Colombia - Social programs and poverty alleviation : an assessment of government initiatives

Report No. 7271-CO Colombia Social Programs and Poverty Alleviation: An Assessment of Government Initiatives December 2,1988 Latin America and the Caribbean Region FOR OFFICIAL USE ONLY Document of the Wbrld Bank Tnis document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALZNTS (as used in this report) Currency Unit - Colombian Pesos (Col$) US$1 - Col$250 VEIGBTS AND MEASURES Metric System GOVERNMENT OF COLOMBIA FISCAL YEAR January 1 - December 31 This report is based on the findings of two World Bank missions that visited Colombia in November 1987 and March 1988. The two teams comprised Messrs/Mesdames. Kazuko Uchimura (LA3C1, mission leader), Robert Buckley (INU, housing finance), Xavier Coll (LA3HR, health and nutrition), Nancy Gillespie (PHR, maternal health), Fernando Vio (consultant, nutrition), Anna Kathryn Webb (consultant, community participation), and Stuart Whitehead (LA3IE, low-income shelter). Ms. Kathy Richman (consultant) carried out a survey of the existing literature on poverty and income distribution in Colombia. The report has also benefited from the advice and guidance of Mr. Marcelo Selowsky who participated in the November mission.  FOR OFFICIAL US ONLY PRINCIPAL ACRONYMS AND TERMS USED IN THE REPORT BCH Central Mortgage Bank ("Banco Central Hipotecario") CAVs Housing and Loan Associations ("Corporaciones de Ahorro y Vivienda") CENAC National Center for Construction Studies ("Centro Nacional de Estudios de la Construccion') DANE National Statistical Department ('Departamento Administrativo Nacional de EstadistUca") DNP Natlonal Planning Departmer:t ('Departamento Nacional de Planeacion") DRI Inte!rrated Rural Development Program ("Desarrollo Rural Integrado") FPDU Financial Fund for Urban Development ("Fondo Financiero de Desarrollo (rbano") FIA liational Savings Fund ("Fondo Nacional de Ahorro") lBI Community-based Day Care Centers ("Hogares de Bienestar Infantil") ICBF Colombian Institute of Family Wellbeing ("Instituto Colombiano de Bienestar Familiar") ICT Land Credit Institute ("Instituto de Credito Territorial") ISS Social Security Institute ("Instituto de Seguros Sociales") NBI "Basic-Needs-Not-Met" Index (a Poverty Measure) ("Necesidades Basicas Insatisfechas") PAN National Food and Nutrition Plan ("Plan de Alimentacion y Nutricion") PLPGE "War on Poverty" ("Plan de Lucha contra la Pobreza y para la Generacion de Empleo") This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  - 2 - PNR National Rehabilitation Plan ("Plan Nacional de Rehabilitacionn) SIM& National Vocational Training Institute ("Servicio Nacional de Aprendizaje") Supervivir The National Child Survival and Development Plan UPAC Unit of Constant Purchasing Power ("Unidad de Poder Adquisitivo Constante") "Vivienda A Low Income Housing Program under the Betancur Popular' Administration  TABLE OF CONTEWNtS Page EXECUTIVE SUMMARY ix CHAPTER ONE: INTRODUCTION 1 Transformation of the Colombian Economy 1 Persistence of Poverty 1 Government's Commitment to Social Programs 2 A Macroeconomic Framework for Poverty Alleviation 3 Priorities in Public Expenditures 4 Limited Absorptive Capacity in the Social Sectors 6 CHAPTER TWO: THE DIMENSIONS OF POVERTY IN COLOMBIA 7 A. Measurement of Poverty 7 NBI--Government's Official Poverty Index 7 Recommendations 8 B. Spatial Distribution of Poverty 9 Households "In Misery" as Defined by NBI 9 Access to Public Services 10 Economic Dependence Ratio 15 C. Characteristics of the Poor 15 Poverty and Malnutrition 15 Inadequate Housing and Services 20 Education and Literacy 20 Employment 20 D. Recent Changes in Income Distribution 20 CHAPTER THREE: AN AGENDA FOR ACTION 23 The Government's Action Program 23 General Comments on New Basic Needs Initiatives 24 CHAPTER FOUR: A NEW APPROACH TO NUTRITION INTERVENTION 29 A. Community-based Day-Care Centers 29 B. Financing of Nutrition Programs 31 C. Recommendations on Program Design 33 Coverage of Children Between 6 and 24 Months 34 Nutrition Interventions in Rural Areas 34 Treatment of Serious Malnutrition Cases 37 Role of Land Credit Institute ('CT) 37 Fostering Community Support 37 D. Policies on Bienestarina 38 Securing Adequate Supply 39 Distribution and Storage 41 E. Information System 41 Absence of Surveillance 41 Work in Progress Recommendations on Establishing a Surveillance System 42  CHAPTER FIVE: STRENGTHENING INTEGRATION WITH THE HEALTH SYSTEM 43 Coverage and Access 43 Problems with Primary Health Delivery System 44 Government Proposals on Health and Bank Recommendations 46 CHAPTER SIX: MEETING THE SHELTER NEEDS OF THE POOR 47 Background 47 Government's Policy on Low Income Housing 50 Main Issues 50 Size&and Feasibility of Implementation 55 Role of Government in Low Income Shelter Provision 56 Agua Blanca--a New Approach to Slum Upgrading 59 Use of Interest Subsidies 61 Priorities for Government in Low Income Shelter 61 CHAPTER SEVEN: BASIC EDUCATION FOR ALL 63 A. Background 63 Prevalence of Illiteracy and Low Educational Attainment 63 Government's Response 66 B. Univer:sal Primary Education 68 D..;ssemination of the "Escuela Nueva" Model 68 Assessment 69 Improving Access in Urban Areas 69 Scbool Feeding Program 70 C. Continuing Basic Education 71 ANNEX I: SOCIAL PROGRAMS OF PAST ADMINISTRATIONS 73 National Food and Nutrition Plan (PAN)- Integrated Rural Development Program (DRI) 73 National Rehabilitationi Plan (PNR) 75 Low Income Housing ("Vivienda Popular") 75 National Child Survival and Development Plan ("Supervivir") 77 ANNEX II: THE EVOLUTION OF THE WAR ON POVERTY (PLPGE) 79 Its Origin 79 Evolution of Program Content 80 REFERENCES 83  TEXT TABLESICRARTS Page Table I-ls Macroeconomic Framework 4 Table I-2: Public Investment Program 5 Table 1-3t Public Expenditures on Principal Poverty Programs by Sector during 1987-1990 6 Table II-ls Distribution of Household Incomes and Expenditures on Food by Income Strata 15 Table II-2s Calorie and Protein Intake Per Adult Equivalent by Income Strata 16 Table II-3: Distribution of Household Incomes 1976-1985 21 TAble IV-1: ICYoF--Projectecl Revenues and Expenditures 1988-1990 32 Table IV-2: Assumptions Regarding Program Costs and Number of Beneficiaries 33 Table IV-3: A Comparison of Bienestarina with Other Foods in Terms of Nutrients and Calories 39 Table IV-4: Bienestarina Requirements through 1990 40 Table V-ls Patient Consultetions by Distance to the Closest Health Post 44 Table V-2: Hospitalization Rates by Region, Income Levels and Degrees of Urbanization 197o,-1990 45 Table VI-1s Relative Shares of Legal and Illegal Housing 48 Table VI-2s Low-Ircome Shelter Plaa 50 Table VI-3: Anticipated Financing Sources for the Low- Income Shelter Plan 1987-1990 54 Table VII-l: Changes in Illiteracy Zqwe among Population 10 Years and Older 63 Table VII-2s Illiterate Population 12 Years and Older by Age Group 66 Table VII-3: Share of 'Basic Education for All' Program in Total Education Sector Expenditures 68 Table VII-4: School Feeding Programs 71 Chart I: New Initiatives in Poverty Alleviation--A Summary of Recommendations 26  HAPS Page Colombia: Househol's "Con Miseria" 11 Colombia: Households without Basic Services 12 Colombia: Death Due to Intestinal Infection 13 Colombia: Households with High Economic Dependence 14 Colombia: Physically Inadequate Housing 49 Colombia: Level of Public Services -- Households with Sewers 51 Colombia: Level of Public Services -- Households with Electricity 52 Colombia: Level of Public Services -- Households with Piped Water 53 Colombia: Households with Children Not Attending School 64 Colombia: Illiteracy 65 Colombia: Hultisectoral Development Programs in Rural Areas 86 Colombia: Spatial Distribution of Poverty by Jurisdiction 87 FIGURES Figure 1: Characteristics of Low-Income Housing 17 Figure 2: Employment Among the Poor 18 Figure 3: Employment of Low-Income Heads of Families 19 Figure 4: Coverage of Hogares de Bienestar Infantil Program 30 Figure 5: Primary School Enrollment Rate by Departments 67  KCUTm SUDIRY Persistence of Poverty Amid Growth 1. Colombia has experienced dramatic changes in its economy and society since the 1960s. The economy has grown steadily and has become progressively diversified. helped in no small part by the country's rich natural resource base and the Government's sound economic management. Furthermore, Colombia, once a predominantly agrarian society, became highly urbanized as rural-urban migration proceeded at a rapid pace In the 1960o and 19708. This process was accompanied by dramatic changes in the country's demographic characteristics--steady declines in fertility and population growth rates as well as in the incidence of infant and maternal mortality. On the educational front, primary school enrollmentAdoubled during this period and the illiteracy rate among those 10 years and older dropped significantly. 2. Serious problems of poverty have persisted throughout this period of sustained economic growth and social change. Income distribution remaina skewed with incomes of the top 202 of households six to seven times those of the bottow 202. Moreover, impressive gains in the social indicators bely the sharp differences in health and educational status among socio-economic groups--fertility is estimated to be three times as high among the poor as the non-poor, and the higher fertility among poorer women is accompanied by shorter birth intervals and higher infant mortality. A high proportion of children from poor communities still have no access to any kind of schooling, and the incidence of i!titeracy is concentrated among adults at the lower end of the income strata. 3. Poverty is as significant and complex a problem today as it was in 1949 when the Bank's first economic mission to Colombia emphasized the need for a coordinated strategy to deal with poverty. Available data suggest that approximately 202 of Colombians live below a conservatively defined poverty line. Nevertheless, as the Colombian economy and society have changed, so has the nature of poverty. The plight of landless peasants remains in many aspects unchanged still. Their situation is to some extent overshadowed by the miserable conditions of the urban and peri-urban poor found in growing numbers in city slums and squatter settlements. Of particular concern are the prevalence of malnutrition as well as the high incidence of infant and maternal mortality among these poor households. Inadequate access to eanitation and safe drinking water perpetuate health problems in poor communities. The Government's Response 4. The Government, recognizing that poor people were not sharing equitably in the benefits of development, has over a period stretching back several decades tried to reduce income disparities. In particular, poverty alleviation has been set as the singie most important item on the policy agenda of the Barco Government (1987-90). 5. The Government's anti-poverty strategy is based on three major elements. First, the strategy is explicitly set within the framework of overall macroeconomic policies. On the premise that the country can best eradicate poverty by custaining  - ii - high economic growth and thereby expanding the overall demand for labor, the Government seeks to achieve steady GDP growth (around 5Z per annum during 1987-1990) combined with continued sound fiscal and external management. Secondly, beyond attaining the overall growth targets, the Government plans to expand specific programs designed to increase the productivity and employment opportunities of the poor, such as land reform, rural development projects and education and training. Thirdly, in order to directly tackle the worst aspects of poverty and to achieve a more immediate impact, the Government is stepping up the pace of basic needs programs specifically aimed at improving the access of the poor to key social services (i.e., health, education, nutrition, and basic utilities) and to adequate shelter. 6. The specific poverty alleviation interventions include (a) programs carried nv'r from previous administrations such as PNR and DRI; (b) the ongoing agrarian land reforms; and (c) a large number of initiatives (both short- and long-term) that collectively fall under a new national program called the 'War on Poverty.' The financing of these poverty-oriented interventions involves shifts of public investments and expenditures away from large capital-intensive infrastructure projects (e.g. power generation) toward the social sectors. 7. The Government's overall approach to poverty appears coherent and sensible. Nevertheless, many new initiatives under the 'War on Poverty' have progressed little beyond the conceptual stage: substantial work needs to be done to translate the still general proposals into specific action programs. To do that, a two-pronged approach is required. The Government needs to implement quickly those programs that can have immediate impact on the most critical and urgent problems facing poor communities (health problems, malnutrition, unsanitary living conditions, illiteracy) while continuing to develop medium- and long-term action programs aimed at achieving productivity increases and job creation. Particular attention needs to be paid to institution builAing. Institutional weaknesses, particularly in the health sector, llmit the pace of poverty program implementation. Obiectives of the Study 8. This study was designed to help the Government focus on designing and implementing short-term, quick-impact, basic needs types of programs. Its objectives are to evaluate and make recommendations on the Government's new short-term interventions in the areas of nutrition and health, low income housing and primary education. Our assessments of four programs under the 'War on Poverty" are summarized below in paras. 9 to 13. This study has intentionally left out discussion of what the Government needs to do in the medium to longer term for employment generation and productivity improvements. The study also focuses on the statistical work underlying the Government's poverty interventions and emphasizes the need for developing and refining the statistical tools. The Government needs more operational critnria for defin!ng poverty, measuring its magnitude and targetting program beneficiaries more selectively.  - iii - Interventions in Nutrition 9. The most innovative and promising among the Government's new basic needs initiatives is a program called 'Hogares de Bienestar Infantil' (HBI) which combines the objectives of supplemental feeding of children in poor neighborhoods with community-based day care organized by volunteer mothers. A government agency, the Colombian Institute for Family Wellbeing, provides the necessary food and the nucrition supplement called lienestarina, trains the volunteer mothers, pays them nominal stipends, and helps them get house improvement loans. An important feature of this program is that the day-care facilities are located in poor neighborhoods so that the beneficiary families can have easy access. Moreover, in addition to dramatically improving the nutritional status of those children covered by the program, HBI frees up their mothers to seek employment outside their home or their older sisters to attend school. The program is targetted largely at single mothers with young children, the most vulnerable poverty group in Colombia. The program now serves nearly 400,000 children across the country. The study concludes that the Government should promote this program as a center piece for its short-term *basic needs" poverty programs. 10. Nevertheless, there are some areas for improvement in the program design. Firstly, the coverage of infants between the age of 6 months to 2 years is low. Since the problem of malnutrition is generally most acute in this age group, the Government should quickly devise new modalities for ensuring adequate provision of Bienestarina for infants in poor communities. The Government might consider a system of home delivery of food and Bienestarina for targetted families or establish specialized day- care facilities for lafants to operate under closer supervision than the regular nEI. Fxcondly, the design of nEI is not well suited to rural communities where settlements are dispersed. In such localities, the Government must develop other approaches including the use of rural school restaurants to feed pre-school age children. Thirdly, as the coverage of HBI is extended, it would become necessary to mobilize the communities more forcefully to participate in the administration of the program. Fourthly, a much closer coordination with the country's primary health care delivery system is crucial to permit the screening and treatment of serious malnutrition cases. Finally, ICBF needs to incorporate an effective information system into the program, not only to monitor nutritional impact, but also to target beneficiaries more selectively and to help make adjustments in the design and admlnistration of the program. Assessment of Government Proposals in Health 11. The main thrust of the Government's poverty alleviation strategy in health is to achieve universal primary health care and to extend the coverage of the social security system to reach low-income families. These objectives do not appear to be attainable in the short-term. Firstly, given the poor utilization of existing primary health care facilities, there is little point in tryin- t, expand coverage by building additional health posts or training more paramedici. Secondly, the proposal with respect to social security is unrealistic; its implementLtion would require far- reaching reforms of the financially-troubled and administratively weak Social Security Institute which is not likely to be achieved in the foreseeable future. What would benefit the poor is for the Government to try to make the public health  - iv - system function effectively in its current capacity by shifting resources from the hospital-based tertiary care to the commmity-based primary health care. Over the medium term. the Government should make a concerted effort to undertake administrative reforms in the Ministry of Health to improve its implementation capacity and rationalize what hba become a complex and unmanageable system of earmarked revenues. The first necessary step is to promptly complete the ongoing study of the health services andertaken jointly by the Ministry of Health, Colombian Institute of Family Wellbeing and the National Planning Department which focuses on the administrative and financial management problems of the sector. A New Approach to Low-Income Shelter 12. The Government's approach to low-income shelter, which focuses on slum upgrading represents a sharp departure from the earlier policy based on direct public sector development of heavily subsidized low-cost housing. This latter policy had totally failed to address the housing needs of the poorest households. The current strategy is to improve the quality of housing in the unregulated housing subdivisions (unauthorized private developments and squatter settlements) through provision of basic services (water, sanitation and electricity) and the upgrading of housing largely with self-help construction to reduce costs. The approach is sensible. However, the prograa 'Human Settlements' as proposed is too large and its implementation will present a challenge both to the Government and local communities. The upgrading of homes in poor neighLorhoods will be a complex process. It involves legalization of land tenure, coordination among numerous public agencies and municipal governments in planning, financing and managing infrastructural investments, and the mobilization of the beneficiaries to undertake self-help construction and to organize various community-based services. Since shelter upg&.c''` has never been tried in Colombia on a large scale, public agencies and communities will need to proceed cautiously, making adjustments in project designs to suit local conditions. The Government should reduce the physical targets for the program and concentrate on formulating an integrated approach to community-based s'um upgrading. Over the past few years, the Central Mortgage Bank, in collaboration with a number of public agencies and non-governmental organizations, has been involved in such an experiment in community building and housing improvement in Cali. While these pilot projects have not been implemented on a large enough scale to test their replicability, they merit a closer examination for their potential for reaching much poorer households than is normally possible under traditional housing programs. Universal Primary Education 13. In the field of education, the Government's explicit priority is universalizatio.a of primary education, with emphasis on extending coverage in rural areas using a system of multi-grade instruction and flexible grade promotien called the 4escueia nueva' model. By comparison, little attention is given to poor urban neighborhoods where large numbers of school-age children do not have access to formal primary education. In many poor communities, the parents pool their resources to organize informal classes (the so-called 'escuela de banco") but the results are generally unsatisfactory. The teachers, poorly qualified to start with, must make do  without the necessary teaching materials. Classes 'lso take place in poor facilities without desks or blackboards. Moreover, sisnce these schools are not officially accredited, those students completing an equivalent of primary education in "escuela de banco" have no chance of advancing to the formal seccndary school system. The Government might consider moving towards standardizing the curricula and the quality of instruction in "escuelas de banco" by distributing free textbooks and offering periodic training for the teachers. Students from these schools might be integrated into the formal system upon passing a specified examination, and over time, the Ministry of Education might take steps to accredit those schools that meet the minimum necessary standards. Alternatively, the Ministry might authorize additional formal primary schools in poor urban neighborhoods. The Government does not accord high priority to adult education which includes skill training and literacy campaigns. This approach is sound because in the long run, it would be much more cost-effective to concentrate on the formal primary education system as the principal vehicle for disseminating literacy and for imparting basic communication and production skills to the poorer segments of the population. In addition to improving access, it is also important to enhance incentives for (or lower the cost of) school attendance for children from low income families. This can be achieved most effectively by expanding the existing school feeding program and implementing the free textbook program proposed under universal primary education. A Summary of Recommendations 14. Statistical Data Base. The poverty measure that the Government utilizes- -the NBI (or the "Basic-Needs-Not-Met" Index)--is based primarily on physical, housing-related indicators and is inappropriate for drawing poverty lines or targetting beneficiaries. What Colombia needs is a system of periodic surveys to generate reliable and relevant data (including those of household incomes and expenditures, employment and nutritional status) in a form that permits a wide range of applications and analysis at a sufficiently disaggregated level. The Government might consider utilizing the Living Standards Survey, a methodology for data collection and generation, developed by the World Bank. The National Statistical Office shoulc be able to tailor the Living Standards Survey to its needs and generate poverty data to complement the NBI measure. 15. Nutrition. Improvements need to be made in taie design of 'Hogares de Bienestar Infantil" program to permit greater coverage for infants between the ages of 0 and 2 years and to accomodate children in rural areas. The year-by-year targets of the program are ambitious. Care should be taken that the pace of implementation does not outstrip the administrative capacity of the Colombian Institute for Family Wellbeing (ICBF), the executing agency. In addition, a greater effort should be directed at mobilizing community support and participation to lessen ICBF's administrative burden. ICBF also needs to adopt a reliable information system that permits more selective targetting of beneficieries and monitoring of program impact. 16. Health. Proposals for universal primary health and social security coverage will be difficult to implement in the short term because of institutional weaknesses in the sector. What the Government needs to do is to make the existing primary health delivery system work by shifting recurrent funding from the hospl al-based tertiary care to the community-based primary care. Health sector finance is one of the key topics of the health sector study currently undertaken by the Ministry of Health. A concerted effort should be made to complete this study promptly and to undertake administrative and financial reforms of the health sector.  - vi - 17. Housing. In the area of shelter, the re-orientation in Government policy from new construction to rehabilitation of existing housing in poor neighborhoods is sensible and commendable. Given that public agencies have had little experience in shelter upgrading programs, the Government ought to proceed cautiously with the implementation of the 'asentamientos humanos" program. Its ambitious targets for 1987-1990 ought to be scaled down. Interest subsidies must be phased out over time to reduce the burden on the financial sector. To make home upgrading affordable to the poorest families without interest subsidies, it is essential to rely on self-help construction; this in turn entails training, technical assistance -nd effective community organization. The most Important first step should be develop a new and inexpensive formula for slum upgrading and site servicing similar to the approach pioneered by the Central Mortgage Bank in Cali. 18. Education. In the field of education, the Government's basic strategy is universal primary education with emphasis on improving access to schools in rural areas. However, large numbers of school-age children in urban and peri-urban areas have no access to formal primary education. To remedy this situation, the Government might authorize (ormal primary schools in those poor urban neighborhoods where no public schools operate today. In communities where private, non-accredited community schools ("escuelas de banco') operate, the Ministry of Education might try to improve the level of instruction in these schools by distributing free text books or offering training to teachers, many of whom are not sufficiently qualified. Furthermore, the existing school feeding program should be expanded to improve incentiveb for school attendance.  Chapter 1 Introduction Tran-rormation of the Colombian Economy 1.01 Over the past two to three decades, Colombia has made important economic gains. The country's wealth of resources has provided a solid base for development and growth. The Colombian economy has grown steadily, weathering cyclic swings and downturns that have accompanied fluctuations in coffee prices due in large part to the Government's prudent economic management. During 1960-1985 the incomes grew at around 5Z annually in real terms. Since the sixties, the country has also experienced a dramatic transformation. The economy has become increasingly more broad-based as its reliance on coffee has been offset by the growing importance of petroleum, mineral and industrial sectors. Sizeable investments have taken place in infrastructure, especially in power and roads, and social services have been developed throughout the country. Far more Colombians today attend schools and have access to health services, safe drinking water and other basic services than ever before. 1.02 During this period, the Colombian society changed profoundly as the rural urban migration which had started ia the 1930s continued at a rapid pace and pushed the urban share of the population from less than 30Z in the 1950s to 702 in 1985. Improvements in social indicators during this period have also been quite spectacular. Total fertility rate declined by about 45Z from the early 1960s to the mid- 19808 and is currently estimated at 3.5Z. This sharp reduction in fertility, attributable to successfttl family planning campaigns as well as to the spread of education, urbanization, and greater female participation :ates in the labor force, served to lower population growth rates from a high of 3.72 between 1964-1978 to less than 1.8Z between 1978-1985. This fertility decline has been accompanied by a similarly rapid decline in rates of maternal mortality (from 254 per 100,000 live births in 1964 to 107 in 1984). Estimates for infant mortality (deaths per 1,000 births) went down from 81 to 61 during the same period. On the education front, primary school enrollment more than doubled, and secondary school enrollment increased six-fold. Persistence of Poverty 1.03 Despite overall economic growth, benefits of development have not accrued to large groups of the poor in both rural areas and cities. Poverty ib as intractable a problem in Colombia today as it was in 1949 when a Bank mission led by Lauc*hlin Currie proposed a "coordinated attack on poverty," mainly in the form of a public investment program. It is as critical an issue today as in 1972, when a report by the Bank's Operations Evaluations Division pointed out  -2- that 20 years of agricultural development had made no significant direct impact in alleviating the poverty of the rural population.l Income distribution also remains skewed today with incomes of the top 202 of households six to seven times those of the bottom 202. It is worth noting that the focus of poverty programs has shifted over the years. In the fifties and sixties, the emphasis was on rural development. Since the seventies, as urbanization progressed and shantytowns and illegal squatter settlements proliferated on the outskirts of large cities, poverty alleviation came to be associated increasingly with the provision of low-income urban housing and related services. These shifts reflected the demographic changes and the increasing concern with urban poverty. 1.04 The impressive gains made in the social sectors also bely the sharp differences in health and educational status among socio- economic groups. For instance, fertility is estimated to be three times as high among the critically poor as among the non-poor, and the higher fertility among poorer women is accompanied by shorter birth intervals, higher infant mortality, and generally poorer health among both mothers and children. Moreover, a high proportion of children from poor communities still have no access to any kind of schooling, and the incidence of illiteracy is high among adults at the lower end of the income strata. (See Chapter 7.) Government's Commitment to Social Programs 1.05 The Government has, for decades, tried to reduce the disparities in social indicators and income levels. Successive administrations since the sixties have actively implemented social programs, especially education and health. In the seventies, poverty alleviation became a more specific goal. The Lopez Michelsen administration (1975-1978), for instance, emphasized the importance of improving nutrition and incomes in rural areas and developed trograms that came to be known as the Integrated Rural Development Program (DRI) and National Food and Nutrition (PAN). The Betancur administration (1983-1986) actively promoted a policy of national integration and peace and sought to achieve its goals by extending public services and productive opportunities to the more remote and underdeveloped regions of the country. This policy was translated into the National Rehabilitation Plan (PNR) in 1986. The Betancur government also implemented a low income housing program refered to as the 'Vivienda Popular' program directed primarily at the urban poor.2 1.06 Under the present Government, poverty alleviation is the single most important policy agenda and the central theme of its four- year Development Plan ('Plan de Economia Social"), covering the 1987-1990 period. The development plan builds on two related national 1/ International Bank for Reconstruction and Development: "Bank Operations in Colombia--An Evaluation," May 25, 1972. See also Lauchlin Currie: Evaluacion de la Asesoria Economica a los Paises en Desarrollo--el Caso Colombiano, Bogota, 1984. 2/ Brief descriptions of these poverty programs are provided in Annex I.  -3- programs: the PNR (a carry-over from the previous Government) and the new initiatives--"War on Poverty" or 'Plan de Lucha contra la Pobreza Absoluta y para la Generacion de Empleo" (PLPGE)--announced early in 1987.3 The importance attached to poverty alleviation clearly reflects a preoccupation with the growing violence and social unrest in the country which many attribute to the persistent disparities in income distribution and poverty. A Macroeconomic Framework for Poverty Alleviation 1.07 The Government's overall approach to poverty is coherent and sensible. In essence, the Government has adopted a strategy aimed at raising the productivity and income of the poor and improving the access of the poor to key social services. It proposes to do so by shifting public investments and current expenditures increasingly toward social sectors. The poverty alleviation goals are to be accomplished within the context of steady GDP and export growth and prudent demand management. 1.08 On the premise that sustained growth and sound economic management are among the most important pre-requisites for poverty alleviation the government has established the following macroeconomic targets: - real GDP growth averaging 5Z a year; - total investment as a percentage of GDP rising from 17% in 1986 to 202 in 1990; - current account deficits of the balance of payments maintained below 3Z of GDP; - consolidated public sector deficits not exceeding 32 of GDP; and - inflation rates stabilized at recent levels (22% a year). The viability of the macroeconomic scenario, as summarized in Table I- 1, depends crucially on the willingness of the private sector to raise investment to around llZ-12 of GDP throughout this period. Given the declining trend for private investment in recent years, policy makers will need to focus on providing adequate incentives for private investors. In this context, the uncertain security situation presents an increasingly serious constraint. Furthermore, the Government needs to continue to exercise restraint on total public sector spending and step up resource mobilization efforts in order to keep the deficits within manageable magnitudes. Fiscal stability will be crucial, particularly because poverty-oriented programs have tended to be vulnerable to fiscal pressures in the past. A good example of this is PAN (para. 1.05), an innovative nutrition program which was launched in the mid-1970s and discontinued for fiscal reasons in the early 1980s when economic growth began to slow down (See Annex I for a brief description of the program.). / Tn 44 F 4v t tinl r PT Prv ,4 4' T  -4- Priorities in Public Expenditures 1.09 The 1987-1990 Development Plan outlines quite specific sectoral priorities that reflect ongoing sector programs. In social services (health and nutrition, education and water supply), the emphasis is on the expansion of basic services. Great importance is attached to the expansion of primary health care delivery and the extension of social security to poorer segments of society. Universal primary education is also an explicit goal. Investments in water and sewerage are likewise designed to support the Government's basic needs strategy. While agriculture is seen as a key sector, public investment levels remain low and secto- policies are not sharply articulated. Priority goes to programs under the ongoing DRI and PNR (para.1.05) aimed at raising productivity and incomes of small farmers and ensuring food security. Significant investments in basic infrastructure (e.g., feeder roads) are envisaged under these rural development projects. Table I-1: Macroeconomic Framework 1986 1987 1988 1989 1990 actual ---------projections--------- GDP Growth (Z) 5.1 5.0 5.0 5.0 5.0 As X of GDP: National Savings 22.0 15.6 16.6 17.9 17.2 Public Savings 6.2 4.3 4.2 4.9 4.9 Private Savings 15.8 11.3 12.4 13.0 12.3 Total Investment 18.0 18.0 19.6 19.8 19.7 Public Investment 8.6 6.8 7.0 7.4 7.4 Private Investment 9.4 11.2 12.6 12.4 12.3 Foreign Savings = Current Account Deficit -4.0 2.4 3.0 1.9 2.5 Public Sector Deficit 0.5 2.5 2.8 2.5 2.5 Financed by: Net Ext. Borrowing 3.6 0.8 0.8 1.0 1.3 Net Dom. Borrowing -3.1 1.7 2.0 1.5 1.2 Source: Plan de Economia Social, 1987-1990 1.10 To make room for increases in expenditures on poverty- oriented programs, the Development Plan aims explicitly to reduce the share of investments in the energy sector while maintaining a lid on overall spending. A particular importance is attached to  -5- rationalizing investments and expenditures in the power sector where an adjustment program currently underway involves financial and institutional rehabilitation and a shift in investments from generation to transmission and distribution. In transport, in addition to directing road investments to rural areas under DRI and PNR, the Development Plan emphasizes improved utilization of existing infrastructure through maintenance and more effective coordination among and use of the various transportation modes. Table I-2s Public Insvstment Program (In Z Shares) 1986 1987-1990 Actual Plan Social Infrastructure 14.3 26.8 Water and Sewerage 5.3 9.9 Education 4.5 7.0 Health and Nutrition 4.5 9.9 Energy 50.3 34.1 Power 25.7 17.6 Petroleum 20.0 14.6 Coal 4.7 1.9 Physical Infrastructure 19.7 28.3 Transportation 16.3 19.7 Communication 3.3 8.6 Agriculture 1.0 10.0 Others 14.7 0.7 TOTAL 100.0 100.0 Source: DNP: Plan de Economia Social 1987-1990, and World Bank estimates 1.11 The Government's efforts to change the focus of investments and ensure efficient resource use are rendered difficult by the rigidities in the budget process (e.g., complicated earmarking of revenues). Institutional weaknesses also limit the pace at which social sector investments can be accelerated in an efficient manner. This is true particularly in the health sector (see paras. 1.12, 3.06). Meanwhile, the momentum of programs in power and infrastructure is strong a:id thus could potentially undermine the effort to control expenditures.  -6- Table 1-3: Public Expenditures on Principal Poverty Programs by Sectors during 1987-1990 (In 1987 Col$ Million) Expenditures Shares in on Poverty- Total Oriented Sectoral Programs Expenditures (Z) Health and Nutrition 455,951 33.0 Education 458,000 38.0 Water and Sewerage * 230,000 85.5 Agriculture * 272,787** 41.2 L Limited to public investment figures ** DRI and PNR Source: Plan de Economia Social 1987-1990 Limited Absorptive Capacity in the Social Sectors 1.12 The Goveranent has made considerable progress in re- orienting the focus of ongoing sectoral programs to fit its overall anti-poverty strategy. Table I-3 shows that the poverty-oriented programs account for relatively large shares of the sectoral totals. However, it is by no meins certain that the Government would be able to achieve these planned expenditures. The cost of these programs and their financing arrangements are not likely to be the principal obstacles; what is more likely to constrain the implementation of these programs is the limited absorptive capacity in these concerned sectors. For example, the Government's new initiatives in the health sector (i.e., expansion of primaiy health care) are important from the standpoint of the basic needs of the poor but little thought has been given to the enormous institutional and financial m!inagement problems in the Ministry of Health and the Social Security 'nstitute. Moreover, many programs under implementation or ir advanced stages of preparation (e.g., nutrition, education) have physical targets that appear ambitious relative to the proven institutional capacity of the executing agencies. The issue of implementation is discussed furlther under individual programs in Chapters 4 through 7. 1.13 In view of the institutional constraints, it is important that the Government first clarify the nature of the poverty problem, secondly, set explicit priorities, and thirdly, design a program of action to address a set of clearly defined issues, distinguishing between those initiatives that can have an immediate impact and those that take longer for their impact to be felt.  - 7 - Chapter 2 The Dimensions of Poverty in Colombia A. Measurement of Poverty 2.01 Poverty persists despite the 30 years of economic gains that have helped raise the living standards of Colombians. As mentioned in Chapter 1, poverty alleviation has been high on the policy agenda of successive administrations and the 'War on Poverty' is the central theme of the development plan for 1987-1990. However, to support its initiatives, the Government must first define the concept of poverty more clearly and second refine its statistical tools for determining its magnitude and geographical distribution and for targetting beneficiaries for poverty programs. The Government now defines poverty primarily in terms of housing-related criteria, based on data from the 1985 Population and Housing Census and estimates that 43Z of Colombians live in 'absolute poverty.' (See paras. 2.02-2.03) This is not an appropriate definition of poverty. It is also unrealistic to target such a large segment of the population for poverty alleviation programs. NBI--Government's Official Poverty Index 2.02 The DANE employs a measure called the 'Basic Needs Not Met' Index (NBI). Using the results of the 1985 Population and Housing Census, the NBI methodology defines poverty primarily in terms of three shelter-related indicators, namely, (a) the adequacy of housing judged on the basis of types of materials used on walls, roofs and floors; (b) the minimum occupation density, and (c) the access to public utilities such as electricity, potable water, sanitation. It furnishes two indicators that serve as proxy for income data; namely, (d) number of dependents per wage earner in the household and (e) number of school age children in a household not attending school.1 A household is defined as living in absolute poverty ('pobreza absoluta") if it meets one of the five indicators mentioned above. Furthermore, a household is said to be in dire misery ('en la miseria') if it meets two or more of the five indicators of poverty. Using these criteria, 38.12 of households (or 43Z of total population) are classified as living in 'absolute poverty', and 17.6? of households (or 21.42 of total population) in a state of misery. The Government uses the first definition in all its official documents. 2.03 The NBI index has a number of serious shortcomings. Firstly, it has a strong bias towards housing and housing-related public service indicators. The problem with using the availability of 11 This methodology is described fully in DANEt Boletin de Estadistica (411/junio 1987).  services as criteria is that, in rural areas where infrastructural investments tend to iag behind, the lack of access to electricity or sanitation does not necessarily in itself indicate poverty. Poverty is more aptly characterized by other conditions such as malnutrition, illiteracy and unemployment or underemployment. Secondly, NBI does not include data on household incomes and expenditures. In large part, the choice of indicators for NBI was dictated by the type of information available from the 1985 Population and Housing Census; questions on household incomes and expenditures were not included in the Census that year. There is no other recent source of information with a nationwide coverage that can be used to analyze the living standards of Colombians. 2.04 Thirdly, and most importantly, NBI does not permit analysis at sufficiently disaggregated levels. The departmental or municipal averages of the various NBI indicators show in global terms the regional patterns and magnitudes (See maps, pages 11 - 14). Nonetheless, they fail to highlight what is probably the most important dimension of poverty in Colombia, i.e., the tremendous disparity in living standards that exists between communities within a single department or a municipio. Pockets of extreme poverty are observed alongside relative affluence in economically advanced departments such as Bogota, D.E., Antioquia or Valle. With this type of information base, it is impossible, for example, to compare the relative deprivation of people living in urban slums with that of landless peasants in a remote rural village, and to decide which group needs assistance more urgently. Recommendations 2.05 The Government urgently needs to further develop the statistical framework for supporting its poverty interventions. We recommend that DANE incorporate in the poverty index all the relevant information for measuring living standards (e.g., household incomes and expenditures, employment, and health and nutrition). What Colombia needs is a system of permanent household surveys to generate reliable and relevant data in a form that permits a wide range of applications (e.g., drawing poverty lines, targetting beneficiaries, monitoring programs, etc.) and analysis at sufficiently disaggregated levels. 2.06 DANE has a great deal of experience in collecting household survey data through periodic wEncuesta de Hogares' but the design of their surveys needs to be modified. Firstly, modules on health, nutrition, expenditure, education, hours of work at home and labor market should be permanent features of DANE's household surveys. Secondly, the coverage of household surveys must be expanded from the current 13 cities to include additional urban areas and rural areas. Thirdly, in view of the additional demands that the inclusion of new modules and expanded coverage place on DANE, serious consideration should be given to reducing the sample size of household surveys from the current 30,000 households. It is not necessary to have a sample this large to obtain accurate results.  - 9 - 2.07 The Government might consider utilizing the Living Standards Survey (LSS), a system of data collection and data generation developed by the World Bank. The Bank established the Living Standards Measurement Study (LSMS) in 1980 to explore ways of improving the types and quality of household data collected by statistical offices in developing countries. The objectives are to develop methods, specifically tailored to the needs of each country, to monitor progress in raising levels of living and to identify the consequences for households of past and proposed government policies. The primary instrument for LSMS is the Living Standards Survey, designed to collect information oi a whole range of indicators for economic wellbeing at the household and community level. The Living Standards Survey is ideally suited for building on and integrating existing household surveys and should serve DANE well in generating additional poverty data to complement the NBI indicator. The Bank's experience with LSS in coumtries like Peru and Ivory Coast indicates that the system is capable of generating high quality data very qui:kly and in a cost effective manner. B. Spatial Distribution of PovertZ2 2.08 This section discusses the geographical distribution of NBI indicators by departments based on the 1985 Census data. The results indicate considerable regional differences. Households ^In Miser4' as Defined by NBI 2.09 Reflecting the bias of the NBI measure towards physical indicators, the incidence of poverty indicators tends to be high in sparsely-populated and predominantly rural parts of the country where infrastructure is still relatively underdeveloped, namely, Sucre, Cordoba, Cesar, La Guajira, and Magdalena in the Atlantic Region, Choco, Cauca and Narino in the Pacific Region, Caqueta in the Central Region, North Santander and Boyaca in the Eastern Region as well as the whole of 'Intendencias' and 'Comisarias' with the sole exception of San Andres, an island and a popular tourist resort (See the map on page 11). There are exceptions such as Bolivar where some 40Z of households are classified as poor though nearly 702 of the department's 1.2 million population live in urban aseas. On the other hand, poverty tends to be least widespread in departments that ars 2/ The analysis in this section is based on DANE's poverty mapping exercise using NBI indicators at the level of departments. Spatial distribution of poverty by 'municipios' is shown on page 87.  - 10 - highly urbanized, e.g., Bogota, D.E.; or in those that have sizeable coffee production, e.g., Antioquia, Caldas, Risaralda. or in those that have both large urban centers and significant coffee production, e.g., Quindio and Valle.3 Access to Public Services 2.10 On the basis of the public services coverage indicator alone, the departments that lag significantly behind the rest of the country include Cordoba (54Z of households without any public services), Sucre (51Z), Bolivar (45Z), Boyaca (442), Choco (692) as well as the wIntendencias' and 'Comisarias' where more than 402 of households have no access to any kind of public services (See the map on page 12). This lack of access to public services (and in particular, the availability of sanitation) in less developed parts of the country is clearly reflected in the prevalence of water-borne diseases among young children. The map entitled 'Death Due to Intestinal Infection' shows that the geographical differences in the incidence of deech due to gastro-intestinal diseases among children of 0 - 4 years of a3e are closply correlated with the degrees of access to public servicas. !3/ Coffee production makes an important difference in these regions, not only because it ensures a steady source of income for the population, but also because the Coffee Federation, in agreement with the Government, sets aside a fixed proportion of its export earnings to support social services and to invest in social infrastructure in coffee growing areas.  -1 1 _ . IBRD 20735R COLOMBIA -. Car;bbeon Soe NICARAGUA HOUSEHOLDS "CON MISERIA" COSA.E I?~~~~~~~~~~~~~~~~~~~~O PANAAAUCOLOMBIA VIA / \*fF I ~~~~~~~* I\ IT-* (' k~~~~~~~~~~~~~~~~~~~~~~~~- , RAZILt PANAMA I,'-~~~~~~~~~~~~~~~~~~~~~~~~~~~0 ~~ _ AN4TIOQIUIA Pacific CHOC6 Ocean 6 CA<A META C A4JAId/ ae( K J, )\ t\\ \ UCA / ! 4~~ ~ ~~~~ ' do-*-' < rUt,:'\ o. f CA U CA J \ 1A E>A a U) s C' 4 BRAZIL PERCENTAGE OF HOUSEHOLDS *LES "CON MISERIA"* dIgo o 50 100 t50 40% - 50% (Highest volue:.C6rdo.o-50.1%) AA.AZONAS 6