the poor with greater incentive to complete schooling (where expected benefits of schooling would exceed direct and opportunity costs). To improve education in Nicaragua, the ongoing Aprende project has four basic elements: curriculum reform, increased provision of school books, school 58 rebuilding, and incentives and standards for teacher qualification and performance. Actual implementation still needs to ensure that these strategies are integrated and targeted to the poor. 5.14 In Nicaragua the school autonomy model is currently in operation, but achieving its full benefits is still closely related to substantial improvements in accountability and local participation. The reality is still far from ideal. There is very little accountability partly because expectations and requirements are not specified, informnation on performance of teachers and schools is not available, and, even in the autonomous schools, parents and staff do not understand their authority and responsibility. Research in Nicaragua' indicates that schools making more decisions themselves (as opposed to the decision being made by regional or central authorities) present significantly higher achievement rates. However, the Government has yet to specify how (and whether) it intends to make all schools autonomous, and by when. The success of school autonomy is dependent upon the mobilization of additional resources at the school which is more difficult in poor, rural areas.2 Since probably the remaining schools to be brought into the autonomy process are precisely the poor, rural primary schools, a specific plan to ensure that they can function adequately, in terms of resources, training, etc., is urgently needed. 5.15. Specific recommendations to help ensure that the school autonomy model can benefit the poor include: * Increase training for parents, directors, teachers, and students on school autonomy. Although the Aprende project has allocated significant resources to this activity, training must be continuous, since it is based on the concepts of local governance and accountability, both of which are new to education stakeholders. * Invest in school directors. Autonomy works well where there is a good director. Since school directors are the main link between parents and teachers, their leadership and managerial skills are crucial to the success of the model. MECD needs to accelerate the identification and training of teachers with leadership qualities more appropriate for a director's position by: (i) training directors in management and communication in order to give them the tools with which to work with parents; and (ii) maintaining and reinforcing the current process of selection of school directors to ensure the application of merit based criteria. * Reform the school supervision system to better use MECD's municipal delegations to support the proper functioning of autonomy. The Aprende project supports efforts to reform the role of these delegations in providing the technical and managerial expertise to help school directors improve performance. 5.16 To enable the education system to better target resources to the poorest and to improve school efficiency, accountability and quality via the school autonomy program, Nicaragua's education system needs to adjust its financial resource allocation: * Place less emphasis on adult literacy. If total budgets are not to increase substantially, MECD should avoid over-funding literacy relative to primary school funding. If too much attention is paid to adult literacy MECD runs the risk of increasing illiteracy among the young, who in turn would become illiterate adults later on. * Review the policy on vocational education. Reconsider the privatization of INATEC, which consumes a disproportionate amount of resources relative to its results. Eliminate the 2 percent tax charged to formal businesses to support INATEC, and perhaps consider tax credits for training expenditures at accredited institutions in its place, or use the current tax to 'King, Elizabeth and Berk Ozler. What's Decentralization Got to Do with Learning? The Case of Nicaragua's School Autonomy Reform. 2 Bruce Fuller and Magdalena Rivarola. 59 finance a voucher system combined with mass information campaigns on training costs and market salaries for selected vocational careers. * Reexamine higher education finance. Under the constitution, universities receive 6 percent of the national budget. This is not only a disproportionate allocation for higher education, it is a highly regressive allocation which primarily benefits the highest income groups. At a minimum, the Government should seek to shift this finance from a subsidy to the supply side (the universities) to a merit-based demand subsidy (scholarships), and complement this effort with the development of education credit markets, and a student voucher system. To achieve this, the Government will need to begin a process of study, and dialogue with the university system to determine the implementation characteristics of such a shift. * Implement a system to measure and report student achievement. This is an important step to provide information as to the cost effectiveness of various education investments, as well as to establish a system of academic accountability. This activity could give substantial leverage to other reforms, such as improved teacher training and reinforcing school councils. * Modernize the existing career tracks for teachers and implement a new system with higher salaries tied to significantly higher training requirements. The new system should also include efforts to attract better talent to the teaching profession with a voucher system or scholarships tied to academic performance and future teaching contracts. * As in the case of health, supporting the extension of coverage and the improvements in service quality and impact would require the Government to commit to significant increases in per capita recurrent finance of primary education services. This incremental finance would need to be directed to the poorest areas of the country and improvements in remuneration linked to upgrading of teacher qualifications. Sustaining Rural Income Growth 5.17 Nicaragua's strong agricultural growth during the mid-1990s contributed importantly to the decline of extreme poverty, especially in the Central region. A closer examination of agricultural performance, however, reveals that this may not be easy to sustain, for two reasons. First, sector performance was driven by unusually favorable export prices, especially coffee. Second, the expansion of agricultural production from 1993 to 1998 primarily reflects increases in labor and land inputs, and only small increases in yields. Land and labor inputs increased in the wake of pacification of the countryside, and the resettlement of former combatants and refugees. This led to increased labor participation rates in rural households, raising household incomes as more income earners contributed to the household budget, and offset the decline in real agricultural wages. Similarly, increased land use reflects the expansion of the agricultural frontier and progress in settling rural property disputes. These factors suggest that the agricultural recovery in the 1990s is primarily based on a one-time shift in underlying factors that are unlikely to persist. 5.18 As the pool of idle rural workers is exhausted, labor force growth in the agricultural sector is bound to decline, while a further expansion of the agricultural frontier would be environmentally unsustainable. Nor can we expect export prices to continue rising at the same pace as before. These factors point toward a slowdown in agriculture growth, with negative consequences for poverty reduction. The only way to sustain agricultural growth and incomes would be by: (i) boosting agricultural productivity; and (ii) improving incentives for non- agricultural rural activities, possibly through improved rural productive infrastructure and better rural credit mechanisms targeted to small producers. In either case it is important to boost economic activity in rural areas to prevent migration to increasingly congested urban areas. 60 5.19 There are two areas of opportunity for boosting agricultural growth in Nicaragua- accelerating technological advance and reducing the remaining anti-export bias in the sector. * Nicaragua's agriculture has very low levels of technology, reflected in low productivity gains. Over 1979-97, the sector shows stagnant or declining trends in value-added per agricultural worker, fertilizer use per hectare and the capital-labor ratio (tractor use per worker) per arable land. While the shift in labor supply since the early 1990s has disguised the impact of this productivity decline on household incomes, a strategy to sustain agricultural income growth must focus on accelerating improvements in agricultural technology. In addition to making more physical capital available, it will also require raising rural education levels in order to raise the capabilities of rural residents to gather and assimilate information about technology, production and marketing alternatives. * Although significant progress was made in the 1990s in reducing trade barriers, there remains a bias against the agricultural export sector, with an estimated effective protection rate of 0.90, versus 1.36 for agricultural importables (basic grains). Further tariff adjustments may reduce this bias and, thereby, stimulate increased production in crops normally characterized by higher productivity growth. In this context, it is important to avoid any major appreciation of the real exchange rate, but reform has also been incomplete. During 1992-98, the price of import-competing goods grew more than exportable agricultural products (import goods are relatively more costly than export products). The incentive to producers, therefore, is to produce import goods (basic grains) rather than export goods. In addition, the incentive structure in agriculture (tariffs and other distortions) is biased against exports. The sum of these two factors results in a strong bias against exportable agricultural goods. 5.20 Reconciling the need to improve living standards while reducing environmental threats presents a particular challenge in the Atlantic region. Due to fragile soils, the region has little agricultural potential and its economy has been based on limited exploitation of its natural resources. Migration into the Atlantic region and expansion of the agricultural frontier, however, are placing increasing pressure on its fragile resource base and biological riches. There is growing concern that the current patterns of frontier development in the Atlantic are unsustainable, non-economic, and the source of increasing ethnic tensions, and social and environmental problems. Of particular concern is the vulnerability of indigenous communities in the region. There is inadequate control over logging and mining operations, natural resource planning and management of protected areas are weak, the legal and institutional framework needs to be improved, and there is insecurity of land tenure and access to natural resources. While the Atlantic region accounts for a relatively small share of Nicaragua's poor, its rich biological resources and extremely high poverty suggest the need for increased emphasis on sustainable development and special attention to the development needs of indigenous communities. Enhancing Non-farm Rural Activities 5.21 The least poor in rural areas are households which are heavily engaged in the non- agriculture sector. In this way, non-agricultural employment may be seen as a route out of poverty. The challenge is thus to increase access of the poor to non-farm activities which yield high and stable incomes. The analysis in this report suggests focusing on rural education and highly selective infrastructure investments, but the role of rural credit and sub-contracting arrangements can also be important. 61 * There is strong evidence that higher incomes in non-agriculture jobs go to those with higher levels of education. While it is not the case that the uneducated are not able to secure any non-agricultural employment, the kinds of jobs tend to be low-productivity, which while helpful as a coping strategy do not substantially raise the incomes of those households. * Access to infrastructure appears to have a significant influence on the likelihood of finding non-farm employment, although causality is difficult to establish. The policy implications, however, need to be approached with care. Providing rural infrastructure is more costly than in urban settings. This suggests two important policy implications. First, the focus of infrastructure providers should not necessarily be on the provision of highest quality (and hence high cost) infrastructure, but rather on a basic package of simple or even rudimentary infrastructure services. Second, the infrastructure investments should be preceded by a careful and detailed analysis of social and economic rates of return in the targeted areas. These rates of return should be used to prioritize investments and to evaluate the opportunity costs of investing scarce fiscal and managerial resources in other programs that may also have a high poverty or economic impact. In other words, a poverty focussed program of investments in rural infrastructure should be fully integrated into a well articulated and prioritized public investment program, consistent with the Government's fiscal capacity. * Improving access to a viable and sustainable rural credit system should also be a priority. An improved rural credit delivery system should adopt a broader focus than agricultural activities and should also encompass the financing of small-scale non-farm rural enterprises. - Nicaragua has considerable experience with maquila enterprises but there is no information on whether sub-contracting arrangements exist. A quick survey of maquila firrns could provide information on whether these arrangements exist and some guidance on a minimal package of infrastructure, training and finance that could encourage this practice. Facing the Demographic Challenge 5.22 One major possibility for improving economic welfare lies with reducing high population growth. At around 2.7 percent, Nicaragua's population growth is very high by international standards. A reduction in average fertility rates by, say, half, would represent a major achievement. This would only have a relatively modest impact on poverty in the short and medium term, however, unless the distribution of that population growth across households also changed simultaneously. That is, if the overall fertility decline were to take place mainly through lower fertility rates among the poorest households, it would lead to a vastly faster reduction in overall poverty levels, for any given income growth rate. 5.23 A recent cross country study3 of the impact of fertility on economic growth and poverty suggests that a sustained decline of 4 per 1,000 in the net birth rate4 over a decade could reduce 3The study by Eastwood and Lipton (1999) used national household survey data for 59 countries. Causality tests confirmed that the causality runs from declines in birth rates (lagged ten years) to GDP growth, and not the reverse. Growth and distribution impacts on poverty of fertility declines were found to be higher in high-fertility low-income countries, and lower in high-income low-fertility countries. By these criteria, Nicaragua is classified as a high-fertility low-income country. 4The authors define the net birth rate as the crude birth rate (normally expressed as live births per 1,000 population) minus the infant mortality rate (normally expressed as deaths per 1,000 live births); that is, NBR=CBR-IMR/1,000. Given the relatively low infant mortality in Nicaragua, this refinement over the use of the crude birth rate has no impact. 62 the incidence of poverty by 6.5-7 percentage points in a country with Nicaragua's income and fertility characteristics. This impact is achieved by two roughly equal effects: a direct growth effect on GDP, and a distribution effect due to declining household dependency ratios and increased per capita household consumption. 5.24 Recent declines in birth rates5 suggest that Nicaragua has an opportunity to reap benefits of future GDP growth and declining dependency ratios over the next 40 years, if it can sustain and accelerate these declines. Already, the evidence shows that smaller families live better. The dependency ratio among the poorest quintile (1.2) is twice that among the richest quintile (0.6). Achieving these gains, however, will require focusing on the links between poverty and fertility, with special attention to the needs of adolescents. 5.25 Nicaragua's high fertility rates represent a major constraint on poverty reduction. Although the demographic transition has begun, fertility declines are concentrated among the non-poor. Consolidating the benefits of the demographic transition will require greater efforts to include the poorest, with special emphasis on adolescents. Failure to do so will have direct poverty consequences, manifested in growing income inequality, reduced prospects for growth- led poverty reduction, concentration of children in poverty and heightened reproductive risks. 5.26 Expanded access to family planning services continues to be a high priority, but access alone is not enough. Household factors are enormously important, including family behavior patterns and the aspirations of young women in Nicaragua's shattered society. The poorest adolescent girls are much less likely to be in school than their non-poor counterparts, which reduces their opportunities outside the home and is directly linked with increased adolescent fertility. In addition, adolescent girls, and particularly those from violent homes, appear to choose maternity as a the only opportunity they see to escape, gain attention or improve their social status.6 5.27 Meeting the challenge to capture the potential human and economic benefits of fertility reductions will require a decisive strategy involving both the Government and as broad a base of civil society as possible. Important elements of a population strategy are: * setting up a public information campaign, * improving outreach of reproductive health and counseling services, * expanding opportunities for adolescents, including participation in community activities, sports and social events, * improving labor market opportunities for women, * keeping children in school longer, and * reducing family violence. Developing Better Household Nutrition and Hygiene Practices 5.28 Malnutrition continues to be high, is geographically concentrated and appears to have changed little in the last decade, despite substantial spending on nutrition programs during the 1990s. Malnutrition indicators typically are closely related to overall consumption levels. In Nicaragua, however, some groups within extremely poor communities exhibit relatively low malnutrition indicators, while child malnutrition is also observed in non-poor communities. This 5 Net birth rates have been declining in Nicaragua, but at a slower rate in recent years: 1975-80: 46; 1985- 90: 38; 1995-2000: 36. From the late 1970s to the late 1980s the decline was 8 per 1,000, whereas from the late 1980s to the late 1990s the decline was only 2 per 1,000. 6 Zelaya (1999) and FNUAP-INIM (1999). 63 suggests that the problem of malnutrition is largely a function of behavioral patterns as well as a matter of food availability or income and disease. This finding points toward possible opportunities for achieving significant improvements in malnutrition, without having to wait for major income changes or expansions in the supply of water and sanitation services. To exploit that opportunity, different types of interventions may be required to modify household behavioral patterns. 5.29 Proposed elements of a strategy to address chronic and acute malnutrition, and to reduce micronutrient deficiencies, include: * Reorient existing nutrition programs to improve impact. Existing nutrition programs are not sufficiently targeted on the neediest; do not deliver a service that promotes a sustainable change in nutritional status; do not define impact goals, define or use monitoring and evaluation mechanisms; and do not coordinate among themselves. * At the national level, program reformulation should emphasize better entry and exit criteria based on consistent targeting criteria across programs, use of evaluations of cost effectiveness and sustainability of different programmatic approaches to guide investment decisions, systematic impact monitoring and evaluation, and improved program coordination. Improve efficiency of existing nutrition programs through the definition of consistent targeting criteria across programs, including individual entry and exit criteria (e.g., child growth). With the new poverty map and proxy means test methodologies, it will be possible to go beyond municipal level targeting to community or household level targeting, which can be combined with individual criteria to improve program efficiency. * At the local level, programs should help families manage their own nutrition by: providing clear information to families on the growth trends of their children and offering feasible recommendations for children's care; supporting activities at the community level to address common obstacles to improved nutrition; coordinating program and service delivery to reach families; and regularly discussing with families progress toward jointly defined goals. It is important to provide adequate nutritional counseling at the household level to reverse ingrained negative nutrition practices, and ensure that the benefits of improved household food security translate into improved nutrition. * Parallel investments in expanding access to safe water and sanitation, improving female education and reducing dependency ratios are important complements to this strategy. * Expand quality control and coverage of fortification programs, and continue promoting the consumption of foods rich in micro-nutrients. Reduce the consequences of micronutrient deficiencies, guarantee universal iron fortification of wheat flour and wheat products, and ensure adequate quality control/regulatory enforcement of all food fortification programs. Also guarantee universal and adequate iron supplementation for pregnant women and children under two years, with appropriate counseling. Improving Gender Symmetry 5.30 The high incidence of domestic violence in Nicaragua may be indicative of the systematic social exclusion of women. Violence against women is pervasive and in some ways accepted by society, often regarded as "a woman's cross to bear in life."7 Beyond reducing the quality of life, household violence also represents a direct obstacle to human capital development and economic growth. One way of addressing this problem is to reduce poor women's isolation and vulnerability by integrating them more effectively into economic life. This would render them less dependent on abusive environments and in a better position to develop a support 7 Ellsber, et. al.(1996). 64 network. Although gender equity already exists in terms of primary education coverage, it is revealing that the estimated rate of return to primary education in 1998 is less than half as high for women as for men (6 percent versus 14 percent), which suggests that the opportunities for engaging in remunerated economic activities are significantly more limited for women than for men. Measures to rebuild Nicaragua's social capital and to support the family (community organization for development, religious, sports and cultural purposes) along with public messages to promote positive role models will be fundamental to improve gender symmetry. In consequence, strategic recommendations are: * Support sustainable changes in negative behavior acting through community and local level organizations. Decentralized mechanisms and community based organizations have comparatively more advantages than the central government to address sustainable change in negative behavior, such as domestic violence, violence among young men, substance abuse, family disintegration, and teenage pregnancy. * Promote changes in gender-based social expectations by acting upon traditionally held gender roles so that both women and men can take advantage of better opportunities, particularly more schooling. Different channels can be used to promote these changes, such as the media, education, and the community. To this aim, projects, in particular in education, community development, judicial reform, social investment funds, rural municipalities and land registration, should define means that will help to desegregate gender roles in their respective areas. - Target males as well as females for three reasons: (a) because gender is a relational category and gender issues influence at the same time both members in a relationship; (b) because males also experience gender-derived losses of opportunities, such as violent behavior, alcoholism, deprivation of education; and (c) because effective women's programs require men's explicit coo?eration and vice versa. - Culturally adjusted interventions should target indigenous communities on their gender disparities and inequities since many of the progresses felt at the level of national societies have not been extended to these communities. Establishing Effect-, e Social Protection Mechanisms for the Poor 5.31 Nicaragua's social protection strategy needs to be more fully developed. With regard to existing programs, the Government's commitment in the social policy matrix presented to the 1998 Geneva Consultative Group for Nicaragua to rationalize the existing social assistance portfolio does not seem to have advanced significantly yet. An important first step would be to prepare an updated evaluation of existing institutions and programs. Evaluations should consider program targeting, and the cost-effectiveness of the program options selected versus design alternatives. 5.32 Looking to the future, Nicaragua needs to define a more comprehensive and coherent strategy. So far, the approach has been largely to provide social assistance, which is reasonable, given Nicaragua's income and poverty characteristics. In practice, however, many programs lack a consistent focus or targeting approach and, consequently, resources may have been spent without achieving a significant impact. The Government needs to define its approach to social protection in Nicaragua: What are the priority target groups? What is the appropriate role and mix of safety nets, social insurance, etc.? Some guiding elements in this process should include: * A detailed classification of the most vulnerable and the nature of their vulnerability. For example, children under 5 are the most over-represented in poverty and appear a priority for social protection. 65 * Definition of a clearly defined benefit tailored to address the vulnerability of a particular group, which may entail direct transfers in some cases and access to insurance markets in others. * Targeting mechanisms that direct programs to the intended beneficiaries. In Nicaragua, there exist greater inequalities within municipalities than across municipalities, so targeting by broad geographic area may not be appropriate. Better targeting may be attained using the new poverty map, which can permit targeting at the sub-municipal level. Targeting mechanisms must include both entry and exit criteria. * Given the wide variety of programs in place, as well as the number of new initiatives, developing a well-designed evaluation system is especially important in the social protection area. 5.33 In the short-to-medium term, the social protection strategy could: * Continue supporting rehabilitation/expansion of basic health and education infrastructure, and stepping up efforts regarding water and sanitation focusing on including the poorest groups; * Complement supply side interventions in health, education, nutrition with demand side interventions (with programs like the Red Social); * Strengthen local and municipal capacity so as to increase their ability to solve the problems of their communities; and * Support groups that are objectively identified as especially vulnerable such as the zero to five years age group, street children, the disabled, etc. 5.34 In addition, increasing attention is needed to develop more effective ways to prevent and mitigate the effects of crises and natural disasters, to develop more effective safety nets and, over the longer term, to focus on developing more effective social insurance mechanisms. If these are agreed priorities for social protection, then the role of existing programs, the scope for their reform/reformulation, and the cost savings or additional resource requirements would need to be identified. References for Main Report Arcia, Gustavo. 2000. Education and Poverty in Nicaragua. Background Paper for the Nicaragua Poverty Assessment 2000. The World Bank Arcia, Gustavo and Vanessa Castro. 2000. Rural Education Experiences. Background Paper for the Nicaragua Poverty Assessment 2000. The World Bank Arredondo, A., and J.S. Chamorro. 1996. Cuentas Nacionales en Salud FUJNSALUD-MINSAL-Banco Mundial. Budinich, Ema. 2000. Public Social Spending and Programs by Sector in 2000. Background Paper for the Nicaragua Poverty Assessment 2000. The World Bank Banco Central de Nicaragua (BCN), 1998. Indicadores Economicos. Managua. Chawla, Mukesh . 2000. Malnutrition among Pre-School Children. Background Paper for the Nicaragua Poverty Assessment 2000. The World Bank . 2000. Demandfor Health Care. Background Paper for the Nicaragua Poverty Assessment 2000. The World Bank. Consorcio FUNSAGLAS. 1997 Estudio del Mercado Laboral en Salud en Nicaragua. Proyecto MINSA- Banco Mundial. Managua. Dayton, Julia. 2000. The Incidence of Public Expenditures on Health and Education. Background Paper for the Nicaragua Poverty Assessment 2000. The World Bank Davis, Benjamin and Murgai Rinku. 2000. Between Poverty and Prosperity: Rural Households in Nicaragua. Background Paper for the Nicaragua Poverty Assessment 2000. The World Bank. Demombynes, Gabriel. 2000. A Short Guide to Poverty and Inequality Measures. Background Note for the Nicaragua Poverty Assessment 2000. The World Bank Eastwood, Robert and Michael Lipton. 1999. "The Impact of Changes in Human Fertility on Poverty". Journal of Development Studies, vol. 36., No. 1, October, 1999 pp. 1-30. Ellsber, Mary et. al. 1996. Confites en el Infierno. Prevalencia y caracteristicas de la violencia conyugal hacia las mujeres en Nicaragua. Red Nacional de Mujeres contra la Violencia, Nicaragua. Departamento de Medicina Preventiva y Salud Nblica, UNAN-Le6n. Departamento de Epidemiologia y Salud Publica, Umed University, Suecia ESDENIC-85 FNUAP-INIM, 1999. "Que mas podia hacer, sino tener un h#o: bases socioculturales de las adolescentes en Nicaragua" Gillespie, Nancy. 2000. Population, Health and Nutrition. Background Paper for the Nicaragua Poverty Assessment 2000. The World Bank GonzAlez, Mary Lisbeth. 2000. Nicaragua Qualitative Poverty and Exclusion Study. Background Paper for the Nicaragua Poverty Assessment. The World Bank. Hausmann, 1999. Ilahi, Nadeem. 2000a. Headship, Gender and Poverty in Nicaragua. Background Note for the Nicaragua Poverty Assessment 2000. The World Bank. . 2000b. Labor Markets and Poverty Reduction. Background Paper for the Nicaragua Poverty Assessment 2000. The World Bank. Kozel and Parker. 1999. Poverty in Rural India. Draft. Kruger, Diana. 2000a. Rates of Return to Education, Background Paper for the Nicaragua Poverty Assessment 2000. The World Bank. . 2000b. Determinants of Primary School Attendance. Background Paper for the Nicaragua Poverty Assessment 2000. The World Bank. . 2000c. Redistribution Effects ofAgricultural Incentive Policies. Background Paper for the Nicaragua Poverty Assessment 2000. The World Bank. L§chler, Ulrich. 2000. A Growth Accounting Approach. Background Note for the Nicaragua Poverty Assessment 2000. The World Bank. MAGFOR. 1999. Gasto del Sector Pzuhlico Agropecuario en Nicaragua. October. Managua. MINSA. 2000a. Andlisis Sectorial del Sector Salud; Project Report for the Inter-American Development Bank Project, Nicaragua Support for MINSA Hospital Modemization, (NI-0024). MINSA. 2000. Analisis del Sector Salud, MINSA/OPS, October. Managua Nicaragua LSMS 1998 Nicaragua LSMS 1993 PROTIERRA. 1999. (PROTIERRA). Diciembre 1999. Evaluaci6n de Efectos y Sostenibilidad PotenciaL Nicaragua, Proyecto Municipios Rurales. Draft. Reproductive and Child Health Project, UNAN-Le6n Sanchez, Susana. 2000. Financial Markets. Background Paper for the Nicaragua Poverty Assessment 2000. The World Bank. Seireg, Mirefte, Craig Loftin, Dagoberto Rivera y Patricia Carrillo con Jose ArgUiello. 1999. Estudio de Fortalezas y Obstaculos que Enfrentan los Programas de Nutrici6n y Alimentaci6n en Nicaragua. Consultoria para el Banco Mundial. Schiff, M. and A. Valdes (1992) The Political Economy of Agricultural Pricing Policiy in Developing Countries, Volume IV: A Synthesis of the Economics in Developing Countries, John Hopkins University Press, Baltimore, Maryland. Sobrado, Carlos. 2000a. Comparing Poverty for the Young and the Old. Background Note for the Nicaragua Poverty Assessment 2000. The World Bank. . 2000b. The Determinants of Poverty in Nicaragua. Background Note for the Nicaragua Poverty Assessment 2000. The World Bank. Sobrado, Carlos, Berk Ozler, Juan Rocha and Dulce Maria Mayorga, with Martha Vargas, Transito G6mez, Joaquin Murillo and Luis Alaniz. 2000. Nicaragua Poverty Map. Background Paper for the Nicaragua Poverty Assessment 2000. The World Bank. UNICEF. 1999. UNICEF Nutrition Assessment. Narrative Paper for the World Bank. How Nutrition Improves, and What that Implies for Policy Decisions. USAID. 2000. Current Nutritional Programs in Nicaragua (information submitted to the World Bank for the study on Gasto Puiblico Social y Programas por Sectores en el ano 2000) World Bank. 1992. Nicaragua Public Sector Expenditure Review. September 11, Report No. 10785-NI, Washington DC. World Bank. 1995a. Republic of Nicaragua Poverty Assessment Report. June 1, Report No. 14038-NI. Washington DC. World Bank. 1995b. Priorities and Strategies for Education: A World Bank Review, Washington DC, 1995. World Bank. 1997. Public Investment Review. May 29, Report No. 16478-NI World Bank. 1998. Project Appraisal Document for the Health Sector Modemization Project, Credit # 3084-NI, Report No.17609-NI (1998) World Bank (1999a). PAD Second Basic Education Project. Report No. 191560-NI World Bank (1 999b). World Development Indicators World Bank. 2000. Nicaragua. Ex-Post Evaluation of FISE. May, draft. Wodon, et. al. 2000. Poverty and Income Distribution in Latin America and the Caribbean, LAC Region, World Bank, Washington, DC. Zelaya B., E. 1999. Adolescent Pregnancies in Nicaragua, the Importance of Education, Umea University Medical Dissertations, New Series No. 639 - ISSN 0346-6612, Epidemiology, Department of Public Health and Clinical Medicine, Umea University, Sweden, and Department of Preventive Medicine, Nicaraguan Autonomous National University, Le6n, Nicaragua.