Senegal - An assessment of living conditions (Vol. 1 of 2) : Main report

Report No. 12517-SE Senegal An Assessment of Living Conditions (In Two Volumes) Volume l: Main Report May 5, 1995 Africa Region Western Africa Department Country Operations __~~~~~~~~~ : A,,_  Currency Equivalent Currency Unit: CFA Franc Period Average: 1992 - CFAF 264.69/USS Period Average: 1994 - CFAF 555.20/US$ ACRONYIMS AND ABBREVIATIONS AGETIP Agence d'Execution des Travaux d'Intdr6t Public BCEAO Banque Centrale de Etats de ['Afrique de l'Ouest CCF Christian Children's Fund CNCAS Caisse National de Credit Agricole du Senegal CONGAD Conseil des ONG de Developpement CPSP Caisse de Perequation ec de Stabilisation des Prix CSA Commissariat a la Securite Alimentaire DAARA Islamic School teaching the Koran DPS Division de la Prevision et de la Statistique DRC Domestic Resource Cost FDEA Femmes, Developpement. Entreprise En Afrique FED Fonds Europeen de Developpement FONGS Federation de Organisations Non Gouvernementales du Senegal GDP Gross Domestic Product GOS Government of Senegal IBRD International Bank for Reconstruction and Development IDA International Development Agency IFRPI International Food Research Policy Institute ISRA Senegalese Institute for Agricultural Research NGO Non-Governmental Organization NPI Nouvelle Politique Industrielle ONCAD Office National de Cooperation et d'Assistance pour le Developpement PAGD Programme d'Appui A la Gestion et le Developpement SAED Societe Nationale d'Amenagement et d'Exploitation des Terres du Delta du Fleuve Senegal et de Vallees du Fleuve Sendgal et de la Falem&e SAL Structural Adjustment Loan SAR Societe Africaine de Raffinage SMIG Salaire Minimum Interprofessionel Garanti UMOA Union Monetaire Ouest Africaine UNDP United Nations Development Program UNICEF United Nations Children's Fund USAID United States Agency for International Development  ADDENDUM .............................i EXECUTIVE SUMMARY .. .........................iv 1. POVERTY PROFILE .......... I 1. 1 The Demographic Context ................................ 1 1.2 Defining Poverty ...................................... 1 1.3 Do the Poor Consume Different Goods than the Non-Poor? .... ....... 4 1.4 Revenue Patterns of the Poor and Predictors of Poverty ............. 7 1.5 Other Characteristics of the Poor and Structurally Disadvantaged Groups (Women) ........................................ 8 1.6 Coping Strategies - The Asset of Choice - A Strong Social Network ... ... 10 1.7 Those Who Fall Through the Traditional Safety Net ................ 11 1.8 Urban Poverty - a Problem Likely to Grow ..................... 13 2. FACTORS AFFECTING THE OPPORTUNITY FOR INCOME ..... ........ 15 2.1 The Implications of Recent Macroeconomic Trends for the Poor ... ..... 15 2.2 The Informal Sector Labor Market: The Primary Employment of the Urban Poor .................................. ..... 18 2.3 Access to Credit - the Stepping Stone to Higher Income ............. 20 2.4 Price Policies for Major Commnodities ........................ 24 2.5 Rice Policy Tradeoffs - Protection at what price? ................. 26 3. TOWARDS LONG-TERM GROWTH: THE ROLE OF GOVERNMENT IN PROVIDING FOR BASIC NEEDS ............................. 32 3.1 Current Health Situation .................................. 32 3.2 Public Health Policy and the Poor ........................... 32 3.3 Factors Affecting the Demand and Supply of Education for the Poor ..... 36 3.4 Long-term Issues: From Here to Universal Primary Education .................... ... 39 3.5 Consistency of Government Investment with Promotion of the Rural Sector .......................................... 41 4. FORMAL SAFETY NETS AND TARGETED PROJECTS ................. 46 4.1 Formal Programs ..................................... 46 4.2 The Role of NGOs, Women's Associations ..................... 49 5. TOWARDS AN OPERATIONAL STRATEGY FOR POVERTY REDUCTION IN SENEGAL .......................................... 52 5.1 From Here to 2015 ................ .. ................. 52 5.2 Short- Medium-term Measures to Decrease Vulnerability and to Protect the Poor ......................................... 53 5.3 Towards More Equitable Growth ........................... 55 5.4 Promoting More Equitable Access to Basic Services ................ 56 5.5 Major Trends Driving Poverty at the Macroeconomic Level ... ... . 58 5.6 Institutional and Process Issues ............................. 59 BIBLIOGRAPHY ............................................. 66 List of Tables Table 1.1: 1992 Poverty Lines ...................................... 3  Table 1.2: 1992 Poverty Indicators ................................... 3 Table 1.3: The Urban/Rural Gap .................... 4 Table 1.4: Monthly Expenditures Shares ...... ........... . 5 Table 1.5: Four Most Important Cash Income Sources Ranked .......... ... . S Table 1.6: Characteristics of the Poor ................................ . q Table 1.7: Summary Characteristics of the Poor .. 14 Table 2. 1: Evolution of Production .. 17 Table 2.2: Evolution of Formal Sector Wages .. 18 Table 2.3: Market Share of HH Spending Less than 6,000 fcfa per capita per month .... 24 Table 2.4: The Link Between Population Growth and Declining Land: 1976-1988 ... ... g Table 3.1: Senegal: Per Capita Real Public Health Expenditure by Region ........... Table 3.2: Selected Education Indicators, Senegal Compared to Sub-Saharan Africa ... 39 Table 3.3: Public Policy and Poor ................ ................... 44 Table 4.1: Summary: Formal Safety Net Programs for the Poor . . 51 Table 5. 1: The Importance of Education Investment in Poverty Reduction .... ....... 54 Table 5.2: Policy Options ............ ............... 63 List of Boxes Box 1. I Alternative Approaches to Poverty Lines .......................... Box 1.2 The Poor's Ability to Take Advantage of Growth in the Rural Sector ........ Box 1.3 Talibes ........................... Box 2.1 Increased Competition and Lack of Capitol ......................... Box 2.2 Credit for Low-Income Women: Combining Business Advice with Credit ... 22 Box 2.3 The Importance of Informal Credit ........... .. ................. 23 Box 3.1 Preventing Malnutrition in Senegal .. 37 Box 3.2 Community Mobilization in Fandene - Grassroots Insurance ..... ......... 38 Box 3.3 Regional Targeting in Education .. 41 Box 4.1 Urban Safety Nets .. 46 Box 4.2 Targeted Projects: Letting the Beneficiaries Define Poverty ..... ......... 48 Box 4.3 Disengagement of the State - the Role of NGOs .. 50 Box 5.1 What Was the Strategy Behind Rapid Poverty Reduction in East Asia . Box 5.2 Five Ways to Better Target the Poor .. 60 Box 5.3 Criteria for Assessing the 'Quality of Participation ...... ............. 61 Lst Qf Figures Figure 1.1 Senegal: Per Capita Monthly Food Expenditures as Percent Total by Expenditure Group ...................................... Figure 1.2 Share of Rice in Total Caloric Intake by Expenditure Group, Dakar, Urban, and Rural ....................................... Figure 1.3 Profession and Consumption Levels ............................ Figure 1.4 Education and Consumption Levels ............................ Figure 2.1 Rural Terms of Trade Index ................................ 17 Figure 2.2 Informal Sector Firm Size, 1981 .............................. 20 Figure 2.3 Senegal: Sources of Household Fuel ........................... 25 Figure 2.4 Senegal: Sources of Household Fuel ........................... 25 Figure 3.1 Senegal: Health Care Expenditure . ............................ 34 Figure 3.2 Senegal: Health Care Financing Source ......... ................ 34 Figure 3.3 Percent GOS Education Expenditure ........... ................ S Figure 3.4 Senegal: Female Welfare .................................. 1  Priority Povertv Indicators 'I ear -0 88 92 95 Poverty Lines - Lower (907) 29 7c l Number of People (million) 2.1 -Base 33 4"l Number of People (million) 2.4 - L'pper 1(10%) 37a% Number of People million) 26 GDP Per Capita 198 -05 725 b CPI t .hg I l 8 -1 8 Urban. Rural TOT Rural P.C GDP (Constant FCFA) 82.966 64,664.'a 68.654 Urban P C GFP 'Constant FCFA) 508.268 412.098/a 373.706 Stunting 29% - Urban 22% Rural 34 % Wasting 5.5 % - Male 6.5 - Female 4.5 %c Public Expenditures on Health/Education 20% 25%/d Infant Mortality (0-5) 86/c Urban 70/c Rural 102/c Immunization Tuberculosis 14 72/a DPT I 21/a Polio 0 7/a Life Expectancy 48/c Primary School Enrollment 58%/b 65% (1998 Urban 98% + /b goal) Rural 38%/b Girls 49%/b Boys 70%/b Total Fertility Rate 6.51c Crude Death Rate 16 Contraceptive Prevalence (all methods) 10% 11% (1996 .____________________________________________ _ _ _ _ _ _ _ _ _ ___________ g o al) % Women with No Prenatal Health Care Urban 4.5% Rural 52.2% Access to Safe Drinking Water 53%/b Urban 79%/b Rural 38% lb a/1985 b/ 1991 c/1986 d/Proposed in most recent Public Expenditure Review  PREFACE The World Bank's 1990 World Development Report on Poverty highlighted the fact that economic growth alone does not necessarily reduce poverty in a given country. To be effective, economic and social policies have to address also the needs and concerns of those, who, because of structural, social and cultural factors, have little voice in the development arena. It is in this light that this Assessment of Living Conditions, one in a series of Country Poverty Assessments that the World Bank is preparing for all its member countries, was elaborated. Until recently quantitative information on poverty in Senegal was very limited. The completion of the first household Priority Survey in 1993, the basis of the Poverty Profile, provides a better statistical source for analyzing the incidence of poverty in Senegal and for a more focused discussion between the Bank, other donors, the Government of Senegal, (GOS) and NGOs on poverty alleviation. It is hoped that this report represents a tool for enriching the Bank-GOS dialogue on this issue, and for moving towards the elaboration of a poverty alleviation strategy in Senegal. The report is based on the findings of a mission to Senegal led by Sarah Keener (Consultant, World Bank, mission leader) in March/April 1992, with the participation of David Kingsbury (consultant, DAI), Abdoulaye Sadio (staff, Direction de la Prevision et de la Statistique, (DPS), (GOS)), Bahjat Achikbache (AFTHR), and FAC-financed consultants Jean-Paul Azam and Alice-Nicole Sidzingre (anthropologist - NGO sector). The report was prepared by Sarah Keener and David Kingsbury with contributions from all mission members as well as Anne Ching (summer intern -incentive and regulatory policies) and discussed with the Government then finalized by Adriana De Leva (Economist, AFSPH). Special thanks go to Mme. Awa Thiongane (DPS) for providing input and substantial time of her staff, as well as to M. Thierno Niane (Direction de la Planification in the Ministere de l'Economie, des Finances et du Plan) for assisting the mission. The peer reviewer is Steen Jorgensen (AF6PH). Mr. Birger Fredriksen is the division chief, Mr. Jean-Louis Sarbib is the department director, and Mr. Francois Laporte is the lead economist. The report was discussed with the Government in February 1995 and presented at a Poverty Workshop organized by the Govermnent in March/April 1995. The Addendum that follows is in response to the Government's request to include the key economic reforms adopted since early 1994 in the Assessment.  SENEGAL AN ASSESSMENT OF LIVING CONDITIONS ADDENDUM 1. In order to complete this Assessment, which was prepared in early 1994, this update enumerates a number of reforms that the Government of Senegal has adopted in the second half of 1994 in the areas of price and trade liberalization, private sector promotion, and labor market liberaLization. These reforms have been or will be introduced in the context of the Pnvate Sector Adjustment and Competitiveness Credit (approved in January 1995) and a proposed Agricultual Sector Adjustment Credit (June 1995). Many of these measures, potentiaUy beneficial to food crop farmers and small entrepreneurs, were recommended in this Assessment, completed in April 1994. Their adoption is a positive sign of the ongoing commitment of the Government of Senegal to pursue the adjustment process, improve the incentive and regulatory framework, and more important, elaborate policies that stimulate productive, income-generating activities of the poor and the most vulnerable groups. Government's recent initiative (Poverty Workshop in March/April 1995) to elaborate a poverty reduction strategy and discuss it with civil society at lage (including NGOs, women and youth groups, and unions), so as to present it at the Consultative Group meeting in July 1995, is another indication of this commitment and a positive step in fostering ownership. Recent Reforms 2. In an effort to eliminate remaining barriers to free domestic and import trade, the Govermnent has taken the following actions: 3. Price controls: liberalized the prices of all goods and services, except for those deemed essentiall; the price of domestic rice has been freed and administered wholesale prices for imported broken rice have been eliminated; the price of wheat flour will be freed in Novenber 1995. The producer prices of cotton will be determined as a floor price by a committee including farmers and marketing agencies. This measure will be extended to groundnuts at the beginning of the 1994/95 harvest period. 4. Non-tariff barriers on imports: eliminated the prior authorization to import on all products (except gold and silver bars and second-hand clothes), including on imports of popularly consumed food items, such as potatoes and onions (December 1994) and bananas and sugar (January 1995). The removal of prior authorization for wheat flour imnports has been agreed upon, but will only become effective in November 1995, while that on rice is being phased out. Protoction of local paddy production will be ensured by an appropriate tariff structure (at present a 16% inport levy) and by tariff surcharges on imports in the event of an extraordinary decline in international prices, so as to safeguard against dumping practices. 'These gooes and services are: bread, locally-produced pharmaceutical dmgs and health services, butane gas and charcoal (for environmental reasons), petroleum products, urban transportation, transport services, electricity, water, and telephone.  5. Special agreements: dismantled a number of monopolies by terminating special agreements (Conventions Speciales or Protocoles dAccord) granted to selected enterprises: SOCOCIM (cement), ICS (fertilizers), SOTEXKA and ICOTAF (textiles), SIPS (packaging), SODEFITEX (cotton), GMD and SENTENAC (wheat flour), SNTI and SOCAS (tomato paste), and SOCOSAC (packing bags). All privileges were also terminated, with the exception of import tariff exemptions granted to ICS and SODEFITEX (essentially two exporting companies) for their inputs. Temporary tanff surcharges, decreasing over three years (end-1997), are applied to three products (tomato paste, wheat flour, and polypropylene bags) in order to allow time to local producers to face foreign competition. A new agreement was signed with CSS (sugar), which eliminates all monopoly privileges on the production, importation, processing, refining, and distribution of sugar, removes all special fiscal and tax benefits accorded to CSS, and liberalizes prices and margins in the sugar market. It is hoped that the removal of these monopoly arrangements will increase competition, thus having a favorable impact on prices of basic consumer goods, and improving the purchasing power of poor households. As the Assessment argues, moving towards a liberalized system of imports and production of rice might lead to greater competition in the importation and marketing of rice, possibly resulting in lower or constant prices because of gains in efficiency. (para. 48-54) 6. Promotion of exports. The Government has also introduced measures to promote exports, including the elimination of prior authorization on the export of cereals, confectionery groundnuts, fresh tomatoes, tomato paste, and all industrial products. 7. Impact of the 1994 devaluation. On January 12, 1994, the Government decided to srenghten its adjustment strategy in coordination with other governments of the CFA zone by adjusting the fixed parity of the CFA franc from 50 to 100 CFAF per French Franc. This decision represented a major step to launch a broad develoment agenda, aimed at accelerating growth in a region which had experienced a serious economic decline. In a small low-income country, such as Senegal, export-led growth is a key pre-condition for accelerated, sustained improvements in living standards and reduction in poverty (others are lower population growth, and more equitable access to basic services). 8. The devaluation has begun to benefit groundnut and cotton producers and certain industries, such as fisheries, cement, textiles, and tourism. Producer prices of groundnuts and cotton have each increased by 50% since 1992, thus having a positive impact on real income of farm families. The impact on consumer prices, especially rice, was dampened immediately after the devaluation as the price increase was limited administratively. Since their liberalization, rice prices have remained at about the same level as the former administered prices, and are expected to stabilize slightly above that level. Consumers, however, have suffered from the increase in rice prices. In fact, surveys. indicate that a switch to traditional cereals away from nce is taking place in many parts of Senegal, and that poor households in urban areas have reduced the number of daily meals, both factors contributing to an increase in malnutrition since January 1994. It is expected that many of the policy reforms supported by the Agricultural Sector Adjustment Credit will contribute to soften the impact on prices of agricultural commodities and minimize the likelihood of disruptive nmrket failure (e.g., the privatization of the groundnuts and rice sub-sectors and the liberalization of the sugar sub-sector). It is clear, however, that these measures will not be sufficient to address the potential negative impact of the devaluation and of these measures on the most vulnerable groups. Thus, targeted interventions that ensure that households at risk of nutritional inadequacies be monitored and receive assistance need to be put in place.  9. Incentive Framework. In order to give enterprises more flexibility, the Govemment also adopted in November 1994 a revised Labor Code, which abolishes the requirement of prior approval by the Labor Inspector in cases of layoffs due to economic hardship or internal restructuring of enterprises. It is expected that this measure will contribute to the creation of jobs. 10. Participatory Approach. With the intention of strengthening the ownership aspects of its economic program, the Government of Senegal is venturing into a process of internal consensus- building, through which representatives from the private sector, labor and the civil service participate in defining and shaping economic and sectoral policies. Such an approach was used in elaborating the reform program of the Private Sector Adjustment and Competitiveness operation and in developing the reforms in the higher education sector. Moreover, in an effort to disseminate information on policy reforms and to explain their rationale, the Government will conduct a two- year communication campaign starting in 1995. 11. Bank Portfolio. The Bank plans to launch a Community Nutrition project in July 1995. This is a poverty-targeted intervention which aims at halting the deterioration in the nutritional status of the most vulnerable groups (malnourished children under three years of age and pregnant and nursing women) in targeted poor urban neighborhoods, providing potable water to unserviced neighborhoods of the nutrition program, and enhancing household food security in poor rural and urban areas during critical periods of vulnerability. This pilot operation utilizes an innovative approach in that it will attempt to deliver community nutrition activities through a private-operated agency, rather than a ministry, that delegates contract management to local entrepreneurs (such as community groups, local NGOs). It is hoped that this approach will not only build capacity among these community groups to deliver these activities more efficiently and cost-effectively, but will also promote community participation, thus ensuring long-term sustainability.  EXECUTIVE SUMMARY# "Goorgoorpufu the wrd in wolof for old man A modem day, it means doing your best Most recently, i signifies "fguring ow a sq to survive through any mecas in tody's economuc crisis" 1. Senegal is a lower middle-income country with a population of 7.4 million, and an average per capita income of US$710. As the former colonial capital of French West Africa, Senegal has historically represented an economic/trading pole in the region, with the largest economy in the Sahel. However, over the past decade, dramatic declines in formal sector activity have combined with longer-term trends (population growth, land degradation, low rainfall, declining world commodity prices ) to plunge the country into a severe, and prolonged, economic recession. Ten years of adjustment and stabilization (from the early 1980's to the early 1990's) produced some key reforms (partial liberalization of agriculture, and trade, elimination of price controls, public enterprise reform, restructuring of the banking system), with particular success in fiscal stabilization through a fairly severe contraction in expenditures. But, this partial structural adjustmnent did not produce the type of economic growth which would have a large impact on poverty. In addition, the policy of internal adjustment aimed at reversing the appreciation of the real exchange rate after the mid-1980's has had an adverse impact on the rural poor. 2. The January 1994 devaluation of the CFA franc represents a substantial departure from past policies of internal adjustmnent. This change in policy offers an opportunity to improve rural incomes through higher prices for the most important revenue source of the poor - groundnuts - and to jump start the economy by increasing competitiveness and growth. This next year will be a crucial time; either the GOS and its social partners will seize the opportunity provided by devaluation to implement policies which maximize the net benefit to the rural population (passing on producer price increases for goundnuts, protecting key social expenditures, reducing regressive levels of taxation, while balancing fiscal and credit control to maintain inflation), or, shorter-term considerations (maintaining the real level of the civil service wage bill, meeting the demands of university students instead of increasing allocation for secondary school education ) will prevail to the detriment of long-term growth and poverty alleviation. 3. What are key elements of a successful poverty alleviation strategy? The region in the world which has had the most success in alleviation poverty is East Asia. What were some key differences in their strategy compared to other regions where economic growth has been strong but where poverty reduction has not been as substantial? The primary elements of the strategy (whether implicit or explicit) were to provide universal primary and secondary education thus narrowing the gap between urban and rural areas and between boys and girls; achieve an early demographic transition through investing resources in family planning; close the income gap between urban and rural areas by investing in basic infrastructure in rural areas (water, sanitation, roads), and by keeping the implicit or explicit level of taxation of the rural sector low (including maintaining a competitive currency, avoiding inefficiencies resulting from monopolistic control over agricultural inputs which could increase prices); and, of course focusing on growth-oriented macroeconomic policies. If the experience in Asia is transferable to Senegal, this implies that the recent devaluation may substantially reduce implicit levels of taxation of the rural sector, and that several distortions in the credit market and in the persistence of inefficient monopolies on basic goods need to be removed, while reforms in improving equitable access to basic health and basic education (particularly in the rural sector ) are essential. This Executive Summary was written in early 1994, prior to the devaluation.  v 4. Profile of Poverty. Until the 1992 Priority Household Survey (PS), it was inpossible to state with confidence how may households were poor. The profile presented here constitutes only a first step in assessing poverty, and thus provide only a fraction of the possible insight to be gleaned from this rich data source. Qualitative studies as well as debate would validate and provide a more accurate interpretation of some of the findings. The profile shows that during the harvest season in 1991/1992 (the most prosperous time of year), 33% of the entire population, about 2.4 million people, did not spend enough money on food (adjusted for estimated levels of auto-consumption) to assure a minhimum caloric intake of 2400 calories per adult equivalent per day. In spite of the general sentiment in urban areas that living standards have deteriorated (googorlu), poverty is still largely a rural phenomenon with over 80% of poor households in rural areas (79% of individuals). However, Dakar is the exception to this rule as it hosts such a large percentage of the total population (21 %) that it also has a significant proportion of poor households (12%). 5. Historically, many donors, NGOs, and Senegal's food security monitoring agency (CSA) focused programs on the drier north of the country which was often associated with low-incomes, due to its poor climate. However, recent data show that in 1992 poverty was most concentrated in Kolda, Kaolack, and Fatick - more southern areas- where over 40% of poor households reside. It is most severe (highest percentage of population below the region-specific poverty line) in Ziguinchor - although it is possible that civil unrest temporarily prevented some farmers from planting and depressed tourism revenues - and in Kolda. With a deteriorating national economy, access to remittance income may play a more central role than previously assumed in raising households above the poverty line; thus, a key to the rdatively higher incomes in regions such as Saint Louis and Diourbel can be found in the stronger traditions among resident ethnic groups such as the Soninke of emigradng abroad. Although farmers and independents are more likely to be poor than formal sector workers, low-income individuals are not generally concentrated in one particular trade. Instead, it is the low level of capital used in the trade that distinguishes the poor (farming with little equipment, selling Cola nuts on the street, serving as a maid). Poor households may also be more likely to have a smaller, less connected family network than the non- poor. The reason for this is that the most important asset to many Senegalese is an extended and well- placed family network from which one can derive jobs, credit, and financial assistance. Participating in ceremonies represent one of the means of "maintaining" this asset. However, pressure on the lucky few in the formal sector to support the many who remain without employment has increased dramatically, this weakening the viability of this source of support, particularly in urban areas. The poor spend a higher percentage of their income on rice than on any time, and on average receive more income from groundnuts than from any other single source. 6. The urban poor (just over 500,000 people) most often live in what has been termed "floating neighborhoods" or unplanned commnunities in peri-urban areas in crowded, unsanitary, housing. Urban poverty has also likely increased in recent years as the land pressure, and a three-year drought in Fatick and Kaolack, as well as economic problems in neighboring countries, push more and more desperate migrants in search of work. Along with about 60% of the active population, the poor are much more likely to work in low-skilled activities in the urban informal sector. This sector is beginning to show signs of saturation for lower-skilled employment. Two signs of this strain are shrinking firm size and a high rate of underemployment (for example, maids being asked to work fewer hours, laundresses may turn to clandestine prostitution, crime, or begging. Unfortunately, one symptom of the economic decline is the growing number of children begging in the street, which UNICEF estimated at up to 100,000 in 1991. 7. The past tax and tariff system has not been consistent with redudng the proportional burden on the poor. One of the "hidden" costs of the economic decline of the past decade was an increasing  vi dependence on regressive taxation of basic goods like fuel, rice, and sugar. For example, petroleum "non-tax" revenue grew from zero in 1984/85 to 34 billion CFAF in 1989/90, while public expenditures on operations and maintenance in key sectors such as health and education dropped to unsustainable low levels. Other inconsistencies with poverty-focused taxation policy continue; while a number of monopoly arrangements continue to govern several basic consumption goods such as tomato paste and sugar, gas fuel - consumed primarily by the non-poor - was subsidized. If one includes only major consumption items, inefficiencies from monopolies, trade restrictions, and tariffs cost consumers between 12,000 and 14,000 CFAF per capita (89 biUion CFAF), or 2.5 - 3 times as much as is spent on health per capita or approximately 47% of the average expenditure of a poor person. However, these high tariff and tax revenues did not translate into the provision of more or better quality basic public services on a per capita basis since an increasing share of these revenues were funnelled into supporting the domestic price of rice while fixing a high price for consumers, needs to be re-examined in light of the high cost it has imposed on the poor, with most benefits accruing to those in the richer region of St. Louis. The recent devaluation, which has changed prices dramatically, provides an opportune context for revisiting this policy. 8. Poor households probably lost more than they gained from the past overvaluation of the real exchange rate, which made imported good inexpensive, and exported goods cheap. The primary reason for this is that key imported consumption goods (rice, sugar, oil) of the poor were heavily taxed while the key revenue source for the rural poor, groundnuts, did not receive a level of price support which would fully compensate for the estimated degree of overvaluation. Rural incomes thus generally declined. 9. Long-term trends in rainfal patterns, and population growth have also made it increasingly difficult to earn a living in the rural sector; as a result, environmentaly sustainable land-conversion techniques as well as sound incentives for preserving the long-term value of land can now be considered a prerequisite to preserving or increasing income opportunity in the rural sector. Over the long-term rainfall levels in Senegal have continued to decline so that by the 1980's over one half of the country's land was no longer viable for rainfed agriculture. At the same time, population density between 1976 and 1988 increased by 40% on average. The current incentive system governing land and natural resource use states that land not under cultivation can be "leased" by the state to outsiders or others, without necessarily requiring permission form the local population. This practice only compounds the incentive to maximize land use, and creates a disincentive to planting forests, thus contributing to problems of soil degradation. Land availability per person has thus decreased substantially over the decade, prompting risk-averse farmers (often poorer farmers with no access to credit) to devote more land to growing food than cash crops (causing a subsequent decline in cash income). As the clearing of the Mbegue forest illustrates, it will likely be the larger, more influential networks of farmers who gain land as availability declines, while poorer more independent farmers will probably be squeezed onto more marginal lands. Downward trends in world prices for peanut oil have also contributed to declines in rural terms of trade index between 1987/88 and 1992/93. Following a disastrous experience with poorly managed large-scale agricultural credit, such credit is unevenly available, and linked to specific donor projects rather than economic returns; in one crop year, for example, almost 50% of total crop credit extended from only two projects, both located in the Fleuve region. Lack of both income and credit means that many farmers cannot afford seed and in desperation engage in detrimental activities such as selling equipment or other key assets to buy seed (i.e. in the region of Fatick). 10. The good news is that in spite of economic stagnation, average health and education indicators have improved over the past ten years; however, this has occurred through increasing reliance on external donors and on households who contribute twice as much as the GOS in terms of health care expenditures. In addition, with the exception of primary health care, the quality of many of  vii the services available to the poor has declined (with wealthier households in the formal sector protected by insurance). Expenditures on medicines for poor households (often on expensive brand-name medicines for lack of an alternative) far outweigh expenditures on health visits. This suggests that pautcal reform is at the heart of hnproving access to health care for the poor. Nonetheless, implementation of cost recovery calls for an examination, using the PS as a base, of the equity of the system of providing central govermnent matching funds on a regional basis, particularly in rural regions characterized by very low levels of cash income. Finally, given the strong correlation between family size and poverty (for a household growing from 10 to 11 persons per capita expenditure decreases by 16%), family planning remains a top priority where substantial progress needs to be made, particularly in light of land constraints faced in most rural regions. 11. Some traditional coping strategies for dealing with a loss of income, or a year of poor harvests, have been constructive: emigration abroad which often brings higher income, diversification of income sources, pooling even small amounts of capital with others to start microenterprises. But, many will ultimately lead to a decline in long-term income or welfare: seeking poorly paid off-farm income in desperation, gathering and selling charcoal (which can further deplete the natural resource base), skipping meals, turning towards less nutritious foods, cultivating marginal lands, emergency borrowing at high interest rates, resorting to prostitution, and sending children to beg instead of to school. Targeted programs for the poor should help people avoid entering such a downwards spiral. 12. The few formal programs or waivers of fees offered to the very poor in Senegal tend not to have very transparent procedures for targeting and verifying who is the poorest. Non-poor households receive more money in "institutional" transfers (i.e. scholarships, aid to orphans, pensions, etc.) than the poor. This suggests that there may be substantial inefficiencies in this system. NGOS have gained experience in targeting, although efforts are scattered. Many NGOs are currently seeking better income generating activities for themselves as well as for their clients, in order to decrease dependence on external finance. An overall positive strategy which has emerged from the economic decline is the increasing adoption of income generating activities for schools, women's groups, charities, and village associations to increase independence from the vagaries of state or external financing. 13. While the Bank is ready to provide assistance in the elaboration of a more focused and cohesive strategy for poverty alleviation, the commitment towards such a strategy needs to come from Senegal. Setting up an institutional "focal point" would be one short-term step which could improve the likelihood of elaboration and ownership of such a strategy. It is important to note that without a change in key trends in terms of economic growth, population growth, or the equity with which basic services and credit are allocated, poverty will Increase to 60% of the population by the year 2015. The Government adopted a population policy in 1988 with the objective of slowing down the population growth in harmony with its economic growth; population growth has shown some decline, (from 3.0 % in 1980 to 2.6 % in 1994, and fertility has decreased from 7.1 in 1975 to 6.6 in 1986, but still remains high at 5.9 births per woman in 1994). Thus, there is a certain urgency to defining a cohesive and focused strategy to improve the likelihood of growth with equity. The solution will not be found simply in pushing more donor assistance; Senegal has received more aid per capita than many other countries. The problem lies more in the distribution and efficiency of that aid in reaching the poorest, and in creating a "voice" for low-income groups through better accountability in targeted and public programs. Some key elements to alleviating poverty over the long-term include adopting policies to improve competitiveness of the economy while addressing some of the tax and pricing distortions which harm the poor, expansion of family planning, particularly into rural areas, and continuing and expanding programs to increase the percentage of literate women. Over the medium to short-term policies to improve the situation of the poor include monitoring cost recovery measures in the three regions identified poorest,  viii and sponsoring an information campaign on generic drugs in poor regions while accelerating the implementation of cost recovery for less expensive generic drugs in health posts. In addition, AGETIP should proceed as quickly as possible with its planned expansion into rural areas during the hungry season, technical assistance to communities wishing to pursue insurance schemes based on tontines could be explored, and beneficiary assessments should be required of any project claiming to target the poor. Finally, participation in projects needs to change from rather passive participation in many government projects, to more active participation of project beneficiaries (including the poor) in project design. Through these measures, accountability of project claiming to help the poor would be tightened using objective criteria and data sources, or more innovative methods of targeting. 14. Organization of the Report. Chapter one starts out by exploring what is known about the expenditure levels, distribution, consumption and revenue patterns of the poor. It is followed by a discussion of the evolution of living conditions, and our key factors influencing income of the poor (opportunity in the informal sector, access to credit, some key elements of tax policy, and the agricultural sector's ability to provide income). Chapter three examines the public sector's contribution to the other aspect of well being - access to social services (health, education) as well as major public expenditure issues relevant to the poor in the agricultural sector. Chapter four briefly touches upon some of the formal safety nets provided by the GOS and NGOs, and chapter five presents some of the key issues and options relevant to elaborating a poverty alleviation strategy in Senegal.  Chapter / - Profle of the Poor 1. POVERTY PROFILE "Lekko, nano sax nelaw dovone" Youir huinger is nev er satiated, Your thirst is neuer quenched, You can never sleep until you are no longer tired" response to rhe question of how living conditions have evolved.' 1.1 The Demographic Context 1. With a 1993 per capita GDP of USS710. Senegal is considered a lower-middle income country. Although it enjoys higher income levels than its Sahelian neighbors, it nonetheless faces many of the same constraints. including: dry arid land, low rainfall, a population which doubles every 26 years, low world comnmodity prices for groundnuts, its major export. which accounts for 60% of farm cash income, and recent economic stagnation. Senegal is more urbanized than other Sahelian countries, with 40% of the population currently in urban areas (about 21% in the capital city Dakar), and with the majority of the rural population (53 %) residing in the agro-climactic region known as the Peanut Basin. Approximately 41 % of all households are of the Wolof ethnic group, followed by 24% Pular, and 147% Serer. Although basic health and educational status of the population has improved over the past ten years (as will be seen in chapter 3). life expectancy (48 years). chronic malnutrition (affecting 23 % of children age I to 5), and other health indicators remain below those for other lower-middle income counries. Diseases such as malaria. diarrhea, and respiratory infections continue to kill, and the potential devastation of AIDs looms on the horizon. 1.2 Defrining Poverty 2. As one of the first African countries to embark on a structural adjustment program, Senegal has substantial experience debating the social costs of adjustment. However, this debate has at times not distinguished between those vulnerable to changes in govermnent policy and those who by some objective standard can be considered poor. The purpose of the next three chapters of this report is to examine who can be considered poor using the best objective measures of poverty available, and the variety of factors, including but not limited to structural adjustment, which have affected living conditions in the past and which will continue to influence them in the future. 3. A first step is to distinguish the poor from the vulnerable. One of the most visible segments of those vulnerable to the changes introduced by adjustment measures - retrenched civil servants and formnal sector workers - have often been at the center of the national discussion of the social costs of adjustment. Although civil servants and formal sector workers have been partially successful at maintaining their living standards over the past five years, they are finding it increasingly hard to do so in an environment of economic decline and increasing pressure to support unemployed dependents. While public employment has not fallen significantly, as of December 1993, government payrolls were becoming more and more difficult to meet in a timely manner. More relevant, however, is that the recent devaluation of January 1994 will reduce the real income of this group, making them vulnerable to this adjustment. Nonetheless. it is important to keep in perspective that while those in the formal sector may have lost from downsizing, recent household-level data shows that in 1992 they were the least likely to fall among the poor as defined here. 4. If one defines the poor as those who are unable to maintain a minimum daily caloric intake (2400 calories per adult equivalent per day) after adjusting for estimated levels of home consumption, then 33 % I Unofficial translation from French adapted from Daouda Diop. Secteur InformallSeceur Refuge, Phinomene de la Pauvrere (Dakar et Banlieu). Republique du Sendgal. Primature. Commissariat Gndmrae i L'Emploi. 1991.  C7taoter I - Profile tf tle Poor 2 of the Senegalese population fell below this minimum at harvest rime in 1992.- This "snapshot" confirms ,hat. as in other Sahelian countries, the majority of the poor live in rural, not urban, areas (over 78 % of the poor reside in rural areas although only 60% of the population is rural). The method of calculation of the poverty line is based on region-specific poverty lines, and is explained in detail in Annex A.3 This profile uses data from the first Social Dimensions of Adjustment household survey which examined numerous household-level characteristics (expenditure, income, health, education, etc.) of 10,000 households nationally and from an agricultural household survey (ISRA/IFPRI 1988-91) which carefully followed a small number of households (approx. 290) over two years in the Peanut Basin and part of Senegal Oriental, collecting information on household characteristics, prices, and pattems of home consumption. This profile presents only a small portion of the analysis possible from these rich data sources, and would greatly benefit from complementarv qualitative information in the future to explain and confirm some of the results. 5. It is essential to remember that this is only a snapshot, and that dramatic changes in income levels over seasons and between years are perhaps as important as the actual level of poverty; for example, between one good and one bad harvest year incomes varied up to 78% with livestock commonly used as an alternative income source in poor harvest years (IFPRI/ISRA, 1993). To illustrate the impact, if the level of auto-consumption were to decrease bv 30%, as could be the case during the "hungry" season, or during a particularly bad harvest, the Box 1.1 percent of the population falling into poverty Alternative Approaches to Poverty Lines would jump from 33% to 60%. This sensitivity stems from the fact that manv of the mral poor If one took two-thirds of the mean expenditure level are clustered close to the poverty line. (as Boeting, et al have done in Ghana in 1990) as a poverty line and one-third as a hard-core poverty 6. The urban-rural gap: As expected, not only line, instead of defining an absolute 'minimum' are there more poor in rural areas, but there are standard of living, the poverty line would increase many more "hard core poor" as measured by the to 5,344 CFAF per month nationally, with 53% of higher poverty gap and Foster Greer and the population falling into poverty, and 2,672 Thorbecke (FGT) index (the higher the FGT index, CFAF or 26% of the population considered hard- the more severe the poverty). This disparity holds core poor. If one followed the approach taken in true for other measures of well being such as the World Development Report of Sl per day, health, access to water and sanitation, and (unadjusted for the value of home consumption not education. Income is distributed less equally in included in the Priority Survey and not based on dakatin. Income asdis asured by e In purchasing power parity) then the poverty line Dakar than in rral regions as measured by the gini would equal 7,500 cfa per month and 61 % of the population would be considered poor. 2 USSI - 264.5 CFAF (1992 exchange rate for 1992 data). 2400 kcal per adult equivalent represents 80% of the 3000 kcaliadult equivalent norms-recommended by the Organisme de Recherches sur l'Alimentation et la Nutrition Africaine (ORANA) which was also used in the IFPRI/ISRA study. 3 The following estimates of poverty are based on dama from the Enquete Sur Les Priorites, funalized in February 1993 and conducted in 1991/92 by the Direction de la Prevision et de la Statistique, Ministere D'Economie, de Finance et du Plan. An "absolute' poverty line, or some fixed measure of expenditure required to meet basic needs (here caloric intake), is used as the basic method throughout this report. This approach is a variation of what Ravaillon (1992) calls the "food energy method.' The absolute poverty approach also has shortcomings in that is requires judgement as to what is most universally considered "minimum". Poverty is inextricably linked to factors beyond caloric intake, based on culture. religion, and individual perceptions and attitudes. It is subjective as well as objective.An alternative approach to estimating poverty would be to focus on relative welfare to see how far away each individual is from average income levels within a country with no judgement as to whether this average is "sufficient" to meet basic needs. However, with this method it may be hard to discern welfare changes over time as average incomes increase. Also, if there are a large number of poor people in a country. poverty could be underestimated.  Cliawter I - Prolfile of the Poor 3 Table 1.1 i992 Poverty Lines coefficient. For comparison. income distribution _ in Senegal is less equal than in India or Cote CFAF Per Poverty Mean Mean D'lvoire. but more equal than in Brazil (Annex Capita Month Line Expenditure Expenditure Ah- Poor Dakar 5.610 16.094 4.825 7. Difference in Poverty Among Regions: [n a Other Crban 3.971 10.33S 3.348 country dependent on rainfed agriculture, it would Rural 2.651 4.154 1.845 be normal to assume that the regions with good Senegal 3.324 8.017 2.247 rainfall would be the best off. However, Map 1.1 reveals one of the most counter-intuitive findings of the poverty profile, included among the regions with a high prevalence of poverty are regions with better rainfall (Kaolack, Kolda. and Fatick where 41% of all poor reside). Conversely, outside of Dakar. the two richest regions (Saint Louis, Diourbel) are in traditionally drought-prone areas, although there have been large-scale investments in irrigation schemes along the river basin in Saint Louis over the past ten years. 8. What factors might explain these results? Probable factors include the high level of monetarization, dynamic commnerce, lucrative trading activities of the Mouride brotherhood in Diourbel, and the historic emigration of certain ethnic groups (Sonink6, Toucouleur) from Saint Louis and generally more diversified income sources in these regions. While Sahelian residents in poor agroclimatic areas have traditionally diversified income to insulate themselves from the full impact of drought, this pattern Table 1.2: 1992 Poverty Indicators Incidence of Number of Gini Poverty Gap FGT Index Poverty Poor Index (Household level) Dakar 16.42% 276,552 47.4% 3.1% 0.87% Other Urban 16.29% 251,782 40.9% 3.9% 1.4% Rural 40.44% 1,875,503 42.0% 16.4% 9.1% Senegal 33.4% (29%) 2,404,800 (52.1%) 13.9% 7.7% 4 Although the interpretation of Gini coefficients from PS data is problematic (as the survey measured only expenditures and not home consumption), other work in Senegal confirms the trend towards greater inequality in urban areas and towards a Gini coefficient of at least .4.- A smaller recent agricultural household survey (IFPRIUISRA. :993) found gini coefficients ot between .35 and .45, with a more skewed distribution in the cities of Tambacounda and Kaolack compared to rural areas. Another study of rural income distribution in the early 1980s revealed a very high gini coefficient (.57) for the rural sector based on a simulation (Rural Income Distribution in Senegal:Changes and lndicators, Gerard Chambas, CRED, University of Michigan. Dec. 1983, Discussion Paper No. 106); another study showed a much lower gini coefficient of .3 for rural areas, .4 for urban areas, and .43 nationally in 1975 (BIT-PECTA: Disparitrs de Revenues Entre les Villes ct Les Campagnes au Scnegal. Addis-Abeba, 1982). 5 The poverty gap index reflects the distance of the average consumption of a poor person from the poverty line. However, the poverty gap index does not provide information on the distribution underlying this average (i.e. one person could transfer income to another poor person and become less poor and the measure would not change). "Severe" is used here to mean the distance of an individual from the poverty line, or the degree of poverty as measured by the Foster Greer Thorbecke index. For an in-depth explanation of these indicators see Ravaillon, Poverty Comparisons, and Kanbur, Ravi, Poverry and the Social Dimensions of Structural Adjussmnent in Cote D'Ivoire. SDA Working Paper Series. World Baant, 1990.  Chanter I - Profile of t/c Poor - 4. confirms the anecdotal - that such income plays a Table 1.3: The Urban/Rural Gap key role in determining whether or not one falls into poverty. Thus, those populations which have U,rban Rural already established strong links to relatives who have emigrated abroad appear better insulated [nfant Mortality Age 0 -5 70 102 from the wide swings in rainfall patterns than Ag , . 23% 33% populations in agroclimactically richer regions Primary School Enrollment -99% 38% without culturally enmeshed traditions of diversification.5 Certain parts of the Senegal River Basin are the most obvious testament to this; in 1984, for the Toucouleur of Dagma (in the Senegal River Basin) transfer remittances reportedly represented 35% - 40% of a family's annual revenue.7 Priority Survey data confirmed this trend, showing substantially higher revenues from transfers for the Sarakhole-Soninke (residing mostly in St. Louis) than for the Wolof (390,000 CFAF versus 188,000 CFAF per households per annum). In addition, the three regions which received the largest proportion of 70 billion in annual transfer income recorded in the PS are also among the regions where the incidence of poverty is lowest - Dakar, St. Louis, and Diourbel (receiving 30%, 17% and 13% of total transfers respectively).' The poverty profile also shows that Dakar hosts a substantial number of poor (12%), in spite of high average income levels. In Ziguinchor a large percentage of the population falls below the poverty line, but it is difficult to determine whether this reflects previous civil strife and the consequent inability of some farmers to plant crops more than a long-term pattern. 9. A relevant question regarding the validity of these regional differentials is whether they reflect one specific year's rainfall and crop patterns, or, longer-term trends in welfare levels. Since only one Priority Survey has been completed to date, one way to "check" these results would be to look at long-term indicators other than income which may shed light on quality of life (tables presented in Annex B). For example, Diourbel, Tambacounda, and Kolda have the highest rates of chronic malnutrition (stunting) of children under 5. Using numerous such indicators other than consumption levels (vaccination, stunting, households per school, access to potable water), Tambacounda and Kolda rank among the regions with the worst overall living standards.9 1.3 Do the Poor Consume Different Goods than the Non-Poor? 10. In many countries, government policies have provided targeted subsidies to the poor by identifying "inferior" goods that only the poor would consume (i.e. cassava in some African countries). In other countries, such as Tunisia, goods have been expressly made inferior and superior through different packaging and quality. In Senegal, at the national level existing data show no one "inferior" good for which quantity consumed increases as income level decreases, although in urban areas anecdotal evidence suggests that day old or week old bread is commonly consumed by the poor. While it is 6 The IFPRI/ISRA study found that in Tambacounda, an area with less of a tradition of migration, income from migration was actually negative, suggesting the possibility that those who had migrated were unable to find lucrative employment. 7 Horowitz, Salem, and Murdock, 1992. 8 However, one weakness with this data is that it includes transfers in polygamous households - from husband to wife for example when the wife resides in a separate domicile. The regional distinction still is valid as there are actually fewer polygamous households in the northem region of St. Louis than in some of the other regions with lower interhousehold transfers. 9 If one takes shorter-term indicators such as seasonal malnutrition (wasting) the highest correlation is not with consumption levels, but with levels of female literacy and Saint Louis, and Diourbel, in spite of high levels of income, rank lowest.  Clhapter I - Profite of(the Poor 5 Table 1.4 NMonthiv Expenditure Shares Expenditures Poor Non-Poor FOOD C70 %) Rice 29.71% [9 24%c Sugar 12.96% 7,68 % Oil 11.20% 10,57q Condiments 9.81 % 11.42%I' Fish 9.22% 1 1. 727c Tea 5.75% 3.61 % Millet 5.45% 4.52 % Other Combined 15.89% 31.24% 100% 100% NON FOOD (30%) Health 17.13% 8.54% Clothing 9.53% 11.84% Transport 7.77% 5.67% Education 2.02% 3.04% Other Combined 63.55 % 70.91% 100% 100% Source: Priority Survey Data Table 1.5 Four Most Important Cash Income Sources Ranked Poor % Income I Non Poor % Income Dakar 1. Non Ag. Activities - A 27% 1. Private Salaries 31% 2. Private Salaries 24% 2. Non Ag. Activities - A 22% 3. Transfers 12% 3. Public Sector Salaries 22% 4 Other combined 29% 4. Other combined 25% Other Urban 1. Non Ag. Activities - A 31% 1. Public Sector Salaries 27% 2. Transfers 15% 2. Non Ag. Activities - A 26% 3. Private Salaries 13% 3. Private Salaries 16% 4. Other combined 41% 4. Other combined 31% Rural 1. Groundnuts 28% 1. Non Ag. Activities 31% 2. Non Ag. 28% 2. Transfers 20% 3. Transfers 15% 3. Private Salaries 10% 4. Other combined 29% 4. Other combined 39% Source: Pnonrty Survey, GOS  Choater I Profile opf ttie Poor 6 Figure 1.1 Senegal: Per Capita Monthly Food Expenditure as Percent Totai bv Expenditure Group V. _3 .s r'cn t J