Senegal - Economic trends and prospects (Vol. 4 of 4) : Human resources

Report No. 1720a-SE l U E C P The Economic Trends and Prospects of' Senegal (In f:our Volumes) Volume IV. Human Resources December 1979 Western Africa Regional Office FORt OFFICIAL USE ONLY Docu rnent of the World Bank This dchcurnent has a restricted distribution and rnay be used by recipients only in the performance of t ieir official duties. Its contents may not otherwise be disclosecl withoUit World Bank authorization.  CURRENCY EQUIVALENTS US $1.00 = CFAF 220.37 CFAF 100 = US$0.454 WEIGHTS AND MEASURES 1 kilogram (kg) 2.20 lbs 1 millimeter (mm) 0.04 inch 1 meter (m) = 3.28 feet 1 kilometer (km) 0.62 miles 1 hectare (ha) 2.47 acres GLOSSARY OF ABBREVIATIONS BCEAO Banque Centrale des Etats de l'Afrique de l'Ouest CEAO Communaute Economique de l'Afrique de l'Ouest CPSP Caisse de Perequation et de Stabilisation des Prix ECOWAS Economic Community of West African States EEC European Economic Community ICS Industries Chimiques du Senegal ONCAD Office National de Cooperation et d'Assistance pour le Developpement SIES Societe Industrielle des Engrais du Senegal SMIG Salaire Minimum Interprofessionnel Garanti SOFISEDIT Societe Financiare S&negalaise pour le Developpement Industriel et Touristique SONACOS Societe Nationale de Commercialisation des Oleagineux du Senegal SONEPI Societe Nationale d'Etudes et de la Promotion Industrielle UDAO Union Douaniare de l'Afrique Occidentale UDEAO Union Douaniere des Etats de l'Afrique de l'Ouest UMOA Union Monetaire Ouest Africaine VAT Value Added Tax GOVERNMENT OF SENEGAL FISCAL YEAR July 1 - June 30  FOR OFFICIAL USE ONLY THE ECON'OMIC TRENDS AND PROSPECTS OF SENEGAL Volume I Summary and Conclusions The Main Report Statistical Annex Volume II The Agricultural Sector Volume III The Industrial Sector Volume IV Human Resources This report is based on an economic mission which visited Senegal in November 1976 composged of Messrs. J. de Leede (Chief), J.C. Brown (para- public enterprises), B.. Horton (industry), Ms. K.M. Larrecq (balance of payments), H. Lubell (human resources), J.D. Shilling (economic modeling), D.R. Steeds (agriculture), and M.P. van Dijk (informal sector), and comple- iaented with findings of the Bank's regular sector work and more recent eco- nomic missions. This volume was written by Mr. Harold Lubell (ILO) on the basis of his participation in the basic economic mission in November 1976 and t:he following unpublished Bank reports: 1Migration and Employment in Senegal 1976 (de Leede, Elkan Bharier) Migration in West Africa 1976 (Zadiaziah, Nair) The Informal Sector in Senegal 1977 (van Dijk) ]Education Sector Memorandum 1977 (Verspoor). 'rhis volume was finali:zed in January 1977; it was updated in 1979 to include new developments. Working Papers on the balance of payments, public finance, and the informal sector are available in West Africa Programs Department II, Division C. The Para-Public Sector Report No. 1619a-SE can be obtained from the ]Reports Desk. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contenst may not otherwise be disclosed without World Bank authorization. W   THE ECONOMIC TRENDS AND PROSPECTS OF SENEGAL Volume IV. Human Resources Table of Contents Page No. SUMMARY AND CONCLUSIONS ..................................e.... . . .... .. i Population ..................... .. ........................ I...... i Income Distribution ..... i Nfigration .. . . . . . . . . .%. . . . . . . . . . . . . . .. . . i Informal Sector Development ...... ii Capital Constraints. . ........... iii Manpower Productivity ........... ......... . ... o, .... ... ... iii Chapter I. INCOME DISPARITIES AND LIVING STANDARDS . .1 A. Income Levels per Earner -o- . ... . 2 Disparities between Sectors... 3 Disparities within Sectors 5 B. Poverty Levels and Living Standards ..13 C. Incidence of the Government Budget. 16 D; Nutrition .l...*. . ... .*.....- ...19 E. Health ............24 Fo Education oo.-.-...... 30 The Demand for Education ..............31 Basic Education ..................... 32 Technical and Science Education at the Middle and Secondary Level - ..... ......... . . . 33 Vocational Training . .o.............. . . .....34 Training of Agricultural Technicians and Professionals ....... o ..... ... ........... .......... 34 Management Training Needs .. 35 Long-tern Planning ....... .... . ...... .36 G. Housing ..... ... . .................. 37 Chapter II. MIGRATION ..... .. . .... . o o ......... 41 A. Regional Population Distribution and Densities .41  Table of Contents (cont.) Page No. B. Ethnic Groups .............................................. 44 C. International Migration .................................... . 46 D. Internal Migration Flows ....... 48 E. The Senegalese Migrant ...... .............................. 53 Chapter III. EMPLOYMENT. ... .......................... 56 A. Employment Trends since 1960 ........................... 56 Modern Sector Employment ............................. 56 Agricultural Employment ..... ......................... 59 B. Employment Projections ...... .................... ............. 61 Population and Labor Force ..... ...................... 61 Urban Employment ................................................... 66 C. Employment Implications of Senegal's Two Growth Axes .... . ............ .................................... 73 Investment Costs per Job ............................. 73 Absorptive Capacity of Dakar ......................... 75 Chapter IV. THE URBAN INFORMAL SECTOR ............ .. .............. 79 A. Employment and Incomes in the Urban Informal Sector ......................................... 79 B. Inputs and Outputs of Informal Sector Producers .... ....... 81 C. Government Attitudes and the Institutional Framework for Policy toward the Informal Sector ............. ...... 83 D. Education and Training for the Informal Sector ............ 85 E. Assisting the Development of the Urban Informal Sector .............. ................................................. 86 Appendix A. SENEGAL'S MIGRATION FLOWS AGAINST THEIR HISTORICAL AND GEOGRAPHICAL BACKGROUND ........................ 89 A. The Regions .......................... 89 Groundnut Basin .. .................................... 89 The Senegal River Valley ..... ........................ 92 Historical Perspective ............................... 93 Modern Migration Flows .. ............................. 94 Casamance and Senegal-Oriental ..... .................. 97  Table of Contents (cont.) Page No. B. The Cities ....................... 98 Urbanization ............................ 98 Dakar ......... . . *. . .0............ 0.........I .. . 99 Saint Louis ............................ , . . . .. . . 103 Ziguinchor ............................ .104 Kaolack ............... , . **.... *...... I.105 S,tatistical Annex .... ......... . ..... ...... 106   List of Tables No. Page 1. Selected rural and urban incomes, 1974-1976 4 2. Gross domestic product (GDP) per capita and per worker, by province, 1975 6 3. Gross domestic pro,luct originating in agriculture and related activities per capita of rural popu- lation and per rural worker, by province, 1975 7 4. Workers and annual wages and salaries in modern sector by nationality, 1974 9 5. Average annual wages and salaries in the modern sector by skill level: Senegalese and other Africans compared with non-Africans, 1974 10 6. Average annual earnings in the modern sector by "macrosector" and by nationality, 1.974 11 ,. Distribution of employees and wage bill in government by salary bracket, COctober 1976 12 El. Distribution of households by monthly household income bracket, greater Dakar, February-March 1975 14 9. Distribution of households by monthly household expenditure bracket, greater Dakar, February-March 1975 15 10. Distribution of households by the level of monthly house- hold expenditure per consumption unit, greater Dakar, February-March 1975 17 11. Net transfers to producers and consumers out of the CPSP, fiscal years 1970/71-1976/77. 12. Nutritional value oE the diet per consumption unit per day of urban families in Dakar by level of total expenditure per consumption ullit, 1975 20 13. Price of selected food products per 1,000 calories and per 10 grams of prote:Ln 22 14. Average daily calor:Le consumption per capita by season in two selected villages in Thies and Sine-Saloum 23  List of Tables (cont.) Page 15. Medical personnel and inhabitants per unit: Senegal, Cap Vert and other provinces, 1966 and 1975 25 16. Health infrastructure: Senegal, Cap Vert and other provinces, 1974 26 17. Health infrastructure: hospitalization capacity, 1974 27 18. Projected distribution by level of education and type of public expenditure 36 19. Total population of Senegal by province and density, April 1976 43 20. Ethnic groups 44 21. Number of transients by place of residence and place of origin 50 22. Rates of migration by province for the 1960-1971 period 52 23. Estimated employment in the modern sector, 1960 and 1975 56 24. Establishments, workers and average earnings in modern sector enterprises (private and semi-private), 1975 59 25. Population and labor force (active population aged 6 years and over), selected years, 1976-1989 62 26. Population by sex and by zone (urban and rural), selected years, 1976-1989 64 27. Labor force aged 6 years and over by sex and by zone (urban and rural), selected years 1976-1989 65 28. Urban employment, Senegal and Cap Vert, 1976 67 29. Population, labor force and urban population, Senegal and Cap Vert, 1976 and 1981. 68 30. Employment in the modern private sector by branch of activity and nationality, 1975 71 31. Modern sector employment by skill level and by nationality (private sector), 1976 and 1981 72 32. Estimated employment in agriculture, animal husbandry and fishing in the Senegal River Delta, 1975 and 1985 74  List of Tables (cont.) Page 3:3. Investment costs per job in proposed projects and some international comparisons 76 34. Incomes per month .in selected activities, a comparison between Grand Dalcar, Medina and Pikine 82 List of Appendix Tables A-I. Sources of income in the Fleuve Region 93 A-2. Comparison of proportion of urban population, 1975 99 A--3. Urban centers of more than 10,000 inhabitants in 1976 100   List of Figures Following Page 1. Lorenz curve of the distribution of workers in the modern economic sector of Senegal by level of wages and salaries, 1974 (WB 21281) 9 2. Lorenz curve of thes distribution of government employees by salary brackets, October 1976 (WB 21282) 12 3. Lorenz curve of the distribution of households by monthly income bracket, Dakar, 1975 (WB 21283) 14 4. Lorenz curve of the distribution of households by monthly househoLd expenditure bracket, Dakar, 1975 15 (WB 21284)   HUMAN RESOURCES IN SENEGAL In a poor cotntry like Senegal, economic development is inevitably linked to the development of human resources. Economic development is achiev- able only if the potential of the country's human resources can be realized, i.f the labor force is enabled to work effectively, if living standards and economic productivity are raised, if the positive energies released by social change are harnessed - in short, if the goals of social development are reached. Those who wish to influence the process of human resource develop- ment must deal with three factors which are discussed in this volume: income clisparities and living standards; the migration movements of the rural popula- t:ion which these income disparities help to generate; and some employment and t:raining aspects of the manpower situation in the modern and in the informal sectors that these population movements influence and are influenced by.   SUMMARY AND CONCLUSIONS Population i. The official estimate of the population growth rate based on the April 1976 population census is 2.6 percent per year; it is projected to rise over the coming decades to a peak of 3.1 percent after the year 2000. Over the 1977-1981 period, the urban population will grow at 4.0 percent per year and the urban labor force at 3.8 percent per year. Of the 458,000 workers in the urban labor force in 1976, 30 percent (140,000) were employed in the modern sector, perhaps 20 percent were unemployed, and the remaining 50 percent were engaged in informal sector activities. The bulk of modern sector employment is located in Cap Vert. Of the urban labor force in Cap Irert, 43 percent (101,000) were employed in the modern sector, 37 percent were in the informal sector and 20 percent were unemployed. Rather optimistic projections covering tbe 1977-1981 Plan period estimated an annual increase of 4.5 percent in modern sector employment. If, as seems likely, the number cf persons engaged in informal sector activities increases at the same rate as the total urban population (4.0 percent per year), the rate of open unemploy- ment in urban areas will show a marginal decline. Income Distribution ii. Disparities of income and living standards are wide both between the rural and urban populations and within the urban population. Within the urban working population, incomes are particularly high for the expatriate non-Africans in the modern sector; but the relative position of the Senegalese and other African employees in the modern sector is also very favorable and was maintained by the Government's active minimum wage and price subsidy policy. Since 1975, however, minimum wages were not increased, and have diminished substantially in real terms. Yet the efforts to protect acquired benefits of employees In the modern sector have placed the work force in the informal sector (together with the unemployed) in the least favorable position in the urban economy. iii. Although the basic needs strategy should be directed principally towards the rural population, the residents of Dakar's inner and peri-urban slums should not be excluded from the target groups. A pilot urban project should be tried out, with slum upgrading, health, mother and child care, and education as main components, to test the feasibility of delivery systems to satisfy urban basic needs in Senegal's circumstances. With the Government's recognition of the desirability of a population policy for Senegal, a family planning element could be made explicit in the health component. Nkigration iv. Rural poverty and income disparities between rural and urban areas have led to an overlapping pattern of seasonal and longer-term migration from rural to urban areas that mark the Senegalese as among the more mobile of Africa's settled populations. Population pressure in the grotndnut basin is pushing people into the cities (particularly Dakar), while the Government  - ii - is attempting to attract some of the displaced peasants to colonization areas in Senegal-Oriental and to the Senegal River Delta. Proposed develop- ment of water storage and irrigation facilities in the middle valley of the Senegal River is intended to absorb the population already settled there and to discourage their out-migration. v. Major efforts to stimulate migration towards the southeastern part of the country will need to be preceded by a campaign to eradicate onchocerciasis (riverblindness) which will require World Health Organization intervention to negotiate, organize and finance part of an international program covering Senegal, Guinea, and Guinea Bissau. Priority should be given to the economic development of the region through promotion of cash crops instead of to migration of selected individuals from overcrowded parts of the country. Provision of infrastructure by the Government should be conditioned on the population growth of the villages, although in some less inhabited areas it may be necessary to start off migration by creating new villages for which a core group of families including a blacksmith and some artisans could be officially recruited. Farmers in such new villages should be encouraged to use hired labor and expand their farms. Rights to land and possible subsis- tence subsidies during the first year should be strictly related to manual land clearing and de-stumping in order to avoid speculative occupancy and destructive use of land. Consideration might also be given to larger-scale cooperative or Western plantation style farming, possibly organized by the religious brotherhoods. vi. Those who work in the urban informal sector and the urban poor have the closest links with rural areas. The reality of employment and incomes in the urban informal sector probably has had a greater influence on the rate of rural to urban migration than the mirage of high paying, secure jobs in the modern sector. Improving incomes and living standards of the urban poor is thus likely to increase the attraction of the city for rural migrants. Since first priority belongs to the rural poor in a basic needs oriented strategy in Senegal, an important objective of a coherent development program to promote rural development and rural resettlement is the prevention of accelerated urban drift. Informal Sector Development vii. The informal sector includes a variety of categories of small producers of goods and services including artisans, small traders, small-scale transporters, sellers of petty services and domestic servants. The artisans are the category about which most is known; they are also the group whose activities are most amenable to improvement through financial and technical assistance. The Societe Nationale d'Etudes et de Promotion Industrielle (SONEPI) has recently been charged by the Senegalese Government with the task of assisting artisans de production and artisans de service (as opposed to the artisans d'art). viii. A strategy for developing the informal sector should have several elements: a more positive attitude on the part of the Government toward informal sector activities; the flexible use of technical assistance, finan- cial assistance and additional training administered through a new sort of urban extension service which could be established using SONEPI as a base;  - iii - measures to increase demand by modern sector enterprises, the Government, and consumers for the outputs of the informal sector; and an increase in our knowledge of the sector. Capital Constraints ix. With underemployment of the rural population during half of the year and 20 percent unemployment among the urban active population, job creation is an important criterion against which new projects should be tested. This is particularly significant since capital is a scarce factor (although in the mediumn term the availability of good projects may be an even more important constraLnt). Investment costs per job in proposed projects show some striking differences between agriculture and industry, and within Lndustry. The recently cancelled Cayar phosphate/petroleum complex would have cost US$175,000 per job created, Dakar Marine US$73,000. The Kaolack lntegrated textile project would cost US$13,000 per job. Investment cost per job that would be created by a fairly labor-intensive approach to development of irrigated farming in the middle valley of the Senegal River is estimated at about US$2,000. :B For the development of the upper and middle valley of the Senegal River, a slow-phased scheme with an emphasis on labor-intensive agricultural techniques is recommended to prepare the farming population for effective use of the irrigation waters that will be made available by construction of the Manantali Dam. xi. There is a recognized need for good Senegalese managers in modern sector industry and admainistration, and to replace expatriate personnel. Good facilities are now in existence to meet demands from the modern sector i-or technical personneL already on the job. However, the general system of technical and vocational education is not at present turning out graduates considered by potential employees to be adequately trained. Manpower Productivity xii. Among the difificulties that face modern sector manufacturers in Senegal is a combination of relatively high wages and low productivity. So long as they operate ornly in a heavily protected domestic market, the less efficient firms can get by with protection from imports and with eventual subsidies from the GovEtrnment in the form of loans or equity participation, at the expense of public savings and Senegalese consumers. Relative ineffi- ciency has become a more serious problem since the Government can no longer afford these subsidies and it has become increasingly necessary for Senegalese industry to compete in export markets. xiii. Industrial wages are relatively high in two senses: (a) in compari- son with incomes earned elsewhere in the Senegalese economy; and (b) in comparison with real labor costs abroad, particularly in Senegal's major West African competitor, the Ivory Coast. The advantaged position of modern sector employees is, however, a fact of considerable political significance. Since it would be politically difficult to attempt to reduce wages, it makes mpore sense not to pass on the benefits of economic growth (modest as it is)  - iv - to urban consumers and to be concerned about increasing productivity for which ways can perhaps be more easily found. xiv. One factor in the low productivity in modern sector industry is sociological: the attitudes of the members of the modern sector work force toward industrial discipline. Absenteeism is said to be a major problem and some plant managers are now suggesting that it may be easier to adjust work procedures and personnel policy to local work attitudes than to struggle with changing local attitudes. They can take this position because of strong economic protection of their firms. xv. A way to increase industrial productivity would be to diffuse labor union insistence on maintaining redundant labor when it is demonstrated that a task can be done with fewer people. At the same time, as much flexibility as possible should be maintained for enterprises in the hiring of temporary (or even daily) workers. To convince labor unions to modify their stance, it would have to be pretty clearly demonstrated that the resulting increase in efficiency would attract enough additional entrepreneurs to increase the total number of jobs available. xvi. The positive aspect of having a growing modern sector in the urban economy is that the incomes of the wage earners serve to maintain a certain level of effective demand for consumer goods that are or can be produced by less advantaged workers in the informal sector. The share of these modern sector incomes that will be spent on goods and services produced by the informal sector can be influenced by widening the range and improving the quality of informal sector production, and by influencing middle income consumer tastes by propaganda and by example. xvii. The employment problem per se is serious but the major human resource constraint on development in Senegal is the shortage of managers throughout the economy, in industry, agriculture, as well as in the Government. A major constraint on industrial growth is the shortage of foremen and technicians, while the informal sector is hampered by the prevalence of primitive (rather than appropriately labor-intensive) techniques. The development of effective facilities and techniques for transmitting packages of skills through both formal and informal education is one of the country's priority needs.  Chapter I. INCOME DISPARITIES AND LIVING STANDARDS 1.01 Economic development in Senegal has meant a secular drift of part of the work force out of low-income, low-productivity rural-based agriculture Lnto higher-income, higher-productivity and largely urban-based industry, commerce and services. It was in the coastal cities of Dakar and Saint Louis that the European traders, soldiers and officials settled; it was to these cities that income flowed from abroad in the form of salaries and other French government outlays and from the interior in the form of the traders' margins on groundnuts for export. It was the relatively high incomes of the cities that provided the effective demand for the outputs of Senegal's incipient industries, and it was the cities themselves that soon provided the pool of relatively skilled and disciplined manpower needed by industry. At the end of lforld War II, Dakar was the primary location for French investments intended to enhance and exploit the city's role as economic capital of French West Africa. Although independence in 1960 meant a shrinking of Dakar's hinterland t:o within the borders of Senegal and a slowing of the pace of growth of urban incomes, the basic disparity in income levels between the coastal cities and the rest of the country continued to grow, with the increasing concentration of Government and import-substitution industry in Dakar. 1.02 Dakar not only generated the relatively high incomes of the modern sector, it also attracl:ed masses of the poor, most of whom ended up making their living outside oi the modern sector. Thus another basic intra-urban disparity was created between the limited-sized advantaged group with secure employment in the moderin sector and the relatively disadvantaged majority working outside it, in the informal sector. 1.03 Overlaying these disparities is the marked inequality of incomes inherent in the fact of contact between the expatriates working out of the European economy and the Senegalese working entirely within the national economy. These inequalities between non-Africans and Africans show up sharply in the modern sector where the bulk of the non-Africans are established. 1/ 1.04 Because of Senegal's relatively slow economic growth and probable overall decline in national product per capita of a rapidly growing popula- tion since 1960, income disparities have not significantly changed and increases in living statndards even for the bulk of those in the advantaged group employed in the mLodern sector have not significantly improved. In 1976, Government salaries hac!, for example, probably done no more than maintain their absolute standar6is of 1958 with a steady decline in more recent years. IBut disparities grow to the extent that other groups, especially some of the peasants, are worse off. There has been an increase in the absoLute number of people in the expanding; and relatively prosperous modern sector but since almost 70 percent of the total population still live in rural areas engaged in agriculture and related. activities, the relative disparities remain. 1/ It should be remenmbered that not all non-Africans have the specific juridical status of "expatriate" even though the terms are interchanged in everyday usage.  1.05 There is in Senegal, as in other African societies, an important cultural factor which partly offsets the statistical inequalities in the distribution of income: a keen sense of family and ethnic solidarity which performs the function of an informal system of income redistribution and social security. Before making too much of income inequalities within the Senegalese society on moral or on practical grounds, it should be remembered that the number of dependents upon each income earner, particularly in urban areas, increases markedly with the level of income, to such an extent that accumulation of individual or family savings is extremely difficult to achieve. Personal incomes are spread pretty thin in Dakar for example, where the average household size is 8 persons, a figure that probably does not include casual visitors who stay for dinner and the night. 1.06 It can be argued that the inequalities that exist today are an inevitable result of economic development from a low level: progress is made more rapidly in some sectors than in others and the beneficiaries of develop- ment are those who participate in the activity of the more dynamic sectors. There is, nevertheless, a growing consensus in the development community that the benefits of development should be more equitably shared than they are today, and that the satisfaction of the basic needs of all sections of the population should be made a central objective of development policy. Basic needs, as defined for the World Employment Conference of June 1976, 1/ include such material elements as minimum family requirements for private consumption of food, shelter, clothing, household equipment and furniture, and essential public services such as safe drinking water, sanitation, public transport, and health and educational facilities. Basic needs as defined by the ILO also include such non-material elements as the fulfilment of basic human rights, increased access to employment opportunities and improvements in working conditions. Building the satisfaction of basic needs into Senegal's develop- ment plan would, among other things, mean increased emphasis on rural develop- ment and on the extension of Government services to rural areas to reach the disadvantaged rural poor and to the peripheral slums of the capital to reach the urban poor. A. Income Levels per Earner 1.07 Although a complete picture of the distribution of income in Senegal cannot be drawn, enough information is available to sketch in the outlines and some of the details of such a picture. Estimates can be made of rural-urban income differences, of income differences between the modern sector and the informal sector in the urban economy, and of regional income differences. Hard data are available on earnings by skill level in the modern sector and on the distribution of wages and salaries of Government employees. The results of a recent household expenditure survey for Dakar are also available. 1/ International Labor Office: Employment, growth and basic needs: A one-world problem (Geneva, 1976), pp. 32-33.  - 3 - 1.08 An estimate of the distribution of personal income in Senegal for 1960 which keeps reappearing in the literatures 1/ has a Gini coefficient 2/ of 0.56. The degree of income inequality as measured by the Gini coefficient will normally be greater for the society as a whole than for any of its parts; the Gini coefficients calculated for the separate sectors within Senegal quoted in the discussion below range between 0.21 and 0.43. Disparities between Sectors 1.09 The broad orders of magnitude of significant group income differ- ences per worker may be lerived from Table 1 which summarizes more detailed data presented further o01. The most striking difference is, of course, between the earnings of non-Africans ("expatriates") and those of Senegalese and other Africans in the modern sector of Senegal's economy: in 1974, for alL skill levels combined, the ratio of earnings of non-Africans to Africans in the modern sector was 7.6:1. A similar comparison for the Ivory Coast on the basis of 1970 estimal:es 3/ yields a ratio of 6.4:1. 1.10 A major factor affecting rural to urban migration should be the diiference between the levels of income that can be obtained in the city and that can be obtained in the countryside. For the average unskilled and probably illiterate ruralL migrant, obtaining one of the higher paying jobs in the modern sector which require some skills would not appear to be a rea- sonable objective. A rural migrant to the city is more likely to have as an income target the wage level of an unskilled worker in the modern sector or the earnings level of a self-employed craftsman or a casual worker in the informal sector. For 1975, an average income per worker in agricu:Lture calculated on the basis of a rural labor force participation rate of 35.5 percent comes to CFAF 88,000 (the corresponding figure for income per capita of rural population bein; CFAF 31,000). 1/ Felix Paukert: "Income distribution at different levels of development: A survey of evidence", in International Labor Review, August-September 1973, p. 114; Irma Edelman and Cynthia Taft Morris: An anatomy of patterns of income distribution in developing nations, Part III of Final report (Grant AID/csd-2236, Northwestern University), February 1971. 2/ The higher the Gini coefficient, the more unequal the income distribu- tion. The Gini coefficient (or Gini concentration ratio) is the ratio of the area between a Lorenz curve and its 450 line (the diagonal) to the entire area below the 450 line. The Lorenz curve plots cumulated percentages of income recipients (x-axis) against cumulated percentages of total income (y-axis). 3/ SETEF for Republique de Cote d'Ivoire, Ministare du Plan: L'image base 1970: Emploi, education, formation, vol. 1: L'emploi (Paris, April 1973), tableau 065, p.115.  -4 - Table 1: SELECTED RURAL AND URBAN INCOMES, 1974-1976 (CFAF '000/year) Item and Sector 1974 1975 1976 Average wages and salaries per worker: Private Sector: Non-Africans 3,392 Senegalese and other Africans: All levels (average) 446 Unskilled workers and apprentices 237 355 /1 Government 487 /2 645 /2 680 /3 Estimated average iacome per worker: Urban informal sector (Dakar): Masters 300 /4 Casual laborers and journeymen 160 /5 Apprentices 54 /5 Agriculture and related activities /6 60 88 /1 Fifty percent above 1974 figure, which corresponds to the rate of increase in the statutory minimum wage (SMIG) in November 1974. /2 Fiscal years ending June 30 (i.e., 1974 = 1973/74). Projected back from October 1976 figure by an index derived from wages and salaries per worker including social charges as recorded in the Government budget and accounts. /3 October 1976. Gains bruts et indemnites. Excludes social charges. /4 Derived from a sample of mostly unmarried room-renters with incomes between CFAF 15,000 and CFAF 55,000 per month (CFAF 180,000 and CFAF 660,000 per year) carried out in 1975 and 1976. /5 Derived from the pre-test for an ILO survey of the informal sector in Dakar carried out in early 1976. Laborers and journeymen assumed to work 200 days per year. /6 The corresponding figures for average income per head of population (based on two active members of a family of six) are: CFAF per capita per year 1974 1975 Agriculture and related activities 22,000 31,000 Sources: Private Sector: SONED: Les activites du secteur economigue moderne du Senegal en 1974 d'apres les resultats du recensement general des entreprises, Tome I; Methodologie. Analyse macrosectorielle (Dakar, June 1976) p. 86. Government: Oct. 1976: Minist're des Finances et des Affaires Economiques, Service de la Solde, computer print-out SMI 505.1, 3 Nov. 1976. Fiscal year figures for 1973/74-1975/76 from Willem Bier: Public Sector Development and Policies in Senegal: 1965-1975 (Washington, D.C.: IMF, July 1976, draft) Appendix A, Tables 2 and 21; and IMF: Senegal: Recent Economic Developments (Washington, D.C., November 9, 1976), Appendix II, Table VII, p. 38. Urban Informal Sector: See footnotes /4 and /5 Agriculture: Mission estimates based on Direction de la Statistique and SONED: Essai d'evaluation de la production de I'agriculture (campagnes 1973/74, 1974/75) (Dakar, 1976); figures provided by Direction de l'Elevage, Direction du Plan, and others; and J. Carol: Rapport de mission sur la planification regionale (Dakar, Direction de la Planification, 1976). For the corresponding figures per head of population, see footnote /6.  - 5 - The ratios of earnings of relevant groups in the urban sectors to this average income per rural worker were as follows: Ratio Unskilled workers and apprentices in the modern sector 4.0:1 Self employed in the informal sector 3.4:1 Laborers and journeymen in the informal sector 1.8:1 Apprentices in the informal sector 0.6:1 Since differences in living costs between the city and the countryside are probably of the order of 5:1, 1/ the income differentials indicated above would not by themselves be large enough to induce much migration. The migrants, however, are also likely to come from rural groups with incomes below the average we have calculated. 1.11 Within the modern sector, the average salary per employee in the Government in 1974 was 9 percent above the average of wages and salaries received by Senegalese and other Africans in the private sector. 2/ Disparities within Sectors 1.12 An indication of regional differences in income levels in 1975 is given by the rough estimates of gross domestic product per capita of popula- tion and per member of the labor force by province (region) presented in Table 2. Gross domestic product per capita in Cap Vert is 5 times that in the poorest province, Senegal-Oriental; gross domestic product per member of the labor force in Cap Vert Is 7 times that in Senegal-Oriental. Thies is just about at the national average. 1.[3 With respect to rural areas, not much is known about patterns of land holding and income distribution within the village. Since there is no land shortage in Senegal,, except perhaps in some of the older and more heavily populated parts of the groundnut basin, it is generally assumed that access to land is fairly equitable. There are, however, a variety of special local situations which result :in major inequalities of rural income. For example, the Mouride brotherhoods and colonization movements concentrate effective land rights and income in the hands of the marabout; the marabout, however, eventually attributes specific land holdings to the young men of the group as they marry. 3/ In the villages of the Senegal River Valley, access to land in 1/ This estimate is, compared to other African countries, relatively high and will be reassessed during the next mission. 2/ There are strong indications that in 1979 the public sector was paying lower salaries than the private sector due to the freeze on public salaries since 1974. 3/ See Appendix A.  - 6 - Table 2: GROSS DOMESTIC PRODUCT (GDP) PER CAPITA AND PER WORKER, BY PROVINCE, 1975 GDP GDP GDP Popula- Labor per capita per worker Province (billion tion force ('000 CFAF) ('000 CFAF) (region) CFAF) ('000) ('000) (1) + (2) (1) + (3) (1) (2) (3) (4) (5) Cap Vert 229.7 951 266 242 864 Casamance 37.3 727 312 51 120 Diourbel-Louga 33.8 825 209 41 161 Fleuve 25.6 520 116 49 221 Senegal Oriental 12.8 278 106 46 121 Sine Saloum 52.2 997 412 52 127 Thias 57.3 684 205 84 280 Senegal 448.7 4,982 1,626 90 276 Sources: Col (1): Aggregate GDP: World Bank mission estimate. Regional breakdown based on J. Carol: Rapport de mission sur la planification (Dakar, Direction de la Planification, 1976), p. 38. Col (2): Projected back from April 1976 provisional census results at provincial average growth rates shown in Table 19 below. Col (3): Derived from col (2) using provincial labor force partici- pation rates implicit in Division des Ressources Humaines: Prolections dfmographigues sur la base de l'enguete 1970-71 (Dakar, August 1975), tableau IV and Projections de la Population active (1971-2000) (Dakar, October 1975), tableau IV.  - 7 - the cuvettes is a matter of caste, with the higher caste peasants holding the better lands and the lower caste peasants (especially the former slave castes) relegated to outlaying and less productive plots whose low yields guarantee the availability of low caste individuals as hired laborers for the more prosperous higher caste peasants. It is also not clear what is the income status of the nomadic and semi-nomadic Peul herdsmen relative to that of the settled agriculturalists. If we nevertheless accept the assumption that landholdings are fairly equally distributed except among the Mourides, the average agricultural income per capita probably gives a good indication of the distribution of the total rural population by level of income. Provincial average value added originating in agriculture and related activities in 1975 ranged from CFAF 25,200 for the year in Fleuve province to CFAF 48,100 in Cap Vert; Thias province shows a figure of CFAF 31,900, just above the average of CFAF 31,400 for Senegal (see Table 3). Table 3: GROSS DOMESTIC PRODUCT ORIGINATING IN AGRICULTURE AND RELATED ACTIVITIES PER CAPITA OF RURAL POPULATION AND PER RURAL WORKER, BY PROVINCE, 1975 GDP GDP in Distribution originating agriculture (percent) :Ln Rural per capita GDP agriculture Popula- of rural originating Province (bil:Lion tion population in Rural (region) CFAF) ('000) ('000 CFAF) agriculture Population (1;' (2) (3) (4) (5) Cap Vert 13.8 183 48.1 8.0 5.2 Casamance 20.2 600 33.7 18.4 17.2 DiLourbel-Louga 17.9 697 25.7 16.3 20.0 Fleuve 10.2 405 25.2 9.3 11.6 Senegal-Oriental ,'.0 244 28.7 6.4 7.0 Si.ne-Saloum 29O.0 846 34.3 26.5 24.2 Thias 16.5 518 31.9 15.1 14.8 Senegal 109.6 3,493 31.4 100.0 100.0 Scurces: Col (1): J. Carol: Rapport de mission sur la planification rfRionale (Dakar, Direction de la Planification,1976), p. 38, adjusted by D. Steeds (IBRD mission). Col (2): Mission estimates projected back from April 1976 provisional census results. Col (3): Col (1) divided by col (2).  - 8 - 1.14 The modern sector of the economy excluding Government administration is located primarily in the urban areas. For this sector, good estimates are available on the distribution of workers and of earnings by skill level from a survey covering 1974 carried out by the Societ& Nationale des Etudes de Developpement (SONED) for the Direction de la Statistique. A summary of these data .s presented in Table 4, separately for Senegalese and other Africans and for non-Africans as well as for both groups combined. ("Macro- sectoral" summaries are also presented for manufacturing, construction building and public works, commerce and services in Statistical Annex Tables 1 to 4). For all modern sector workers combined, the skewness of the dis- tribution of wages and salaries is marked: the 3 percent of the work force in the managerial group received 20 percent of the total wage bill; the 24 percent in the unskilled group received 10 percent of the total wage bill. A Lorenz curve drawn for this distribution from the macro-sectoral data underlying Table 4 (see Figure 1) yields a Gini coefficient of 0.43; a Lorenz curve derived from similar data for the Ivory Coast for 1971 yields a Gini coefficient of 0.46. 1/ This particular distributional pattern combines three elements: (a) the usual spread in compensation between one skill level and the next; (b) the differences in the payments made to "expatriate" non-Africans and to Africans; and (c) the concentration of expatriates in the higher skill groups. At any given skill level, non-Africans obtain from twice to over two and a half times the rate of pay of Senegalese and other Africans as calculated in Table 5. Overall, however, the average payment to non-Africans is 7.6 times the average payment to Senegalese and other Africans, since of the non-Africans, 65 percent are in the top two skill groups (managers and tech- nicians); and the non-Africans constitute 54 percent of the personnel in these groups. Gradual replacement of expatriates by Senegalese in these and in the middle-skilled jobs will raise the average level of pay received by Senegalese as a group, and will also reduce the aggregate wage bill of the enterprises concerned. The disparity between non-African and African employees varies from one macrosector to another (Table 6): of the four macrosectors, it is highest in construction (a ratio of 11.9:1) and lowest in commerce (a ratio of 6.0:1). 1.15 For Government employees, a detailed breakdown of the number of employees and the wage bill ("masse salariale") is available month by month from the computer services of the MinistAre des Finances et des Affaires Economiques. The distribution of employees and of salaries by narrow salary brackets for the month of October 1976 is shown in Table 7. The corresponding Lorenz curve (which yields a Gini coefficient of 0.306) is shown in Figure 2. The average salary received in that month was CFAF 56,700 (CFAF 680,400 at an annual rate). According to the table, 68 percent of Government employees received approximately this amount or less; the aggregate received by them constituted 47 percent of the wage bill. The top 10 percent of Government employees received 25 percent of the total Government wage bill. 1/ See World Bank: Employment and salaries, income distribution, draft annex (by W. Stolper) to Ivory Coast basic economic report (1976, typescript), p. 17.  Table L: WORKERS AND ANNUAL WAGES AND SALARIES IN THE MODERN SECTOR BY NATIONALITY, 1974 Annual Wages and Salaries Category and Workers (Dec. 1974) (1974) Ski.ll Level /1 No. % Aggregate Per Worker CFAF (CFAF '000) million % 1. Senegalese and other Africans: Employers, managers 639 1.08 1,492 5.65 2.334 Technicians 1,049 1.77 1,581 5.99 1,507 Foremen 3,219 5.44 3,202 12.13 995 Skilled workers 18,286 30.89 8,951 33.91 490 Semi-skilled workers 21,341 36.05 7,695 29.16 360 Unskilled workers, 14,660 24.77 3,473 13.16 237 apprentices Total 59,194 100.00 26,394 100.00 446 2. Non-Africans: Employers, managers 1,240 41.20 5,727 56.08 4.618 Technicians 718 23.85 2,381 23.32 3,316 Foremen 626 20.80 1,634 16.00 2,610 Skilled workers 305 10.13 363 3.55 1,190 Semi-skilled workers 109 3.62 101 0.99 927 Unskilled workers, 12 0.40 6 0.06 500 apprentices Total 3,010 100.00 10,212 100.00 3.392 3. All workers: Employers, managers 1,879 3.02 7,219 19.72 3.842 Technicians 1,767 2.84 3,962 10.82 2,242 Foremen 3,845 6.18 4,836 13.21 1,258 Skilled workers 18,591 29.89 9,314 25.44 501 Semi-skilled workers 21,450 34.48 7,796 21.30 363 Unskilled workers, 14,672 23.59 3,479 9.51 237 apprentices Total 62,204 100.00 36,606 100.00 588 /1 Skill levels: English translation of French categories. English French Employers, managers Employeurs, cadres superieurs Technicians Techniciens superieurs Foremen Techniciens, agents de maitrise Skilled workers Ouvriers et employes qualifies Semi-skilled workers Ouvriers et employes specialises Unskilled workers, Manoeuvres, apprentis apprentices Source: SONED: Les activites du secteur economigue moderne au Senegal en 1974 d'apres les resultats du recensE!ment general des entreprises, Tome I: Methodologie. Anaylse macrosectorielle (Ds&kar, June 1976), p. 86. 12/76  FIGURE 1. LORENZ CURVE OF THE DISTRIBUTION OF WORKERS IN THE MODERN ECONOMIC SECTOR OF SENEGAL BY LEVEL OF WAGES AND SALARIES, 1974 100% Percentage of Wages and Salaries 50% 50% 1 00% Percentage of Workers Gini coefficient: = 0.42888 Source: Statistical Annex Tables 1 to 4. World Bank - 21 281  - 10 - Table 5: AVERAGE ANNUAL WAGES AND SALARIES IN THE MODERN SECTOR BY SKILL LEVEL: SENEGALESE AND OTHER AFRICANS COMPARED WITH NON-AFRICANS, 1974 Average annual wages Ratio of non- and salaries (CFAP '000) Africans to Senegalese Senegalese and and other Non- Differences other Africans Skill Level Africans Africans (2) - (1) (2) + (1) (1) (2) (3) (4) Employers, managers 2,334 4,618 2,284 1.98 Technicians 1,507 3,316 1,809 2.20 Foremen 995 2,610 1,615 2.62 Skilled workers 490 1,190 700 2.43 Semi-skilled workers 360 927 567 2.57 Unskilled workers, apprentices 237 500 263 2.11 Total 446 3,392 2,946 7.60 Source: SONED: Op. cit., p. 86.  - 11 - Table 6: AVERAGE ANNUAL EARNINGS IN THE MODERN SECTOR BY "MACROSECTOR" AND BY NATIONALITY, 1974 Ratio of non-Africans to Senegalese SeneBgalese All and other and other Non- Workers Africans Macrosector Afr:Lcans Africans (averages) (2) t (1) (1) (2) (3) (4) Manufacturing 408 3,557 552 8.7 Construction 3:10 3,685 413 11.9 Commerce 545 3,248 762 6.0 Services 500 3,182 606 6.4 Total 446 3,392 588 7.6 Source: Statistical Annex Tables 1 to 4, quoting SONED: Les activites du secteur ec.onomigue moderne au Senegal en 1974 d'apras les resultats du recensement 0en6ral des entreprises, Tome I: M&thodoloRie4 Analyse macrosectorielle (Dakar, June 1.976), pp. 86, 88-91.  - 12 - Table 7: DISTRIBUTION OF EMPLOYEES AND WAGE BILL /1 IN GOVERNMENT BY SALARY BRACKET, OCTOBER 1976 Cumulative Average Employ- Aggregate Distribution Distribution Salary Bracket Salary ees Wage Bill (percent) (percent) (Cl?AF/month) (CFAF/ (No.) (million Employ- Aggregate Employ- Aggregate month' CFAF/month) ees Wage Bill ees Wage Bill I - 10,000 674 325 1.1 0.72 0.04 0.72 0.04 10,001 - 15,000 14,211 834 11.1 1.86 0.47 2.58 0.51 15,001 - 17,500 16,525 383 6.3 0.85 0.25 3.43 0.76 17,501 - 20,000 18,873 253 4.8 0.56 0.19 3.99 0.95 20,001 - 22,500 21,590 895 19.3 1.99 0.76 5.98 1.71 22,fO1 - 25,000 23,982 1,483 35.6 3.30 1.40 9.28 3.11 25,001 - 27,500 26,333 1,303 34.3 2.90 1.34 12.18 4.45 27,501 - 30,000 29,220 938 27.4 2.09 1.07 14.27 5.52 30,001 - 32,500 31,366 2,521 79.1 5.61 3.10 19.88 8.62 32,501 - 35,000 34,130 2,177 74.3 4.84 2.92 24.72 11.54 35,C01 - 37,500 36,464 2,438 88.9 5.42 3.49 30.14 15.03 37,501 - 40,000 38,697 2,256 87.3 5.02 3.42 35.16 18.45 40,001 - 42,500 41,246 2,439 100.6 5.43 3.95 40.59 22.40 42,501 - 45,000 43,843 2.217 97.2 4.93 3.81 45.52 26.21 45,001 - 47,500 21,591 1,723 79.8 3.83 3.13 49.35 29.34 47,501 - 50,000 48,856 2,753 134.5 6.12 5.28 55.47 34.62 50,001 - 52,500 51,340 1,977 101.5 4.40 3.98 59.87 38.60 52,501 - 55,000 53,458 1,706 91.2 3.80 3.58 63.67 42.18 55,001 - 57,500 55,939 2,113 118.2 4.70 4.64 68.37 46.82 57,501 - 60,000 58,564 1,156 67.7 2.57 2.66 70.94 48.48 60,001 - 70,000 64,904 3,764 244.3 8.37 9.58 79.31 59.06 70,001 - 80,000 75,298 2,348 176.8 5.22 6.94 84.53 66.00 80,001 - 90,000 85,811 1,924 165.1 4.28 6.48 88.81 72.48 90,001 - 100,000 95,356 926 88.3 2.06 3.46 90.87 75.94 100,001 - 110,000 104,462 1,031 107.7 2.29 4.22 93.16 80.16 110,001 - 120,000 115,130 694 79.9 1.54 3.13 94.70 83.29 120,0.31 - 130,000 125,205 488 61.1 1.09 2.40 95.79 85.69 130,001 - 140,000 134,783 437 58.9 0.97 2.31 96.76 88.00 140,001 - 150,000 145,192 312 45.3 0.69 1.78 97-45 89.78 150,0(31 - 175,000 161,772 395 63.9 0.88 2.51 98.33 92.29 175.001 - 200,000 183,824 272 50.9 0.61 2.00 98.94 94.29 200,001 - 250,000 219,048 231 50.6 0.51 1.99 99.45 96.28 250,001 - 300,000 274,700 83 22.8 0.19 0.89 99.64 97.17 300.001 + 448,450 161 72.2 0.36 2.83 100.00 100.00 Total 56,695 44,956 2,548.8 100.00 100.00 /L Gains bruts et indemnites. Excludes charges sociales, which add about 40 percent to the basic wage bill. Source: Ministere des Finances et des Affaires Economiques, Service de la Solde, computer print-out SMI 505.1, November 3, 1976.  FIGURE 2. LORENZ CURVE OF THE DISTRIBUTION OF GOVERNMENT EMPLOYEES BY SALARY BRACKETS, OCTOBER 1976 Percentage of 100% Wages and Salaries 50% 50% 100% Percentage of Employees Gini coefficient = 0.30602 Source Table 7. World Bank - 21282  - 13 - B. Poverty Levels and Living Standards 1.16 As indicated above (Table 2), gross domestic product per head of population came to CFAF 90,000 for all of Senegal, ranging by province from CFAF 41,000 in Diourbel to CFAF 242,000 in Cap Vert. Income originating in agriculture per head of rural population for all of Senegal came to CFAF 31,400, ranging by province from CFAF 24,200 for Fleuve province to CFAF 48,100 for Cap Vert. In 1976, the World Bank has estimated an absolute poverty level for rural Senegal at CFAF 14,900 (US$69) and a relative poverty level of CFAF 25,170 (US$117). 1/ The average figure for income per head of rural population is above the absolute rural poverty level in all the provinces. We have no estimate of the range of per capita rural incomes within each province. However, the averages for two provinces, Fleuve and Diourbel-Louga, which together have 31.6 percent of the country's rural population, are at or below the relative poverty level. 1.17 Some indications of living standards in greater Dakar are given by the results of a household survey carried out in February-March 1975 by the Institut Universitaire de Technologie (IUT) of the University of Dakar. 2/ For the sample of 544 households covered in the IUT survey, the average number of persons per household came to 8.6, the average number of children per household being 3.2 and the average number of adults, 5.4. Counting a child as the equivalent of one half an adult with respect to consumption, the number of "consumption units" per household came to 6.0. Average monthly income per household was recorded at CFAF 87,143, of which CFAF 69,200 consisted of earned incomes and CFAF 7,059 of transfers. Average monthly expenditure per household came to CFAF 89,215 of which CFAF 80,445 consisted of consumption expenditure and CFAF 8,769 of transfers. The distribution of households by monthly income bracket is shown in Table 8 and its Lorenz curve (with a Gini coefficient of 0.391) is shown in Figure 3. The number of consumption units increased steadily with size of income, from 4.6 for the lowest income bracket to 7.8 for the highest income bracket; in other words, the higher the household income level, the greater the number of dependents (as well as earners) per household. A corresponding distribution of households by monthly expenditure bracket (with a Gini coefficient of 0.310) is shown in Table 9 and in Figure 4. In this case, the number of consumption units per household peaks at 6.8 (the equivalent of 10.1 persons) in the CFAF 125,000-149,999 month:Ly expendi- ture bracket. 1/ In the World Bank estimate, the absolute poverty level is taken as the cost of a diet of 2,300 calories and 100 grams of protein per person per day, plus an additional 42.8 percent for non-food expenditure. The relative poverty level is taken as one-third of national aggregate personal income per capita. 2/ Institut Universitaire de Technologie, Division Tertiaire, Universite de Dakar: Etude Budget - Consommation, II. Enquete: Budgets familiaux (Dakar, June 1976), p. II-6.  - 14 - Table 8: DISTRIBUTION OF HOUSEHOLDS BY MONTHLY HOUSEHOLD INCOME BRACKET, GREATER DAKAR, FEBRUARY - MARCH 1975 Average Cumulative Ll Consump- Income per Percent percent tion Household Income Household Aggregate Aggregate Units Bracket (CFAF per Income per House- Income per House- House- (CFAF per month) month) Bracket holds bracket holds holds 0 - 24,999 17,161 1.9 9.6 1.9 9.6 4.6 25,000 - 49,999 37,292 10.1 23.7 12.0 33.3 5.4 50,000 - 74,999 60,840 16.3 23.3 28.3 56.6 6.0 75,000 - 99,999 87,215 13.6 13.6 41.9 70.2 6.0 100,000 - 124,999 111,737 14.6 11.4 56.5 81.6 6.3 125,000 - 149,999 137,914 9.6 6.1 66.1 87.7 6.6 150,000 - 249,999 189,128 18.8 8.6 84.9 96.3 7.2 250,000 and over 357,728 15.1 3.7 100.0 100.0 7.8 All levels 87,143 100.0 100.0 6.0 /1 Gini coefficient estimated at 0.391. Source: Institut Universitaire de Technologie, Division Tertiaire, Universit& de Dakar: Etude budget-consommation, II. Enguete: Budgets familiaux (Dakar, June 1976), p. III-12.  FIGURE '. LORENZ CURVE OF THE DISTRIBUTION OF HOUSEHOLDS BY MONTHLY INCOME BRACKET, DAKAR 1975 Percentage of 100% Income 50% 50% 1 00%/, Percentage of Households Gini coefficient = 0.39063 Source: Table 8. World Bank - 21283  - 15 - Table 9: DISTRIBUTION OF HOUSEHOLDS BY MONTHLY HOUSEHOLD EXPENDITURE BRACKET, GREATER DAKAR, FEBRUARY - MARCH 1975 Percent Distribution Lj Household Aggregate Expenditure Expenditure Consumption Persons Bracket per House- Units per per (CFAF per month) Bracket holds Household Household 0 - 24,999 0.4 2.2 3.5 5.1 25,000 - 49,999 7.2 17.8 4.6 6.8 50,000 - 74,999 14.0 22.3 5.7 8.4 75,000 - 99,999 16.5 18.6 6.4 9.1 100,000 - 124,999 15.5 13.6 6.7 9.5 125,000 - 149,999 14.4 10.5 6.8 10.1 150,000 - 249,999 22.7 12.1 6.6 9.2 250,000 and over 9.3 2.9 6.6 9.3 Total 100.0 100.0 6.0 8.6 L1 Gini coefficient estimated at 0.310. Source: Institut Universitaire de Technologie, op. cit., pp. II-10 and III-5.  FIGURE 4. LORENZ CURVE OF THE DISTRIBUTION OF HOUSEHOLDS BY MONTHLY HOUSEHOLD EXPENDITURE BRACKET, DAKAR, 1975 Percentage of Expenditure 100% 50% 100% Percentage of Households Gini coefficient = 0.30979 Source: Table 9. World Bank - 21284  - 16 - 1.18 The distributions of households and consumption units and of aggre- gate expenditure by level of monthly expenditure per consumption unit accord- ing to the IUT survey are shown in Table 10. The table illustrates the inverse relationship between household size and expenditure per consumption unit in the households. The table may also be used to estimate the proportion of households in greater Dakar below a poverty line defined (as for the rural population) in terms of consumption expenditures per person. We calculate the level of absolute urban poverty to be at least 5 times the rural level, in view of the differences between rural and urban prices. The resulting figure of CFAF 6,210 per month implies that only some 5 percent of greater Dakar households (or 7 percent of greater Dakar's consumption units) are, according to Table 10, below such an absolute poverty level. C. Incidence of the Government Budget 1.19 Of the various distributive effects of Government expenditure and receipts, the operations of the Caisse de Perequation et de Stabilisation des Prix (CPSP) have a particular significance for the distribution of income between rural producers of agricultural commodities for export and the mostly- urban consumers of imported and locally grown agricultural produce and pro- cessed foodstuffs. CPSP profits on operations involving export commodities (particularly cotton and groundnuts) imply a tax on rural producers, and conversely losses imply a subsidy. CPSP profits on imported locally produced foodstuffs imply a tax on consumers, and losses imply a subsidy. In periods of high prices for agricultural exports, the CPSP has made profits and imposed a considerable implicit tax burden on Senegalese agriculture: in 1973/74 the profits on groundnuts and cotton reached CFAF 15.5 billion (see Table 11). In that same year, the CPSP's subsidies to consumers on rice, sugar and groundnut oil amounted to CFAF 11.1 billion, resulting in a clearly identifiable net transfer of resources from rural producers to urban consu- mers. In 1974/75, the CPSP's profits on groundnuts and subsidies on groundnut oil fell drastically and overall there was a small net transfer to producers as well as a moderate net transfer to consumers. In 1975/76, however, with declining world prices for groundnuts, the CPSP carried out its price stabilization function by taking a severe loss on its operations and paying a substantial subsidy to rural producers who in effect began to recoup some of the transfers they had made in the good years. Faced with the prospect of impossible losses for the CPSP, the Government took the hard decision in late 1974 to eliminate subsidies on rice and sugar by raising the selling price to consumers, at the same time moderating the blow to modern sector employees by raising the minimum wage (SMIG) to compensate for most of the loss in purchasing power. The net losers at that point were the urban residents working (or unemployed) outside the modern sector. The Government's distributional policy in the mid-seventies was clear: it did its utmost to protect the benefits already acquired by the work force in the modern sector which forms a significant political clientele in the Senegalese situation. However, there are at the moment clear signs that the Government is changing its policies in this area. A comprehensive income policy is being worked out in the framework of a medium-term economic adjustment program.  - 17 - Table 10: DISTRIBUTION OF HOUSEHOLDS BY THE LEVEL OF I4ONTHLY HOUSEHOLD EXPENDITURE PER CONSUMPTION IJNIT /J1, GREATER DAKAR, FEBRUARY - MARCH 1975 Expenditure per Percent Distribution /2 Consumption Consumption Unit Aggregate House- Consumption Units per (CFAF per Month) Expenditure holds Units Household 0 - 4,999 1.0 2.6 3.9 8.9 .5,000 - 9,999 12.9 22.1 28.4 7.7 110,000 - 14,999 19.9 24.3 26.4 6.5 1.5,000 - 19,999 17.4 17.6 17.0 5.8 20,000 - 29,999 21.7 16.7 14.8 5.3 310,000 - 44,999 11.8 8.8 5.4 3.7 4.5,000 - 59,999 7.2 4.4 2.4 3.2 60,000 and over 8.1 3.5 1.7 2.9 Total 100.0 100.0 100.0 6.0 L1 Adults and adult equivalents (a child equals one half of an adult). L2 Gini coefficients: for households, 0.210; for consumption units, 0.343. Source: Institut Universitaire de Technologie, op. cit., p. III-6.  - 18 - Table 11: NET TRANSFERS TO PRODUCERS AND CONSUMERS OUT OF THE CPSP, FISCAL YEARS 1970/71 - 1976/77 (in billions of CFAF) Recipient and Commodity 1970/71 1971/72 1972/73 1973/74 1974/75 1975/76 1976/77 Transfers to producers (subsidies = -): Groundnuts 4.2 3.8 4.8 14.3 2.9 3.4 4.9 Cotton 0.3 0.1 0.3 1.2 0.1 n.a. n.a. Program -0.8 -0.5 -2.4 -1.2 -4.2 -5.9 -9.6 3.7 3.4 2.7 14.3 -1.2 -2.5 4.9 Transfers to consumers (mostly urban) (subsidies = Rice -- - - -5.0 -1.7 5.1 3.8 Sugar -0.1 -0.2 -0.5 -1.7 -1.1 2.4 1.0 Groundnut oil -1.3 -1.6 -2.7 -4.4 -0.6 -0.4 -1.6 Flour -- -- -- -- -1.3 -2.4 -- -1.4 -1.8 -3.2 -11.1 -4.7 4.7 3.2 Surplus 2.3 1.6 -0.5 3.0 -5.9 2.2 8.1 Note: Table to be up-dated with information requested from the CPSP. n.a. not available  - 19 - D. Nutrition 1.20 For making any intra-Africa international comparisons, particularly between Senegal and the Ivory Coast, it should be remembered that there are two completely different types of diet within West Africa. In the dry (Sahel) regions, the diet is based on cereals; in the wet (forest) regions, the diet is based on starchy foods (tubers, roots and plantains). This is illustrated by two nutrition surveys cited by Perisse 1/: an average cereal consump- tion of 202.3 kg per capita per year with an average tuber and root consump- tion of only 1.1 kg in the Senegal River Valley 2/; and an average cereal consumption of only 9.8 kg per capita per year with an average tuber and root consumption of 666.1 kg in Bongouanou (Ivory Coast) 3/. In the Senegal River Valley case, 77 percent of total calorie intake was supplied by cereals; in the Ivory Coast case, 83 percent of total calorie intake was supplied by tubers and roots. The cereal-based diet has, of course, a higher protein content than the tuber-based diet; but this is an advantage only if the diet contains sufficient calories to permit absorption of the protein as protein, and not as an energy source. 1.21 A measure of the extent of urban poverty is given by the estimates of nutrition levels derived by the FAO from the results of the IUT's 1975 household budget survey of Dakar. 4/ The FAO has converted the per capita food consumption estimates of the IUT household budget survey into quantities per capita on the basis of market prices collected by the Direction du Com- merce Int6rieur et des Prix and from quantities into nutritional values using standard tables of equivalents by product. The results are summarized in Table 12 which shows the average quantities of calories, proteins, animal proteins and lipids consumed per consumption unit by level of total expendi- ture per consumption unit. Calculating the average daily caloric requirement of the urban population at 2,000 calories (5 percent below the estimated requirement of 2,100 for Senegal as a whole), it is evident from the table that the average consumption of 2,341 calories of Dakar is above the require- ment. However, for the two lowest expenditure categories, which include 31.4 1/ Julien PFriss&: L'alimentation en Afrigue intertropicale: Etude critique a partir des donn6e des enquetes de consommation 1950-1965. Thesis presented to the Faculte de Pharmacie de l'Universit6 de Paris, 1966, p. 19. 2/ Quoted from J.L. Boutillier et al: La moyenne vall6e du Senegal: Etude socio-economiique (Paris, Presses Universitaires de France), 1962. --/ Quoted from Service de Statistique de Cote d'Ivoire: Enquete nutrition - niveau de vie. Sub-division de Bongouanou 1955/56. 4/ FAO: La demande alimentaire [au Senegal) (Rome, November 1976, mimeo draft).  Table 12: NUTRITIONAL VALUE OF THE DIET PER CONSUMPTION UNIT PER DAY OF URBAN FAMILIES IN DAKAR BY LEVEL OF TOTAL EXPENDITURE PER CONSUMPTION UNIT, 1975 Item Level of Total All Expenditure per 0 5,000 10,000 15,000 20,000 30,000 45,000 60,000 Consumption Consumption unit -4,999 -9,999 -14,999 -19.999 -29,999 -44,999 -59,999 + Units Population (%) 3.8 27.6 26.5 17.2 14.8 5.9 2.5 1.7 100.0 Nutritional values per day: Calories 1,294 1,630 2,183 2,827 3,172 3,612 4,300 3,457 2,341 Proteins (gr) 27.7 46.0 60.1 69.2 84.7 103.0 131.6 114.5 Animal Proteins (gr) 5.2 17.6 24.7 29.7 38.3 44.5 63.5 63.5 Lipids (gr) 47.4 55.0 80.4 107.9 134.7 140.3 151.0 144.1 Source: FAO: La demande alimentaire [au Sen&gal] (Rome, November 1976, mimeo draft) p. 2, based on Institut Universitaire de Technologie: Etude budget consommation, Tome II: Budgets familiaux (Dakar, June 1966).  - 21 - percent of the consumption units in the IUT sample, the average consump- tion is less than the requirement. The FAO's guess is that perhaps half oE these (or 15 percent 1/ of the population of greater Dakar) suffer nutri- tLonal deficiencies. According to the staff of the Institut de Technologie A:Limentaire, caloric deficiencies are significant in some of Dakar's peripheral urcban slums. Small chiLdren in these areas suffer in addition from other nutritional lacks: protein deficiencies due to taboos on giving small children f:Lsh and eggs; and vitamain deficiencies due to lack of access to the fruits and leafy vegetables that would be available to the rural poor. The higher income groups, on the other hand, suffer from some of the ailments of over- consumption (particular:Ly of oils); diabetes and high blood pressure are major health problems in Dakar. 1.22 The commodity composition of the diet of Dakar has indirect distri- butional implications as well. Within total food expenditure in Dakar, according to the 1975 IIJT survey, 27 percent went for cereals and cereal pIoducts and 31 percent for animal products. 2/ 3/ Of the expenditure on ce!reals only 7.0 percent: (1.9 percent of total food expenditure) went for millet and sorghum; rice, bread and wheat flour accounted for 91 percent (24.7 pe!rcent of total food expenditure). Even among the poorest groups, millet and sorghum accounted for only 7.4 percent of expenditure on cereals (2.6 percent of total food expenditure). As incomes rise, relative demand will decline for (locally produced) millet, while demand will increase for (imported) rice, wheat, maize, sugar and milk. 1.23 Part of the difficulty of marketing millet in urban areas is the tediousness of its preparation and the relatively high cost of industrially prepared couscous. The price per calorie of the main cereal products, as estimated by the FAO, show millet and maize to be the cheapest nutritional energy sources; but the price per calorie of prepared couscous is the same as for whole rice and one and a half times that for broken rice (see Table 13). In order for couscous based on locally grown millet to cut in on the price advantage of rice and the established position of rice in the Dakar diet, a major improvement in the industrial preparation of couscous from millet and a marked reduction in the market price of the prepared couscous will be required. 1/ FAQ, ibid., p. 10, where, however, this figure is presented as an assumption, not as a description of reality. 2/ FAQ, ibid., p. 5, tableau 2. 3/ Institut Universitaire de Technologie, op. cit., tableau 4 adjusted to same basis as FAQ table.  - 22 - Table 13: PRICE OF SELECTED FOOD PRODUCTS PER 1,000 CALORIES AND PER 10 GRAMS OF PROTEIN Price Price Protein Price per per kg Calories per 1,000 per kg 10 g of Food Product (CFAF) per kg Calories (grams) Protein (CFAF) (CFAF) Broken rice 100 3,560 28 98 10.2 Whole rice 150 3,570 42 81 18.5 Millet 38.5 3,780 10 97 4.0 Couscous 100 2,350 42 60 16.7 Maize 56 3,680 15 78 7.2 Wheat flour 90 3,510 26 105 8.6 Bread 128 2,540 50 80 16.0 Potato 60 710 84 15 40.0 Manioc 65 1,100 59 9 72.2 Sweet potato 49 960 51 13 37.7 Edible peanuts 70 5,490 13 232 3.0 Oil 198 8,840 22 - - Sugar 250 3,870 65 - - Meat 375 1,980 189 139 27.0 Fresh fish 107 690 155 115 9.3 Dried, smoked fish 629 3,610 174 634 9.9 Fresh milk 157 790 199 38 41.3 Source: FAO: La demande alimentaire [au Senegal] (Rome, November 1976, mimeo draft), p. 6.  - 23 - 1.24 The nutritiona.L situation in rural areas is less well known than that of greater Dakar ex,zept for Senegal's fundamental rural problem: the seasonal shortage of food grains in the major areas of rainfed agriculture and particularly the groundnut basin. 1/ Millet stocks are normally abundant immediately after the harvest; they are run down during the long dry season and during the May-June rainy season. Consequently, the stocks are extremely low during the critical season (August-October) prior to the harvest (the soudure) when energy requirements to carry out the harvest are the highest. The results of two village nutritional surveys quoted by the FAO show this seasonal pattern clearly. Table 14: AVERAGE DAILY CALORIE CONSUMPTION PER CAPITA BY SEASON IN TWO SELECTED VILLAGES IN THIES AND SINE-SALOUM Tivaouan Fatick (Thias) (Sine-Saloum) Item and Season 1966 1968 Calories consumed daily per capita: January - March 2,044 2,193 May - June 1,880 1,942 August - October 1,528 1,848 Percentage of energy requirement (2,100 calories) covered: January - March 97 104 May - June 90 92 August - October 75 88 Source: FAO: Etat Nutritionnel au S&negal (Rome, November 1976, mimeo draft), p.7. 1.25 The normal seasonal shortage of food grains in the rural areas of Sahelian Senegal is fundamental to several aspects of the distribution of income and wealth and to the geographical movements of sectors of the popula- tion. The basic problem of the soudure is resolved or circumvented in a number of ways. During the dry season, when there is little or no work to be done in the countryside, the young men of the village migrate to smaller or 1/ The World Bank is preparing a nutrition project for Diourbel and is studying the Casamance region, in the framework of which new data are being established.  - 24 - larger towns in search of income, 1/ partly in order to reduce the depletion of the stocks of foodgrains needed to carry the village's agricultural work force through the next period of the soudure. During the lean period, part of the villagers' money income is spent on basic foodstuffs (at seasonally high prices) from outside the village to supplement dwindling supplies of self-produced foodgrains needed for subsistence. 2/ It is said that some villages keep part of the millet stock in special granaries whose contents are consumed only during the short period of extreme exertion of harvest time; reduction in food intake and belt tightening occur during the slack season when less physical effort is required. Finally, the peasant who manages to retain an adequate stock of millet through the year is the one who can make his demand for manpower effective at the critical moments of seeding and harvesting. 1.26 A factor in Senegal's nutrition situation which is generally con- sidered favorable is the wide spread consumption of fresh fish and of dried and smoked fish; this enhances significantly the protein content of the staple diet except in some of the more isolated inland villages to which the commer- cial network does not extend. There is, however, an as yet unresolved ques- tion of the biological value of the protein content of dried and smoked fish prepared by artisanal methods. 3/ 1.27 The regional income disparities noted above are compounded by disparities in the availability of public services, which originally spread outward from the French colonial capital cities, first Saint Louis and then Dakar. These disparities are illustrated in the next sections. E. Health 1.28 The regional inequality of the distribution of medical facilities is heavily biased in favor of Cap Vert (essentially Dakar). Of the 192 doctors in Senegal in 1966, 130 (68 percent) were located in Cap Vert; by 1975, the Cap Vert share (225) of the higher total number of doctors (307) was up to 73 percent. In 1975 Cap Vert had 64 percent of the country's mid-wives and 67 percent of its medico-technical personnel. The number of inhabitants per doctor in Cap Vert in 1975 was 3,700 as compared with 42,300 in the rest of the country; while the number of inhabitants per mid-wife was 5,800 in Cap Vert, in the rest of the country it was 52,900. 1/ J. Roch: "Les migrations economiques de saison sache en bassin arachidier senegalais", in Cahiers ORSTOM, Serie Sciences humaines, vol. XII, No. 1, 1975. 2/ Perisse, op. cit., p. 99. 3/ FAO: Bilan alimentaire 1974 (Rome, Nov. 1976, mimeo draft), p. 5. (This problem is being studied by the Institut de Technologie Alimentaire in Dakar).  - 25 - Table 15: MEDICAL PERSONNEL AND INHABITANTS PER UNIT: SENEGAL, CAP VERT AND OTHER PROVINCES, 1966 AND 1975 Number of Medical- Cap Vert Year and Health Personnel as % Inhabitants Per Unit Category Cap Other of Cap Other of Personnel Senegal. Vert Provinces Senegal Senegal Vert Provinces 1966: Doctors 192 130 62 67.7 18,341 4,276 47,832 Pharmacists 14 13 1 92.9 251,533 42,753 2,965,600 Nurses 1,221 449 772 35.8 2,884 1,238 3,841 Mid-wives 151 95 56 62.9 23,321 5,850 52,957 Tech. personnel 1,244 868 376 69.8 2,830 640 7,887 1975: Doctors 307 225 82 73.3 14,044 3,732 42,339 Pharmacists 37 34 3 91.9 116,527 .24,697 1,157,267 Nurses 2,563 936 1,627 36.5 1,686 897 2,134 Mid-wives 329 212 117 64.4 13,014 3,960 29,674 Tech., personnel 1,761 1,174 587 66.7 2,448 715 5,914 Source:: World Health Organiz,tion: Profil sanitaire du Sen&gal (Dakar, September 1976), Annex V, quol:ing Service de Sante: Rapport annuel 1974. Oi. the 9 hospitals in Senegal in 1974, 4 were in Cap Vert although this preponderence was partly compensated by the 33 smaller health centers located in the other provinces. Of the 60 maternities ia the country in 1974, 17 (28 percent) were in Cap Vert as were 24 (36 percent) of the country's 66 mother anid child protection centers. 1/ 1/ Protection materne:Lle et infantile (PMI).  - 26 - Table 16: HEALTH INFRASTRUCTURE: SENEGAL, CAP VERT AND OTHER PROVINCES, 1974 Other Facility Senegal Cap Vert Provinces Hospitals 9 4 5 Health centers 34 1 33 Health posts 428 62 366 Maternities 60 17 43 Mother and Child Protection centers /1 66 24 42 /1 Protection maternelle et infantile (PMI). Source: World Health Organization: Profil sanitaire du Senegal (Dakar, September 1976) Annex III, quoting Service de Sant&: Rapport annuel 1974. The population per bed in hospitals, maternities and other facilities was 767 for Senegal as a whole ranging from 298 for Cap Vert to 2,121 for Senegal- Oriental. The numerical advantage of Cap Vert is marked not only in terms of numbers of persons per health facility but also in terms of the distances of access to such facilities.  - 27 - Table 17: HEALTH INFRASTRUCTURE: HOSPITALIZATION CAPACITY, 1974 ]'rovince and Population Type of Beds per Facility No. Percent bed Cap Vert 2,916 50.4 298 Casamance 468 8.1 1,458 ])iourbel 444 7.7 1,577 Fleuve 751 13.0 569 Senegal-Oriental 127 2.2 2,121 Sine-Saloum 601 10.4 1,494 rhias 478 8.2 1,285 Senegal (Total) 5.785 100.0 767 Hospitals 3,359 58.1 Maternities 1,536 26.5 Other 890 15.4 Source: World Health Drganization: Profil sanitaire du Sen6gal (Dakar, September 1976), Annex IV, quoting Service de Sant&: Rapport annuel 1974. 1.29 The incidence of disease and especially infant mortality shows corresponding regional disparities, some of which are due to the differences in availability of health services and some simply to climatic and ecological differences. For example, infant mortality rates in rural areas of Senegal, estimated at 20 percent in the age group under 1 year and 35 percent in the age group 1-4 years, are high by international standards and twice as high as in urban areas of Senegal. 1/ The extremely high rate in the 1-4 year age 1/ FAO; Etat nutritionnel au S6negal (Rome, November 1976, mimeo draft), p. 3.  - 28 - group is presumed to be linked with the precarious nutritional state of young children once they are weaned, which increases their vulnerability to measles and malaria. There are strong seasonal variations in infant mortality in both rural and urban areas, with the highest rates during and just after the rainy season when infections, diseases and malaria are the most rampant and food stocks in rural areas are depleted. The lower infant mortality rates in urban areas would appear to be due to the much higher level of medical and health infrastructure in urban areas (especially in Cap Vert) than in rural areas. 1.30 A CINAM-ORANA survey carried out in 1960 showed rates of incidence of malaria outside of Cap Vert ranging from 31.5 percent in Thies to 90.1 percent in Senegal-Oriental. It is estimated that 70 percent of the country's population is still affected by malaria and parasitic diseases. 1.31 One of the more important of the parisitic diseases is riverblind- ness (onchocerciasis) which is endemic in parts of Casamance and in Senegal- Oriental. The disease is, in fact, the major obstacle facing Government plans to extend settlement and the "groundnut frontier" eastward to the upper Senegal River (the new-lands project). Mapping the geographical extent of the disease and its eradication are essential to the eventual success of the new-lands program. Since the disease is also endemic in Senegal's southern neighbors, Guinea Bissau and Guinea, a mapping and eradication effort must be an international-regional effort which can, hope- fully, be organized by the World Health Organization. The Senegalese Govern- ment has submitted a request to the World Health Organization for such a regional project. An international committee has been set up to study the onchocerciasis problem in West African countries not yet covered by a program. 1.32 The Government's health program has had two major orientations: (a) the promotion of integrated preventive, curative and social medicine and health education, with polyvalent training of medical and health personnel; and (b) improvement of the coverage of the medical and health network through the development of basic health services, health education and an adequate decentralization at the provincial level. The accent has been put on two important programs: (a) the development of rural maternities and village pharmacies in the provinces: in Thies (since 1972) and in Sine-Saloum (since 1974); and (b) a nutritional and health protection program focused on the mother-and-child group most vulnerable to disease and malnutrition. As of mid-1976, there were 99 rural maternities and 25 village pharmacies planned or functioning in Thies and Sine-Saloum; the nutritional and health protection program had 35 principal centers and 105 secondary centers functioning in all of the provinces. 1/ The USAID program in Senegal is reinforcing the Sine-Saloum program with a $3.5 million rural health services development project, to start in 1977, which is to provide a model for assuring adequate health care at the village level. The project will construct some 90 clinics throughout Sine-Saloum province and will provide necessary basic equipment and staff. 1/ WHO: Profil sanitaire du Senegal (Dakar, Sept. 1976), p. 6.  - 29 - 1.33 One of the marked health problems in Senegal usually linked to diet deficiencies is the p:revalence of anemia which has been found to affect about 48 percent of the total population and 74 percent of pregnant and nursing women. 1/ In the case of Senegal the prevalence of anemia is probably due to malaria and parasitic infections. Another endemic diet-related disease is goiter which, however, could be easily eradicated by the iodization of salt or by injection of iodatead oil in isolated communities. 1.34 With respect to population policy, Government efforts have so far concentrated on reducing infant mortality rates through the mother and child protection centers, of which there were 66 in 1974: 17 principal centers in provincial and departmental capitals, 32 secondary centers in smaller towns and villages, and in Dakar 7 USAID-financed private health centers and 10 centers run by the municipality. It is not at all clear how (or whether) the mother and child protection service is coordinated with the rural maternity and the nutritional and health protection programs. 1.35 The mother and child protection program is primarily a preventive health service directed toward pregnant and lactating mothers and toward pre-adolescent children. With regard to women, the objective is to monitor pregnancy and post-natal care; with regard to children, the service carries out periodic examinations of new-born babies and a systematic program of vaccinations and check-ups to protect against a variety of childhood diseases. Information and advice are made available on family planning methods and options. 1.36 The only clinic in Senegal which now provides complete family planning services is a private institution at Croix-Bleu in Dakar financed by Pathfinders International (US). The clientele includes women from all social classes, but only few from the lower income groups can avail themselves of these services because of their relatively high cost (e.g., CFAF 1,000 for a gynecological examination and a further CFAF 1,000 for fitting an IUD). USAID is currently arranging to establish a pilot family planning clinic on the premises of the main mother and child protection center in Dakar and has included in the country program for USFY 1977/78 a $1 million project to provide comprehensive family planning information through existing clinics (rural as well as urban) and to train personnel in family planning techniques and health management. 1.37 As the Government programs for reducing infant mortality become increasingly effective, policy makers will have to face the implications of the inevitable increase in the rate of population growth resulting from the cultural lag between a fall in mortality rates and a subsequent fall in birth rates. 1/ PAO: Etat nutritionnel au S6n4gal (Rome, November 1976, mimeo draft), p. 3 quoting data collected by Dr. Krikawa in 1974.  - 30 - F. Education 1.38 The Senegalese formal education system resembles that of other Francophone West African countries. The system is modeled after the tradi- tional French system and only recently have attempts been made to reform and adapt it gradually to the social and economic needs of the country. 1.39 The proportion of the national budget allocated to education and training increased considerably until 1970, when it stabilized at approxi- mately 25 percent, representing 3.3 percent of GDP. The Government has decided that education's share of the national budget will not be allowed to increase further, and a policy of limited growth of enrollments for the next plan period has been adopted. As a consequence, there will be very little room for expansion of and improvements in the present formal system. 1.40 Government policy with regard to education has two major components. First is a firm determination to make the educational system more responsive to manpower requirements, particularly in the modern sector, hence the priority given to the development of technical education at all levels. Second is the Government's aim at improving the quality of education rather than expanding it, especially in the primary grades. The major constraint to education development is this emphasis on quality combined with the fact that recurrent expenditure for education will not be allowed to increase more than the rate of growth of the Government's recurrent budget. Universal primary education still remains a long-term policy objective, but the above constraints have forced the Government to abandon this goal, at least for the near future. 1.41 The base of the Senegalese school system is still very narrow, with a primary enrollment rate of around 40 percent, and enrollments increas- ing at approximately the rate of population growth. About 20 percent of primary school graduates are admitted into a four-year middle-school program. The Government is at present experimenting with a new form of practical middle-level education (enseignement moyen pratique - EMP) for the other 80 percent. Originally this type of education was conceived to provide a practical education for primary school drop-outs and adolescent illiterates as well, but this aspect is currently being deemphasized. Secondary schools offer a three-year diversified curriculum leading to the baccalaureate which gives access to higher education. Students who have obtained the baccalaureate are all assigned to an institution of higher education by a "Commission d'orientation". Although this Commission is set up to direct students accord- ing to national priorities, its manoeuvering ability is limited by the quali- fications of the graduates.  - 31 - Post-Seconudary Orientation of Baccalaureates in 1975 Number of Baccalaureates Planned Department Assigned Enrollment Law and Economics 452 150 Medicine and Pharmacy 103 100 Sciences 222 350 Literature and Humanities 439 240 1.42 The largest institution of higher education is the University of Dakar which was originally conceived for 3,500 students but enrolls at present about 8,000, of which nearly 70 percent are Senegalese. The vast majority of the students are enrolled in the social science and language departments and only 10 percent in science. A second university in St. Louis will be con- structed exclusively for studies in literature and humanities. The Demand for Education 1.43 In some regions of the country the enrollment ratio for primary education is falling due to lack of demand. This is the case in the provinces of Louga/Diourbel, Senegal-Oriental, Sine-Saloum and Fleuve. The reasons for this decline have not yet been sufficiently analyzed. The peopLe may have become disappointed with the "French" education that the formal school system is providing and are rejecting it. Many Muslim parents send their children to the Koranic school where they receive religious instruction and are taught the traditional values of their community. According to the 1970/71 census, only 3.4 percent of the population in rural areas had attended primary schools, compared with 34 percent who had been to the Koranic or Arabic school. Since 1969/70, a number of schools have been closed in rural areas. Instead of one school in every village, the Ministry of Education is trying to have three or four six-grade schools in every "arrondissement". It should be noted in this context that the role of modern private education at the primary level is also diminishing although the Government supports private institutions. All this seems to indicate that in the rural areas in Senegal the demand for education is lagging. Since urlban demand for primary education is still strong, the system is becoming more urban based than before. 1.44 Beyond the primary level, the situation seems to be radically different and several indicators point to a considerable excess demand for middle-level and secondary education:  - 32 - (a) the high rates of repetition in the top grade of both primary schools and middle schools, repetition which is in general voluntary with students trying for a second chance at entry into a higher level of education; (b) the rapidly increasing importance of non-religious private schools at these levels, where parents pay the full cost of education; and (c) the fact that all students who pass the final examination at one level of education continue their studies at the next; only those who fail enter the labor market. The excess demand is caused by three factors: education in Senegal is free; few job possibilities exist for this age group, keeping the opportunity cost of the time spent on further education very low; and the expected income increments generated by the additional years of schooling are considerable. Although income expectations are excessive in view of job availabilities, the low private cost of education will continue to encourage young people to take the risk of long studies and years of unemployment. 1.45 All evidence points to a large discrepancy between the private and the social returns of post-primary education. At one time, this was justified because of the general shortage of Senegalese with post-primary education. However, at the moment there is an overall surplus with shortages only prevailing in some specific fields. The high cost to society of post-primary education is reflected in the budget allocations for education and in the increasing number of educated unemployed. Bringing the private returns of post-primary education in line with the social returns could be done by increasing the cost of education for the beneficiaries and reducing income levels of the urban elite in proportion to other population groups, a case which also can be defended on equity grounds. Basic Education 1.46 The duration of the primary school course is six years. The language of instruction is French, but plans exist to change to local languages. The curriculum contains few practical subjects and is largely determined by the requirements of the middle school entrance examination. During the school year 1975/76, total enrollment was estimated at 312,000 (of which 40% were girls). The system is basically oriented towards urban life and modern sector employment. Rural areas are at a disadvantage not only because buildings are often poor, furniture scarce and educational materials lacking, but the best teachers are attracted to the cities. 1.47 The average national primary school enrollment ratio conceals considerable regional variations. The highest enrollment ratio (64%) is found in Cap Vert 1/, the most urbanized region, and the lowest (13%) in 1/ A recent survey found however that the rates of schooling in some of the poorest parts of Dakar are well below 50 percent.  - 33 - Louga-Diourbel, the center of the largest Muslim brotherhood. Senegal- Criental and Sine-Saloum also have rates of schooling well below the national average. Casamance strongly deviates from this pattern with high enrollment rates compared to the economic structure of the region, but with relatively low rates for girls. 1.48 The 1977-1981 Development Plan proposes to limit the expansion of primary school enrollments to 3 percent, which is the rate of growth of the school-age population. The concern for the quality of primary education has led to a discontinuance of the recruitment of teachers with qualifications below the level of "instituteur", such as "moniteurs" and "instituteurs adjoints". This policy precludes any significant expansion of primary educa- tion even in the long run. Realization of universal primary education will therefore require profcund changes in the Government's primary education policy. Reform plans have existed since 1971 but have never been implemented, and the 1977-1981 Plan does not even mention them. 1.49 The stagnation of primary education in rural areas is not only socially regrettable, but it is also likely to have a detrimental effect on the rural development programs that the Government is undertaking to give increased operational responsibilities to cooperative groups of farmers. M!oreover, the administrative reform now being implemented aims at: Governmental decentralization and an. increase in regional and local responsibilities. All these programs could be jeopardized if they are not accompanied by an attempt to deliver a basic education package to large groups of people in rural areas. T'his would imply a dual effort: on the one hand, effective ways should be devised to deliver such a package to children in rural areas by either adapt- ing the existing primary system or by developing alternative systems, and on the other hand, the literacy programs in rural areas should be expanded with a view to preparing adult farmers for additional responsibilities. For educa- tional and cultural reasons, the Government has decided that these literacy courses be given in the local languages. However, the farmers have asked for literacy training in French since this helps them to communicate with the administrative services, the agricultural development cooperatives and market- ing organizations. Therefore, a prerequisite for a successful acdult literacy program would be the increased use of local languages at the village level by Government and parastatal agencies. Technical and Science Education at the Middle and Secondary Level 1.50 The education. reform of 1971, which attempted to improve the quality of the teaching of science and technology, created two types of middle schools, a general one: "Collages d'enseignement moyen g6neral" (CEMG), and a technical one: "Collages d'enseignement moyen technique" (CEMT). In reality, the difference.s between the two are minimal and little attention is given in either kind of school to the teaching of practical subjects. The Government's recent Pla.n proposes an ambitious expansion program for the middle schools which includes construction of (a) 16 new schools, (b) addi- tional facilities for three existing schools, and (c) improvement of facil- ities in 18 schools; the bulk of this program is devoted to the CEMTs. Enrollments in the CEMGs are not expected to increase much (0.8 percent annually) but the enrollments in the CEMTs are expected to increase by 13.5 percent a year between 1975/76 and 1980/81.  - 34 - 1.51 At the secondary level, the general lycee remains the preferred choice of nearly all middle school graduates. The technical lycee has until now been a second-best option and as a result does not attract the most qualified students. The programs are based on those used in the French technical school and have the dual objective of trying to prepare the students for further studies as well as for direct employment. Over the period 1975/76 - 1980/81, enrollments in the general lycees are planned to increase by about 4.7 percent per year but enrollments in technical secondary education by about 13 percent per annum. Unfortunately, the latter type of training does not adequately prepare the students for industrial employment, certainly not at the supervisory level. The practical training is insufficient and there is, therefore, a need to revise the curriculum. The proposed expansion of general secondary education is very difficult to justify by manpower data. Vocational Training 1.52 The main institution for pre-employment vocational training in Senegal is the "Centre de Qualification Industrielle" (CQI), which gives industrial training in the following fields: general mechanics, motor mechanics, electricity, metalwork electronics, and cooling techniques. The students are recruited from middle school graduates between the ages of 16-21. The main problem with the CQI seems to be that it has not established contacts with industrial employers in Senegal and the profile of the graduates does not seem to meet employers' expectations. Some pre-employment vocational courses are also given in centers attached to the technical lycees. The Plan proposes to create a national training office which would coordinate at a national level the different vocational training activities in the country. It is of prime importance that employers be involved in this activity. Training of Agricultural Technicians and Professionals 1.53 Training adequately qualified personnel in the fields of agricul- ture, forestry, animal husbandry, and fisheries is among the top priorities of the Senegalese Development Plan. Although at present no detailed projections are available of the total need for technicians and managers for the projects in these sectors, rough estimates have been made for the requirements of the public sector, including the rural development corporations (SAED, SOMIVAC, SODEFITEX, etc). These calculations do not include the requirements of the modern sector, nor the projections of the requirements for agricultural extension workers who, according to Government plans, would eventually be replaced by trained farmers. 1.54 At present all training institutions for agricultural technicians are under-utilized with the exception of the National School of Horticulture. This is also the only school which does not guarantee its students future civil service employment. In the other schools, enrollments are restricted by the capacity of the civil service to absorb the graduates. To end present under-utilization and bring down unit costs, all employment guarantees should be withdrawn and enrollments be open. Competitive selection of graduates for public employment would certainly provide incentives for better school results. The vast requirements for technicians once irrigation projects get underway, and current requirements of the private sector in the form of some agro- businesses that are planned, would create sufficient alternative employment opportunities for graduates.  - 35 - 1.55 Junior technician students are recruited among middle-school grad- uates. The length of the course has recently been extended from two to three years. Full agricultural technicians are recruited on the baccalaureate level and are being trained in the "Ecole Nationale des Cadres Ruraux" (ENCR) in a three-year course which provides guaranteed civil service employment. The equipiment in ENCR is inadequate and practically unusable, and the buildings are in poor condition. The "Ecole Nationale d'Economie Appliquee" (ENEA) trains staff for the civil service at the technician level for community development work particularly in rural areas. Recruitment takes place every other year at the baccalaureate level; 20 percent of the places are now reserved for junior technicians with at least four years' experience. The length of the course is three years. It has recently been decided that teacher training for the practical middle-schools (EMP) will take place at the ENEA. Occasionally, the ENEA organizes courses for junior technicians at the request of the employing agencies. 1.56 The most urgent manpower needs in the agricultural sector are found at the professional/management level. Senegal does not have an institution of higher education in the Eield of agriculture. Plans for the creation of such an institution, "l'Institut National Agronomique du Senegal" (INAS) are quite advanced and several donors (FED, FAC, USAID, and Brazil) seem to be interested. The Institute will be located in Thies not far from the National Agriclultural Research Center (CNRA) and the ENCR. The duration of the course is foreseen to be five years. In order to keep the recurrent costs at an acceptable level, it seems advisable that the Government consider merging the INAS and the ENCR. This would also establish an organic link between the training of technicians and the training of professionals and managers. Consequently it would be advisable to reserve a certain number of places in the INAS for technicians with a number of years of professional experience. It might also be worthwhile to study ways that the ENEA and also the Agricul- tural Research Institute in Bambey could be associated with the training programs of the INAS. 1.57 Regarding the fast growing fishing industry, a bottleneck has developed because of the insufficient training of Senegalese fishermen. The utilization and maintenance of new equipment demand skills which the majority of the fishermen do not possess. Fishing further away from the coast requires greater knowledge of navigation, and participation of family enterprises in the market economy necessitates knowledge of basic business management and accounting techniques. A survey of the fishing sector will be undertaken short:Ly and should determine detailed training needs for the different occupa- tions in this sector. Management Training Needs 1.58 There is a genera:Lly recognized need in Senegal for well-qualified high- and middle-level managers in both private and para-public enterprises and in the civil service. The Government is particularly concerned about the sLtuation in the para-public sector, whose deficits represent an increas- ing burden on the Governmenl: budget. Accounting procedures are often inade- quate and management is poor. Managers usually have a reasonable general education, but their skills have been acquired on the job and they lack knowledge of modern techniques of management and accounting which would be applicable in the Senegalese situation. There definitely seems to be a  - 36 - case for the creation of a management development institute which would concentrate initially on in-service training of functioning managerial staff of private and para-public sector enterprises, but which could also be used for the training of high- and middle-level civil servants. 1/ As a first step toward the establishment of such an institution, a survey should be undertaken to identify the potential "clientele" and to determine their training needs. Long-term Planning 1.59 Close analysis of the Government's medium-term education plans reveals that the changes in budgetary allocation to different types of educa- tion will be only minimal. The improvement in technical education will be to some extent at the expense of primary education rather than of general secondary and higher education in specializations which will be increasingly in oversupply. 1.60 The following table illustrates the implications of the proposed policy for the distribution of educational expenditure. Table 18: PROJECTED DISTRIBUTION BY LEVEL OF EDUCATION AND TYPE OF PUBLIC EXPENDITURE (in percentage) 1976/77 1980/81 1995/96 Primary 41.9 40.0 36.7 Middle level - general and technical 8.6 10.0 7.9 Middle level - practical (EMP) - 0.6 14.1 Secondary general 3.3 3.6 2.6 Secondary technical 1.5 2.5 3.5 Other secondary 11.4 11.6 10.0 Higher education 21.0 22.7 20.7 Other expenditure 12.2 9.4 4.5 100.0 100.0 100.0 Source: UNESCO Sector Mission 1.61 In the long run, only a third of the expenditures on education (or 2.3% of GDP) would be spent on primary education, reflecting the limita- tion of available resources and first priority for advanced education. The key problem seems to be the role of EMP. With this program the Government hopes to retain primary school graduates in rural areas, and stimulate rural development. EMP is at present still in an experimental phase, but it is very doubtful whether the above objectives can be achieved. In any case, the Government should realize that development of EMP reduces the funds available for primary education, and that a limitation of expansion of EMP combined with a reduction of admissions into general secondary education and university studies in humanities and languages, would free considerable funds which could be allocated to primary education. 1/ World Bank Group project.,  - 37 - G. Housing 1.62 The housing poLicy in Senegal is biased in favor of the workers in thes modern sector, in particular towards the more affluent among them. Public funds and recurrent subsLdies are concentrated on types of houses that only the highest paid 10 percent can afford. Rents are tightly controlled and do nol: increase with inflatLon. The rural housing policy gave poor results and wi:Ll be replaced by a cooperative approach, but public funding for rural housing will remain very low. The housing needs of the urban poor have generally been neglected. Since the Government has earmarked most slum areas close to Dakar's center ior modern urban development, it is not improving these quarters, and from time to time, slum dwellers are moved to the out- skirts of greater Dakar. Since the old policy has not met expectations and is creating severe financial' problems, the Government is shifting the emphasis towards self-help construction programs. 1.63 The urban housing problem has vast dimensions in Senegal. The Government expects that t:he urban population will grow at a rate of 4.5 percent over the 1971-2000 period. But this growth will range between 7 and 3 percent for the different: towns, with greater Dakar in the upper range. By the year 2000, about 43 percent of the Senegalese population may live in towns as compared to 33 percent at present. The quick growth of the urban popula- ticin was due, to a large extent, to migration induced by the uneven urban/ rural income distribution, but gradually the large difference between urban and rural mortality rates is also becoming an important factor. 1.64 The total housing stock in urban areas can be estimated at about 280,000 units of which 60 percent are poor quality. About 90 percent of the existing units are private. This figure points to the dynamism of the private housing investment sector. At present, the annual average of good quality houses built by the private sector with official approval is about 2,000, roughly the same number as those built by public agencies. At least one thousand units should be added to account for illegal construction. The annual private investment in urban housing can be estimated at CFAF 4 billion, i.e., CFAF 50 billion over the 1960-72 period. A striking characteristic of these private housing investments is that they are 90 percent self-financed. Public housing investments are considerably less. From 1960 to 1972, the Government invested CFAF 18 billion in housing through its two public housing agencies SICAP and OHLM 1/, and CFAF 2 billion on basic urban infrastructure. These investments, which account for about 3 percent of the average total public expenditures, accommodate only 10 percent of the top wage earners of the urban population. 1/ SICAP - Soci&t& Immobiliare du Cap Vert; OHLM - Office d'Habita- tions A Loyer Mod&re. See paragraphs 1.67 and 1.68.  - 38 - 1.65 Although there is no comprehensive data on the demand for urban housing in Senegal, one could use as a proxy the figures available for greater Dakar, where 50 percent of the urban population lives. According to the two housing agencies (SICAP and OHLM), the unsatisfied demand for these subsidized houses is about 20,000 units for medium income groups. The housing shortage, however, is especially severe in low income groups. Since roughly 60 percent of greater Dakar's population is living in overcrowded, poor quality dwellings, one can estimate the housing shortage for low income groups at about 80,000 units. The above figures would imply that on a countrywide basis, the urban housing shortage is on the order of 200,000 units. Conservative estimates show that this shortage is likely to increase over the forthcoming years at an annual rhythm of 10,000 units. A quick comparison between the building volume of quality houses in the past and the growth of the existing backlog demonstrates that a large proportion of the urban population will have to live in substandard housing for a long time to come. 1.66 Conscious of its limited involvement in housing and of the increas- ing housing shortage, the Government sought to tackle the situation within the framework of the Fourth Plan (1973/74-1976/77). About CFAF 20 billion for housing projects were inscribed in the investment program, an increase of about 300 percent as compared with the investments which took place over the last decade. CFAF 10 billion were allocated to the Cap Vert region (where Dakar is located) of which 3 billion were appropriated to the first sites and services project in Dakar. So far, the achievements are below expectations. In the first three years of the Fourth Plan, only CFAF 9 billion out of the 20 billion programmed were committed or, in physical terms, 40 percent of the Plan's targets. The major constraint was the lack of capital, a problem that became even more pressing after 1974 when building costs started to skyrocket. As a result of this situation, the Fourth Plan has done very little to improve the housing situation. 1.67 To implement its middle class housing program, the Government set up two housing agencies. The most important one (OHLM) is a publicly owned institution created in 1960 to implement the Government's low-cost housing program. Over the 1960-1974 period, OHLM built about 7,500 low-cost housing units mostly in Dakar for about CFAF 10 billion. Within the OHLM, a special unit, the Direction des Parcelles Assainies, was created in 1972 to manage the sites and services scheme that aimed at developing 10,000 plots for self-help construction near Dakar. OHLM is financed through national housing taxes and concessionary loans, mainly from France. Over the 1960-1974 period, OHLM received about CFAF 7.3 billion in housing taxes and CFAF 4.2 billion in loans to cover its investment and operating costs. Despite the subsidies and low interest capital aid, the financial situation of OHLM is poor. Rents are too low to cover investment and operating costs of the rented housing units. Moreover, the French aid agency (CCCE) decided to withdraw its financing. These problems are compounded by the dramatic increase in construction costs which went up by 80 percent between 1973 and 1976. As a result, OHLM is lacking long-term resources and is constantly running into liquidity problems. Although rents are far below the economic cost, OHLM's rental units are still not affordable to the overwhelming majority of the urban population. As previously mentioned, only 10 percent of the wage earners can afford OHLM units. Not surprisingly, applications for OHLM units stem from individual earnings from CFAF 300,000 per month to CFAF 700,000.  - 39 - 1.68 SICAP is a semii-public institution created in 1950 to cater to the medium-income groups' demand for housing. From 1950 to 1973, SICAP built about 8,620 units in the Cap Vert region. Financing for SICAP's operations came also from Government funds and French loans on soft terms. As in the OHLM case, CCCE withdrew its financing so that SICAP is badly in need of long-term funds. 1.69 Besides budgel:ary and foreign concessionary funds, some financing for housing was provided by the domestic banking system. The national devel- opment bank (BNDS) granted about CFAF 100 million in housing loans annually over the last decade. The analysis of BNDS's financial accounts shows that il:s long-term loans are steadily decreasing, not only in relative terms, but also in absolute terms. In 1968, BNDS long-term loans including industrial activities stood at about CFAF 2 billion which declined to CFAF 1 billion in 1974. When compared wit:h the roughly CFAF 1 billion annual contribution that France used to channel into the building program, it is clear that BNDS is not in a position to replace these funds. 1.70 Faced with steeply deteriorating housing and financial conditions, the Government decided to put more emphasis on sites and services projects. About 22,100 plots are expected to be developed by OHLM and the DLrectorate of Urbanism. The wisdom of involving two agencies in this program can be dis- ptuted. The estimated tcital development cost is about CFAF 4.6 billion with the cost of prepared sites going from CFAF 225,000 in Dakar to CFAF 190,000 in the provinces. OHLM's middle-class housing program will be stabilized at 1,200 units a year with a total investment cost during the Plan period of CFAF 16.2 billion. SICAP's ,crogram will decrease from 700 to 650 units a year with an investment cost of CFAF 6.6 billion over the four Plan years. Including some smaller items, the total four-year investment program amounts to CFAF 27.8 billion. 1.71 This program resulted from a 72 percent cut-back on the original proposals which were unfeasible by lack of financing. About CFAF 16 billion of the reduced program should come from domestic resources. A new treasury account will be opened for this purpose funded from housing taxes 1/, rents, and sales of land and houses to the occupants. Also, a new specialized credit institution will be created (Credit Foncier). 2/ Foreign concessionary financing is expected to contribute 78 percent to the sites and services projects whereas the middle-class houses may attract 28 percent foreign financing. It is doubtful whether CFAF 16 billion domestic savings will be available for housing over the Fifth Plan period. The Government has considered using foreign commercial loans and giving guarantees and subsidies with the purpose of lowering the interest burden on the beneficiaries. However, this is difficult to justify in view of the general shortage of funds and the relatively high average income level of the beneficiaries. On the 1/ Under the present legislation, half of the housing tax is earmarked for OHLM, i.e., about CFAF 1.2 billion annually. 2/ In June 1979 a majority privately owned Housing Bank was founded to fulfill this function.  - 40 - contrary, subsidies on the middle-class housing programs should be stopped by charging the beneficiaries the full cost of these houses and basing the rents on current construction costs and commercial interest rates. The Government should promote long-term purchase contracts through which it could recuperate the public funds invested, and encourage the private sector to increase its share in this sector by making prepared building sites available at full cost. Moreover, houses should eventually be allocated according to the ability of the beneficiaries to accumulate over a number of years, a certain down pay- ment, savings which would help to relaunch the building program. 1.72 Such a policy would still not provide for the rural population and the urban poor. It would be unrealistic to assume that the Government could have a substantial impact on these categories within the Fifth Plan period. The important conclusion drawn from the analysis of demand was that a large part of the Senegalese population will have to cope with substandard housing for a long time to come. The policy should therefore be based on the improve- ment of slums and rural shelters rather than on their replacement. Because of the large number of people involved, even these improvements should be gradual so that part of the cost can be borne by the beneficiaries.  - 41 - Chapter II. MIGRATION 1/ A. Regiona. Population Distribution and Densities 2.01 A superficial observer could easily conclude that Senegal has no population problem. Vast areas in the southeast are sparsely populated and official estimates give a low overall population density of 26 persons per square kilometer or about 42 persons per km2 of arable land. But such imlpressions and indices are almost meaningless in the Sahel region to which Senegal belongs. About half of the farmers live under marginal conditions crowded together in the groundnut basin where possibilities for yield in- creases are limited by t:he dry climate. In the south and southeast, rainfall conditions are more favorable and would seem to allow for much higher popula- tion densities. However, in the valleys endemic diseases have kept the population down, and the often fertile plateaus have been rendered largely uninhabitable by lack of drinking water. Senegal should, therefore, not be placed in the category of countries without a land constraint. The ratio of population to usable land is uneven and demographic growth can only be absorbed by steady progress towards developing new agricultural techniques and arable land. 2.02 Six more-or-less homogeneous geographical zones can be distin- guished in Senegal, but since statistics have an administrative base, this report has to use the seven provinces (Cap Vert, Casamance, Diourbel 2/, Fleuve, Senegal-Oriental, Sine-Saloum and Thias) as a proxy for the geographi- cal zones. 2.03 The first and most urbanized zone of Senegal is the axis from Dakar, situated at Africa's extreme west point, to Thids, roughly 50 km in the intterior. This zone will be described with the demographic data of the province Cap Vert, leaving out the city of Thias. The second geographical zone is the groundnut basin, which is defined as the strip of land east of Cap Vert between the 600 and 800-mm isohyet stretching roughly 200 km into the interior. A characteristic of this region is its dependence on groundnut and millet cultivation. In this report the groundnut basin is identified with (a) the province of Diourbe]., including a large part of the sparsely populated pastoral zone further to the northeast; (b) the province of Thies, whose western part is already linked with the urban economy of Cap Vert; and (c) the province of Sine-Saloum, half of which benefits from 800 to 1,100 mm of rainfall allowing for scime agricultural diversification. A third geographic zone is composed of Senegal's major valleys, listed in north-south order: 1/ This section is a reduced and modified version of Section I of World Bank: Migration and Employment in Senegal: An Introductory Report (Washington, D.C., September 24, 1976). 2/ Diourbel has recently been split into two new provinces: Diourbel and Louga. Most data are, however, available only for the previous adminis- trative boundaries..  - 42 - the Senegal River, the Sine-Saloum River system, the Gambia River and the Casamance. Only the Senegal River valley can be identified with a province, that of Fleuve. This most northern province includes a desert-like part of the pastoral zone. The Casamance valley is part of Casamance province which is separated from the rest of Senegal by the enclave state, The Gambia. Although groundnuts are still the major crop in Casamance, this province distinguishes itself from the rest of Senegal by possibilities for agricul- tural diversification, thanks to annual rainfalls between 1,000 and 1,600 mm. The fourth geographical zone is the underdeveloped area in the southeast of the country, which includes the Senegal-Oriental province and the eastern part of Casamance. In the following analysis Senegal-Oriental is considered representative of this geographical zone although this province also includes the city of Bakel on the Senegal River with a type of economy and strong outward migration typical of the river valley. A fifth zone, the coastal strip, with rich fishing grounds and underground fresh water reserves, could not be separated out in the statistics nor could the sixth zone which is the pastoral zone mentioned above. 2.04 Senegal's population is unevenly distributed over the different provinces with Sine-Saloum containing the largest proportion of the total population and the Senegal-Oriental the lowest (see Table 19). However, when allowance is made for the differences in area, the Cap Vert region, where the capital city Dakar is located, emerges not surprisingly as the densest (1,790 persons per km). Senegal-Oriental not only has the lowest proportion of the total population, but also the lowest density (less than 5 persons per km). In general, the density of the population decreases systematically with distance from the Atlantic coast.  - 43 - Table 19: TOTAL POPULATION OF SENEGAL BY PROVINCE AND DENSITY, APRIL 1976 Estimated Annual Degree of Population Density Total Population Urbanization Growth Rate per Surface Province '000 Percent (percent) (percent) _ km2 (km2) (1) (2) (3) (4) (5) (6) Senegal 5,085 100 32.5 2.6 26 196,722 Cap Vert 985 19 84.2 4.6 1,790 550 Casamance 736 15 14.9 1.6 26 28,350 Dioubel and Louga 843 16 18.5 2.9 25 33,547 Dioubel (425) ( 8) (97) (4,359) Louga (418) ( 8) (14) (29,188) Fleuve 528 10 26.9 2.1 12 44,127 Senegal-Oriental 286 6 11.4 3.6 5 59,602 Sine-Saloum 1,008 20 15.1 1.5 42 23,945 Thies 699 14 34.2 2.9 106 6,601 Sources: ccl. (1): Bureau national du Recensement: Resultats provisoires du recensement g4n6ral de la population d'avril 1976 (Dakar, 22 July 1976, mimeo), p. 4, tableau 1: R&partition de la population de droit par region. ccl. (2) Division des Ressources Humaines: Prolection de la population aLnd (3): totale pa.r r&aions et par strates (1971-1981 Dakar, Feb. 1976). ccil. (4): Mission estimates. ccl. (5) arid (6): World Heaelth Organization: Profil sanitaire du Senegal (Dakar, September 1976), Annex table I.  - 44 - 2.05 The level of urbanization in Senegal is high compared with other countries in Africa or LDCs in general. In 1970/71, about 30 percent of the total population was living in towns of 10,000 people or more, of which nearly two-thirds were in the capital city of Dakar. An additional 14 percent lived in semi-urban areas (1,000 to 9,999 people), communities which have the characteristics of large rural villages. The rest of the population generally lived in villages with an average population of 150 people. In most parts of the country the rural population is organized in family groups called carres composed of the chef de carre, his younger brothers and the children. A carre normally consists of four to five family units with each family unit averaging about six people (or a total of 25 to 30 persons). B. Ethnic Groups 2.06 Some 90 percent of Senegal's African population belong to six major ethnic groups: Wolof, Serer, Peul, Toucouleur, Diola and Manding. The Wolof is by far the largest accounting for 40 percent of the population. Few Wolof, Serer, or Diola live outside Senegal. Many Toucouleur do, but the great majority live in Senegal. The Manding, Bambara and Soninke are groupings whose main centers are found primarily in neighboring countries. The 500,000 Peul (referred to as Fulani in Anglophone countries) are spread across the more arid parts of Western Africa and are estimated at 7 million in total. Table 20: ETHNIC GROUPS (in percent of total population) 1960 1970 Wolof 36.2 40.5 Serer 19.0 16.0 Peul ( Fulbe) Fulani 12.3 Toucouleur 9.2 22.8 Diola 7.0 8.4 /1 Manding, Bambara 6.4 8.6 /1 Soninke 2.1 3.7 Others 7.8 TOTAL 100.0 100.0 /1 Includes some assimilated immigrants. Source: Harold D. Nelson: Area Handbook for Senegal (Washington, D.C., U.S. Government Printing Office, 1974) and 1970/71 demographic survey.  - 45 - 2.07 Most Wolof live in Diourbel, Thies and Sine-Saloum which together constitute the northern and central groundnut basin. The Serer live mostly in the southern part of Sine-Saloum, which is the southern groundnut basin. The Toucouleur live along the Senegal River in the Fleuve region, though they do not extend into the delta area. The Diola live in west Casamance, the Manding in the southeastern part of the country while Peul are scattered all over Senegal. 2.08 A long history of intermingling and intermarriage has made ethnicity of much less importance in Senegal than elsewhere, but caste and class are important. In contrast with many other African countries, caste and class often override ethnic affiliations. Another reason for so much less inter- ethnic rivalry may also be the fact that in Senegal the proportion of the population who live in towns (32 percent) is much higher than the average for Africa. An additional reason may be the unifying influence of the propor- tionately large Muslim population. Even among Serer only some 15 percent are Christians (though the proportion among highly educated Serer is much higher, which is why it is often mistakenly said that the Serer are predominantly Christian). Many Senegalese say that ethnicity has little bearing on behavior patterns and that official policy is to foster national loyalty rather than to dwell on ethnic differences. This may be true, but the fact remains that in a migration study, ethnicity is too important to be overlooked. 2.09 The Wolof are not only the largest group, but also economically the most dynamic. Their language is the lingua franca in towns and mixed vil- lages. Many Serer have chosen to call themselves Wolof, which might explain the shift in their share of the population between the 1960/61 and 1970/71 surveys. But, in general, ethnic groups have preserved their own languages. Even in Dakar there are a number of distinct ethnic quarters. For example, in one quarter virtually everyone is Diola. There is even some measure of occupational specialization by ethnic groups. In rural areas the Wolof and Serer are mainly involved in groundnut production but with some differences in technology. The Peul were once nomadic herdsmen, and while they continue to be primarily pastoralists, the majority are now settled and interspersed with the other ethnic groups. Insofar as they migrate they do so only sea- sonally in search of pastures for their cattle or because of droughts. The 1972/73 droughts wiped out 15 to 25 percent of their herds, and the herdsmen had to seek refuge wherever they could. There is usually less integration of the Peul with other ethnic groups than among ethnic groups living in the same area because of recurrent conflicts between the Peul and farmers over the presence of the former's cattle near cultivated fields. 2.10 Marked differences exist in the degree of urbanization according to ethnic groups. With les3 than a third of the total population living in towns, 40 percent of the Wolof reside in cities, but only 20 percent of the Serer. The majority of the urbanized Serer and Wolof live in the Cap Vert region which is adjacent to their region of origin. Part of this difference in degree of urbanization can be explained by the tendency of some Serer to register as Wolof, but the difference is too large not to indicate a basic  - 46 - difference in behavior. Dakar itself was originally Lebou territory, a fishing people belonging to the Wolof ethnic group who managed to maintain a relatively important role in this modern town despite their small numbers. Out of the 27 percent urbanized Toucouleur, a vast majority also lives in Cap Vert due to the long standing Toucouleur migration to Dakar. Notwithstanding these high proportions, about half of the population of Cap Vert is Wolof, while the Toucouleur and Serer represent respectively only 21 and 11 percent of Cap Vert's total population. C. International Migration 2.11 Although the volume of international migration is not precisely known, it is safe to assume that it has had only a marginal influence on Senegal's population. At present, the number of immigrants in Senegal is estimated to be about 355,000 of whom 300,000 are from other African countries and 55,000 are from outside Africa, many of them from France. The number of Senegalese emigrants is estimated to be about 210,000 of whom 175,000 are living in other countries in Africa and 35,000 in non-African countries, principally France. This gives a net immigration balance of about 145,000 people. It must, however, be remembered that these figures include children of migrants born after the date of migration and that the period of migration could extend from zero to 30 or 40 or more years. Estimates of the number of Senegalese living abroad and foreigners living in Senegal therefore, cannot provide a basis for measuring the impact of migration on the current rate of population growth, but the magnitude of the numbers indicate its marginal importance. 2.12 Over the last 10 years, however, the effect of international migration on total population growth is most likely to have been positive. The number of immigrants increased from about 129,000 in 1960 to about 222,000 in 1970. Taking into consideration deaths among the immigrants, these figures imply a net immigration of about 114,000 persons during 1960-1970, or an average gain of 11,400 persons per year. 2.13 The little information available on emigration flows indicates that the annual net outflow has been lower than suggested by the number of Senegalese living abroad. Since 1960 when the former French Sudan was dis- solved into a number of independent states, quite a large number of civil servants must have returned to their homelands. More recently, in 1972, the Government of Zaire requested the large number of Senegalese immigrants residing there to repatriate. For these specific reasons backflows must have been considerable. In 1973, 26,000 Senegalese were living in The Gambia, but over the preceding 10-year period this number had increased by only 400 people a year. The largest emigration over the last 10 years was probably to France, growing from negligible numbers in 1960 to 30,000 in 1975, and to the Ivory Coast where about 5,000 Senegalese reside. The movements of the Toucouleurs and Peuls who live on both sides of Senegal's borders are difficult to assess; since there are no apparent reasons for net emigration of these groups; maintenance of the historical distribution over the different African countries over the last 16 years may be assumed.  - 47 - 2.14 Senegal is obviously not part of West Africa's predominant migration pattern from the arid regions in the north to the tropical zones in the south. The Arab orientation and differing ethnic background of Senegal's northern neighbors, and the different political systems that the bordering countries in the south used to have are factors that explain the absence of substantial north-south movements. West Africa's second major migration pattern - from the interior to the coast - has manifested itself in Senegal mostly as a domestic migration flow. The directions of international migration by Afri- cans into and out of Seinegal have been as follows: of the 220,000 who were born in an African country outside Senegal, the largest number of immi- grants came through the country's southern borders - about 75,000 from Guinea BLssau, 57,000 from Guinea, and 33,000 from The Gambia which is surrounded by Senegalese territory. 'rhese numbers represented a significant proportion of the population in the smaller countries of emigration. For example, the immigrants from Guinea IBissau represented about 15 percent of the population oi' that country; immigrants from The Gambia constituted 7 percent of the population of that count:ry in 1970. Of the emigrants who left Senegal in 1970, 50,000 went to the Ivory Coast, 25,000 to both Benin and The Gambia, 20,000 to both MauritanLa and Guinea, 15,000 to Mali and 20,000 to other countries. 2.15 African immigrants are distributed unevenly within Senegal. More than 46 percent havre settled in Casamance and represent about 16 percent of the total population of this southern province. The recently ended war of independence in Guinea Ilissau and the kinship of population on the two sides of the frontier explain this situation. Only 11 percent of the immigrants have settled in Senegal--Oriental but this group represents an even higher percentage (17 percent) of the province's total population. Cap Vert and Sine-Saloum have each received 17 percent of African immigration which is negligible compared to the total population in these regions. Dalcar seems to be the preferred place by immigrants from Mali and Mauritania. However, Mauritanian shopkeepers are found in villages all over Senegal. 2.16 The number of non-Africans in Senegal rose rapidly from about 20,000 in the late 1940s to 61,000 in 1960/61. In the early 1960s their numbers declined, so that by 1970/71 there were about 55,000 non-Africans of whom 30,000 to 40,000 were Europeans and about 15,000 to 20,000 were Lebanese. 1/ Over 75 percent of them now live in Dakar and 64 percent of the men work in commerce or other services. 2.17 Of all residents born outside the country, 57 percent live in rural communities and 12 percent in towns under 10,000 inhabitants. These high percentages, however, are almost exclusively due to foreigners of African origin. This group accounts for only 0.6 percent of the population of Cap Vert and must therefore not be considered an important factor in the urban drift. Of foreigners in rural areas, about half lived in Casamance, 24 percent in Sine-Saloum and 16 percent in Senegal-Oriental. 1/ Harold D. Nelson: Area Handbook for Senegal. Prepared by Foreign Area Studies of the American University (Washington, D.C., U.S. Government Printing Office, 1974).  - 48 - D. Internal Migration Flows 2.18 The Senegalese appear to be an extremely mobile people, moving seasonally, cyclically and secularly in response to different economic condi- tions. At the time of the 1970/71 survey, as many as 503,000 persons or 15 percent of the total population in the country was living in a region dif- ferent from the one in which they were born. This mobility is part of the values and traditions of the Senegalese people and has been acquired under the pressure of a harsh climate which demands a great flexibility to survive; ability to move ranks high among the tools of survival. 2.19 Seasonal mobility in Senegal is induced by the long dry period in which agricultural activity gradually comes to a complete standstill for a five-month period. Seeding is compressed into a very short period in June to allow for sufficient growing time. Weeding is concentrated into the first half of the three to four-month growing season. In the drier parts of the country, the farm cannot sustain for a full year all the manpower needed in the peak season, and part of the agricultural population has to find secondary income during the dry season. This pattern leads to a short "pull" for agricultural manpower around June and a "push" during January-May when the harvest is done. Activity in the cities is largely counter-seasonal. During the summer months the schools are closed and many offices in Dakar are in low gear because of vacations. Also, expatriates usually take their home leave during this period and release their servants. A new antiseasonal factor is tourism, which flourishes in winter and has its low in the hot summer months. The consequence of rural push coinciding with urban pull in the winter, and rural pull occurring at the same time as urban push during the "hivernage" in summer, is a substantial temporary rural-urban migration. This typical Sahel phenomenon is of major importance for all questions related to migration and urban drift. 2.20 From the 1970/71 demographic survey, which was carried out in three stages covering the same groups of families with intermissions of about half a year, it can be concluded that every year 12 to 15 percent of the Senegalese are on the move, of whom roughly one third cross provincial boundaries. 1/ The first stage of the survey, the summer of 1970, coincided with one of the worst crop failures in the history of Senegal which certainly reduced the demand for seasonal labor in that year and may explain the relatively low number of passagers (transients) in that period. But also for the normal summer of 1971, the survey found that the number of passagers in rural areas was equal or lower than during the winter (which is the dry season). The only exception was the Casamance region which has a more year-round type of agriculture. One could conclude from this evidence that the push from the 1/ In the second stage of the survey 26 percent of the passagers (tran- sients) recorded during the first stage were still in the same location; in the third stage, 46 percent of the passagers recorded during the second stage were still in the same location. Thus an important part of the passagers were non-seasonal migrants, about 4 to 5 percent of the total population every year.  - 49 - agricultural sector during the dry winter season is more important than the pull during the wet sumier season. However, it should be borne in mind that the peak demand for agricultural labor is a matter of three to four weeks, much shorter than the full six-month survey periods. 2,21 For Cap Vert a more regular picture emerges when comparing the three stages of the survey. 'Cable 21 summarizes the results of the three stages. Only the data on passag?rs are shown and not those of the absentees since only the first category concerns physically counted people. About 80,000 passagers were in Cap Vert during the dry winter season, and 40,000 during the wet sulmmer season which coincides with the slack period for Dakar. The temporary m:Lgration within the regions (indicated in the boxes) is roughly two-thirds of total movement. 1/ 1/ See footnote page 48.  - 50 - Table 21: NUMBER OF TRANSIENTS BY PLACE OF RESIDENCE AND PLACE OF ORIGIN /1 Place of Origin Place of Senegal- Residence Stage L2 Cap Vert Casamance Diourbel Fleuve Oriental Sine-Saloum Thies All origins Cap Vert 1 6.0 1.7 6.8 3.5 0.5 5.0 6.6 30.1 2. 45.0 4.0 7.0 7.0 0.9 8.6 9.4 81.9 3 27.3 1.7 3.4 3.8 0.3 4.2 2.8 43.5 Casamance 1 1.5 10.0 0.1 0.1 0.2 0.4 0.1 12.4 2 2.4 28.9 0.3 0.5 0.5 0.9 0.6 34.1 3 2.5 37.2 0.1 0.4 0.4 0.8 0.2 41.6 Diourbel 1 1.9 0.1 7.2 0.9 0.1 1.8 0.6 12.6 2 2.5 0.4 26.7 1.1 0.2 3.4 1.4 35.7 3 1.9 0.2 18.2 0.7 0.1 1.5 1.4 24.0 Fleuve 1 0.7 0.1 0.8 0.1 0.4 0.4 9.9 2 2.7 0.2 2.1 14.7 0.2 0.7 1.3 21.9 3 2.1 0.3 0.9 2 0.0 0.6 0.4 13.5 Senegal-Oriental 1 0.2 0.3 0.1 0.3 [ .7 0.4 0.1 5.1 2 0.7 0.5 0.3 0.7 13.6 0.6 0.0 16.4 3 0.4 0.4 0.1 0.3 0.7 0.3 14.1 Sine-Saloum 1 1.6 0.6 4.4 0.9 0.8 15.9| 1,7 25.9 2 3.1 0.9 5.9 1.5 1.0 54.6 1.5 68.5 3 4.3 0.9 4.2 1.0 3.1 53.4 2.0 68.9 Thies 1 4.1 0.2 1.6 0.7 0.0 1.2 |67 14.5 2 4.8 0.4 3.8 1.2 0.4 3.0 .2 38.9 3 3.4 0.2 1.7 1.0 0.0 2.6 [ 23.8 /1 A transient is defined as a person being in his present residence for less than 6 months. In the next stage of the survey, six months later, he has either moved again or is counted as a permanent resident. /2 The 1970/71 survey was done in three stages, interviewing the same families in every round. The first stage was held between May and November 1970; the second stage between November 1970 and May 1971; the third stage between May and November 1971. Source: Direction de la Statistique: Enquete d6mographigue nationale 1970-71, Special tabulations: tji&rations: Mouvements passagers, chiffres extrapoles Serie 1/3 (Dakar, 1975).  - 51 - 2.22 In the case of Senegal it is not justifiable nor possible to distin- guish sharply between cyclical and secular migration. On the contrary, all non-seasonal migration must be seen as part of a two-way flow. The main long-term migration flows from the rural to urban areas occur in waves set off by crop failures, low farm prices or booming economic conditions in towns. Flows to the countryside occur when crops are good, farm prices high or urban economic conditions depressed. The notion of "temporary" and "permanent" migration should also be used with caution; a large part of temporary migra- tion can probably be explained by unrealized expectations of better economic conditions while some types of migration appear to be temporary but are in fact part of a permanent cycle. 2.23 The main long-term migration flows are westward towards the Atlantic seaboard, with Dakar as the principal center of attraction. Other places of attraction are the four next largest towns: St. Louis on the Senegal River, Thias close to Dakar, Kaolack on the Saloum, and Ziguinchor on the Casamance River - all in the west of the country. Since the urban centers expand possibilities for secandary farm products and non-farm incomes during the dry season, the western part of the country is also more attractive for farmers. 2.24 Long-term migration can be analyzed by comparing place of residence with place of birth. In 1970, nearly 37 percent of the population of the Cap Vert region was born elsewhere with a fair spread over the different pro- vinces, while only 8 percent of those who were born in the region lived in other provinces. The number of inter-regional migrants in Dakar is nearly half of inter-regional migration for the country as a whole and therefore represents by far the most important long-term flow of people. Next in order of importance of in-migration are the Thias and Sine-Saloum regions, but the proportion of in-migrants in these areas does not exceed 15 percent. On the other hand, the impact of out-migration was greatest in the Fleuve region; in 1970 more than a quarter of the population born in the region had moved out. The impact of out-migration on the population in the neighboring Diour- bel region was also significant (20 percent). 2.25 During the 1960s inter-regional population mobility increased considerably. Comparing the results of the 1960/61 and 1970/71 demographic surveys, the proportion of people living in provinces other than where they were born increased from about 11 percent in 1960 to 15 percent in 1970. The percentage increased in all provinces except Sine-Saloum: from 30 to 37 in Cap Vert; 9 to 14 in the neighboring Thias region; and from 0.3 to 9.8 in Senegal-Oriental. The proportion of out-migrants also increased in most regions: from 17 to 26 percent in Pleuve, 5 to 12 percent in Sine-Saloum and 4 to 8 percent in Casamance.  - 52 - Table 22: RATES OF MIGRATION L/ BY PROVINCE FOR THE 1960-1971 PERIOD (percent) In- Out- Net- Province migration migration migration Cap Vert 28.5 3.4 25.1 Casamance 1.9 5.3 -3.4 Diourbel 4.2 8.8 -4.6 Fleuve 2.1 16.0 -13.9 Senegal-Oriental 13.6 5.8 7.8 Sine-Saloum -0.7 7.7 -8.5 Thies 9.6 7.8 1.8 All provinces 7.6 7.6 - 1L Number of migrants divided by the average population (excluding foreign immigrants) during 1960-1971. 2.26 For some regions the demographic development during the 1960s represented a break with past trends. Sine-Saloum with 15 percent migrants in its population had a net loss through migration of 58,000 people. This outflow was mainly directed to Cap Vert (25,000) and Thies (12,000). A fairly large number (9,000) of migrants from this region moved eastward to Senegal- Oriental. These figures indicate a reversal of the migration stream, a reversal which was related to the urban attraction of the Dakar-Thias region (obviously Kaolack the provincial capital of Sine-Saloum could not compete), and a further shift of the "groundnut frontier" which was still in the south of Sine-Saloum shortly after the war, but has since moved to Senegal-Oriental. This shift is also clearly demonstrated by the figures on Senegal-Oriental itself. In 1960/61 barely 500 people living in this province had been born outside of it. In 1970/71 the number was 22,600. Despite increasing out- migration, the population of Senegal-Oriental increased over these 10 years by 13,000 people. 2.27 In the other provinces past trends prevailed. Cap Vert had a net gain of 124,000 people over the ten years which represented by far the largest flow in absolute numbers. The net gain of Thias was 8,000 people. Next to Sine-Saloum, Fleuve had the greatest net outflow of 46,000 people. This outflow was mainly directed to Dakar, but about 8,000 went to Senegal- Oriental. Whereas the migration from Sine-Saloum into Senegal-Oriental related to expansion of groundnut cultivation, it can be assumed that migra- tion from Fleuve relates to the prosperous zone of Bakel in the northeast corner of Senegal-Oriental which is in fact part of the geographical zone of the Senegal River basin.  - 53 - E. The Senegalese Migrant 2.28 To understand migration the profile of the migrant should be ana- lyzed and placed against the background of his origin and destination. 2.29 Most of the rural migrants are poor and uneducated but not without alternatives or financial support. One of the reasons is that Senegalese farmers still can find some land if they are willing to move. Except for the few large estates of the marabouts, the religious leaders of the Mouride sect, land is fairly equally distributed. 1/ Traditionally the right to land belonged to the person (or family) who burned the natural vegetation. The head of the family normally gave permanent user rights to persons who were willing to clear the trees on a piece of this land. In many villages, the village chief still has some land to allocate. He gives out land to make his village larger and himself more prestigious. Membership in the extended family also gives some rights to land or to the income of other family members. 2.30 The majority of the rural/rural migrants are, however, not permanent settlers, but people working in a temporary relationship with the farmer. These relationships take two forms. A special case are the "Talib&s", who are youngsters given to the marabout for religious education, and who earn their living by working on the marabout's fields. More common are temporary workers who are fed by t:he farmer and get some land to use plus the necessary tools and animals in exchange for a number of days' work on the farm. Since the worker is there for a short time only, he usually sells his crop to the farmer. If the temporary labor originates from another ethnic group (usually non-Senegalese) he is called "Nav6tane"; if he belongs to the same ethnic group he is called "Sourgha". With the increasing mobility of people this distinction is becoming more vague. 2.31 If married people have previously worked as Sourghas in the less populated regions, they sometimes go back and ask the head of the village for some land of their own and to stay. Most Sourghas are, however, unmarried and come to earn the dowry for their bride. This system is, in some traditional communities, reinforced by a custom which denies rights of land to any unmarried man or which connects full manhood to a period of working as a Sourgha. The system ensures that the elder people in the village either get the work of the young generation or the cash from the dowry. 2.32 The capacity of farmers to use temporary labor depends not only on the size of their land but - since they have to feed the workers -- also on the stock of food that is left over after the dry season. In completely monetized regions the farmer may buy this food, but in most cases he tries to store a sufficient supply of millet. When a region becomes over-populatecd, the farmer may get into a situation where he cannot even feed his own family during the last months before the new crop comes in. This can play an important role in migration decisions. Crop failures, mostly because of droughts, bring margi- nal situations to an immediate crisis. 1/ See Appendix A.  - 54 - 2.33 If land is becoming scarce and the farmer cannot feed his family during the full year, he, or members of his family, have to look for a second- ary income in other sectors during the dry season, or for permanent migration. If they decide to go to town, they are often helped with shelter, jobs and immediate necessities by relatives living there. Help in the provision of shelter is reflected in the reproduction of ethnic communities in Dakar. Assistance in finding jobs is reflected in ethnic concentrations in certain occupations as, for example, newspaper selling (Toucouleur) and housework (Diola). This support allows the migrant to be without employment for some time, which gives him greater bargaining power in the job market. On the other hand, members of extended families in the towns can usually come back to the village and count on subsistence and shelter. This produces a two-way stream of people and income transfers which broadens the economic base of the group. 2.34 The mechanisms of urban migration in Senegal are very similar for both the lower and higher educated migrant. Both categories can get support from their village of origin or from relatives already established in town. The most likely difference between the educated and uneducated urban migrants is that the educated will have less inclination to return to the countryside. Education of the classical type can therefore be an important cause of urban unemployment. In Senegal - except for St. Louis - this problem has not yet assumed large proportions. The adult literacy rate was only 10 percent in 1970. Secondary education reached only 15 percent of the relevant age group. On the other hand, the modern sector has been growing by 3 to 4 percent annually during the last five years, creating still sufficient new jobs each year for this particular group. The number of rural people with secondary education is still too low to influence migration flows signifi- cantly. 2.35 Some evidence can be found in the population surveys that migration of people who would qualify for the modern sector is still very low. About 85 percent of the migrants who had been in their present residence for less than six months had no more than a few years of religious education or no formal education at all; more than half had an agricultural background. An insigni- ficant number (3 percent) had secondary education or more. Finally, the share of inter-regional migration (which can be used as a proxy for long distance migration) out of total short-term migration was about equal for all levels of education (about 35 percent). This supports the view that education in the rural areas is so scarce that the higher educated can easily find better paying jobs in their own regions. The relevant factor in inter-regional migration turns out to be the degree of urbanization. Lower educated people, in particular, more often cross regional frontiers if they are urbanized (about 45 percent) than if they are completely rural (about 35 percent). This profile disappears for those with primary education or more. 2.36 A good part of migration is done by young unmarried people. Once they are settled they marry and raise a family. Married migrants often leave wives and children in the rural areas where they live cheaply and produce their own food. Only if the migrant is successful does he send for  - 55 - his family. Particularly in Cap Vert, migrants under 15 years of age were a much lower percentage of the total migrant population than the share of their age group in the total population. Only in Fleuve was the difference not too large. This may indicate that the inflow of population in that province is of a type that implies less economic risk, which is the case for b.ackflows from other areas, movements of nomads in the pastoral zone of Fleuve, or families settling in irrigation perimeters. 2.37 Most migration appears to be related to desires for better income. The relatively low percentages of passagers that give jobs as motivation for their move (30% in the first stage, 13% in the second, and 20% in the third stage of the survey) is inconclusive evidence. Since an average farm house- hold consists of six people, the "job" may be the main motivation for a relatively small number, while the others will say they come for family reasons. One should al3o bear in mind that a relatively large group of mLgrants who are trying to get a job in town are living temporarily at the expense of family membe:rs and may prefer to give a more personal touch to the reasons of their stay. The same applies to those who go back to the villages. Most of them have some productive activity and will stay if they can improve their situation. We are therefore inclined to add the majority of the 60 to 80 percent of migrants who say they move for family reasons to the category of people who move for economic reasons. 2.38 It would appear that migration as it occurs in Senegal is mostly determined by income possibilities at the margin rather than by the diffe- rences in the average of rural and urban incomes. People attracted by the higher average income in the modern sector must have some grounds to think they may qualify for modern sector jobs, an expectation usually derived from family background or education. Such migration would give rise to "waiting-line" unemployment for which there is no strong evidence. On the contrary, for the higher educated there appears to be no particular indication of movement over regional frontiers. Uneducated migrants may have some hcope for modern sector Employment but know they do not qualify and have to accept the income possibilities which the informal sector offers. If the possibilities in the informal sector decline, the migration inflows would decrease and outflows increase. In Senegal, travel distances are quite short, and the inter-regional network of extended families provides fairly accurate information on the different labor markets in the country. Right of land does not seem to count as a reason for immobility. Although hard to prove with quantitative evidence, all indications are that the population will respond to relatively small changes in income opportunities.  - 56 - Chapter III. EMPLOYMENT A. Employment Trends since 1960 Modern Sector Employment 3.01 On the basis of the quite unreliable estimates available for earlier years, we calculate that employment in the modern sector in Senegal grew at an annual rate of perhaps 1.6 percent from 1960 to 1975 (see Table 23), in other words at a rate certainly no higher than that of the total national labor force and well below the rate of growth of the urban labor force. It follows that the bulk of the new additions to the urban labor force have either been absorbed into informal sector activities or have increased the pool of unemployed. Table 23: ESTIMATED EMPLOYMENT IN THE MODERN SECTOR, 1960 and 1975 Average 1960 1975 1975 Annual (Index: Growth 1960 Rate _100) (%) Government (incl. 45,000 62,000 137.77 2.1 public enterprises) Private 75,000 88,000 117.33 1.1 Total 110,000 140,000 127.27 1.6 Source: 1960: Derived from American University, Special Operations Office, Foreign Area Studies Division: Area Handbook for Senegal (Washington, D.C., US Government Printing Office, August 1963), p. 381. 1975: Preliminary estimates of Division des Ressources Humaines, October 1976. 3.02 Employment in industry stagnated during the early 1960s, although the estimates are sufficiently erratic to leave considerable doubt about actual movements (see Statistical Annex Table 6); it was only in 1969 that the figure for industrial employment finally got onto a trend that carried it well past its 1959 level. Over the 15-year period from 1960 to 1975, the average annual growth rate in industrial employment came to 3.0 percent. Indeed it was only in 1975 that industrial employment reached the level of 25,000 projected for 1964 in the First Plan. 1/ Value added in industry in 1/ Republique du Senegal, Assemblee Nationale: Plan Quadriennal de Devel- oppement 1961-1964 (Dakar, March 1961), p. 107.  - 57 - constant prices fluctuated around a rising trend but the average amnual rate of increase declined from one quinquennium to the next; it was 6.5 percent from 1960 to 1965, 4.8 percent from 1965 to 1970 and 4.4 percent from 1970 to 1975, averaging 5.2 percent over the whole 15 year period, implying an increase in productivity (value added per worker) over the period of 2.1 percent. From 1970 to 1975, the rate of increase in productivity was only 0.3 percent per year. 3.03 Industry in Senegal suffered in the early 1960s from two sets of misfortunes: the closing of a number of markets of francophone West Africa in which Senegal had enjoyed free entry as part of the French West: Africa during the 1950s; and the fluctuations in domestic purchasing power as a result of a cyclical series of drought years and poor groundnut harvests. In more recent years, Senegalese industry increased its range of import substi- tution but further development in that sphere continues to be limited by the constraints of domestic (particularly rural) purchasing power. More recently, Senegalese industry is not doing too well in competition with the Ivory Coast. 3.04 Employment in modern sector construction, which came to around 11,500 in 1959 1/ and in 1961 2/ later showed an absolute decline and was only 9,500 in 1975. 3/ Employment in modern sector commerce and services increased by only about 0.5 percent per year. Government employment (Goverrnment admin- istration and public sector enterprises) increased from around 45,000 in 1960 4/ to around 58,000 in 1968 5/ (3.2 percent per year) but has since grown by only 1.0 percent per year to around 62,000 in 1975; over the 15-year period, this implies an average annual growth rate of 2.1 percent. 3.05 Part of the erratic nature of the employment estimates for the modern sector is due to differences in coverage of and response to surveys. There is, however, a significant additional element of ambiguity introduced by uncertainty with respect to the size of the "temporary" work force. The SONED survey of the modern sector in 1974, for example, notes a "seasonal" work force in the modern sector of between 34,000 and 37,000 per cquarter 6/ 1/ CINAM/SERESA: Rapport general sur les perspectives de dgveloppement au S&n6gal. lere partie, (Dakar, July 1960, 2e edition January 1961), p. 11-7 (13). 2/ BIT: Rapport au Gouvernement de la Republique du Senegal sur l'elabora- tion d'un programme de main d'oeuvre (by Leon Reneau) (Geneva, 1962), p. 9. 3/ Preliminary estimate by Division des Ressources Humaines, October 1976. 4/ American University, Special Operations Office, Foreign Area Studies Division: Area Handbook for Senegal (Washington, D.C., U.S. Government Printing Office, August 1976), p. 381, adjusted for expatriates. 5/ Ministere du Plan et de l'Industrie: Troisiame plan quadriennal de developpement &concmigue et social 1969-1973 (Dakar, July 1969), p. 317. A similar figure may be derived from ILO: Memorandum au Gouvernement de la Republique du Sen&gal sur l'organisation de la planification de la main d'oeuvre (by Edouard Floch), pp. 22-29. 6/ See Statistical Anrtex Table 5.  - 58 - in addition to the 62,300 permanent employees counted as of 31 December of that year. Since the turnover within this floating work force is not known, the total number of persons actually working (or any derived estimates of full time equivalents or rates of under-employment) remains something of a mystery. 3.06 The most recent data on the structure of employment in the modern sector by branch of activity are contained in an analysis by the Division des Ressources Humaines 1/ of declarations submitted to the Direction des Impots as of December 1975. Employment, numbers of establishments and average earnings by branch of activity are summarized in Table 24. The table shows considera- ble variation in average earnings per worker from one branch to another, partly as a function of the different proportions of expatriates among the personnel. Construction, which uses a high proportion of relatively unskilled labor, shows the lowest average earnings of any of the branches. The analysis by the Division des Ressources Humaines also presents an interesting set of breakdowns by size of establishments (see Statistical Annex Table 8). The larger-scale establishments employing 100 workers and more account for over two-thirds of total modern sector employment (33.5 percent in establishments with 100-499 workers and 35 percent in establishments with 500 workers and more). The share of Cap Vert in the modern sector is most marked in the 20-49 worker size class; but for all size classes with 10 workers and more, Cap Vert accounts for between 85 and 90 percent of both establishments and of workers. 1/ Division des Ressources Humaines: L'emploi et les salaires dans les secteurs priv& et semi-prive au mois de decembre 1975 (Dakar, Octobre 1976), from which summary data are reproduced in Statistical Annex Tables 7 and 8 below. An alternative source of data on numbers of establish- ments by size of establishment (employees per establishment) is the Direction de la Statistique of the MinistAre du Travail which produces such tabulations by branch of activity and by province every two years (see Statistical Annex Tables 9 and 10 for 1974 data).  - 59 - Table 24: EST'ABLISHMENTS, WORKERS AND AVERAGE EARNINGS IN MODERN SECI'OR ENTERPRISES (PRIVATE AND SEMI-PRIVATE), 1975 Average Annual Branch of Activity Establishments Workers /1 Earnings - ________________ ______________ __________ ('000 CFAF) Agriculture, fishing 59 7,702 433 M[ining 18 1,416 778 Manufacturing 159 20,481 465 Electricity, water, gas 30 3,213 1,031 Construction 119 9,500 393 Commerce, hotels, restaurants 971 17,035 638 Banking, insurance 71 2,464 1,212 I'ransport, communicaticin 162 6,718 891 Other services 329 9,105 636 Unspecified 1 4 -- I'otal 1,919 77,638 601 1! Including expatriates and other non-Africans. Source: Division des Ressources Humaines: L'emploi et les salaires dans les secteurs prive et semi-prive au mois de decembre 1975 (Dakar, October 1976), tables I-A, II-A, V-B. See also Statistical Annex Table 7 below. Agricultural Employment. 3.07 The employment situation in agriculture is succinctly described by the results of the secc,nd round of the 1970/71 demographic survey. 1/ The s.urvey was carried out during the period December 1970-May 1971, which is the slow season for agricultural activity, with the result that among the rural population 15-64 years of age who defined themselves as being in the labor force, 58.4 percent were recorded as "unemployed" since they had not worked d'uring the previous wee.k. Most of the rural unemployed were in fact part of the agricultural labor force, having previously worked in agriculture. E'urthermore, the figure underestimates the degree of seasonal under- employment in Senegalese agriculture since a considerable number of seasonally unemployed agricultural workers were also to be found in urban and semi- urban areas. 3.08 The definition of the labor force in rural areas, where most of the population is pulled into agriculture at the peak season, also suffers from 1/ Ministare des Finances et des Affaires economiques, Direction de la Statistique and Ministbre du Plan et de la Coop6ration, Division des Ressources Humaines: Enquete demographigue nationale 1970-1971: Analyse des resultats du 2ame passage portant sur la population active (Dakar, May 1974).  - 60 - elements of ambiguity. The extent of female participation in the labor force is particularly difficult to ascertain with any precision as is shown by the remarkable (and unlikely) variation in female participation rates from one province to another that appears from the replies to the 1970/71 demographic survey. It is also illustrated by an incident that arose in an agricultural experimental unit in Koumbidia (Sine-Saloum). When a system of lifting water with animal traction was installed, the chef de quartier decided to limit use of the water thus obtained to watering the animals and had the women continue to get household drinking water from wells by hand in order to keep them from idleness (desoeuvrement). Subsequently, truck gardening was introduced and the women were permitted to take advantage of the more abundant source of water to grow vegetables. 1/ 3.09 In the Sahelian regions of Senegal, particularly in the central groundnut basin, agricultural employment will continue to depend on the seasonal rainfall pattern. Given this overriding constraint there are, nevertheless, several ways of increasing the returns to work in agriculture and also of extending the length of the agricultural season: (a) through the application of animal and chemical fertilizer; (b) through the provision of improved agricultural equipment and a judicious degree of mechanization; and; (c) in the river valleys, particularly that of the Senegal River, through water control and irrigation. 3.10 In the areas cultivated by the Serer, animal manure is traditionally used in a systematic way to restore soil fertility; and although chemical fer- tilizer is hardly effective for cultivating groundnuts in Senegalese conditions, it is effective in growing millet. The integration of livestock raising into family farming is an option which has several pay-offs in the availability of oxen and cows for traction, the creation of manure for fertilizer, and even- tually, the sale of animals for meat as an additional source of income. Agricultural experimental work in a Wolof-settled area in the south of Sine- Saloum has put major stress on developing this aspect of the family farm. 2/ 3.11 Traditional farming in Senegal used only manpower and hand tools until around 1927 but systematic efforts to generalize the use of animal drawn farm equipment were made in the 1930s and after World War II in the 1950s. Horse and donkey drawn seeders and weeders found ready acceptance, and since 1955 the use of ox-drawn equipment has been spreading rapidly. 3/ At the animal-traction level of technology, the introduction of farm implements raises productivity and the incomes arising from farm employment and has 1/ Institut de Recherches Agronomiques Tropicales et de Cultures Vivrieres (I.R.A.T.): Bilan succint d'un essai de developpement experimental: Les unit6s experimentales du Sine-Saloum au Sen6gal: 1969-1973 (Bambey: Centre National de la Recherche Agronomique, June 1974), p. 25. 2/ Institut de Recherches Agronomiques Tropicales et de Cultures Vivrieres (IRAT): Bilan succint. . . , op. cit. 3/ J. Monnier: La mecanisation au S6n6gal: Effets sur la production et 1'emploi (Bambey: Centre national de Recherches agronomiques, November 1974).  - 61 - probably reduced the pressures leading to out-migration from rural areas. 1/ The animal husbandry needed for raising and maintaining animals for traction a:Lso serves to spread out the productive agricultural working season. 3.12 Motorized traction has so far been introduced only by Governmental development agencies in preparing heavy soils in the river valleys and parti- cularly in the Senegal Rtiver Delta. In the case of the Delta, the direct effect has been to increase employment by opening up new areas for settlement. However tractorization :Ls an issue for the near future in the development of the middle valley of the Senegal River. 3.13 Water control on a large scale is far beyond the scope of the individual peasant and has required Government intervention for the planning and construction of the primary instruments: dams and irrigation facili- ties. These facilities also permit the introduction of industrial crops such as cotton and the indusl:rial farming of food crops such as rice and tomatoes which open up a limited volume of additional employment opportunities in the afifected areas. B. Employment Projections Population and Labor Force 3.14 The results of the population census of Senegal carried out in April 1976 indicate an annual growth rate of 2.6 percent during the 1970s, rather than the 2.2 perc:ent previously used in official forecasts. For the ccoming decades, the population growth rate is now expected to rise to a peak of 3.1 percent after the! turn of the century and then to fall baclc somewhat. Over the next three plan periods, the rates are projected to increase as shown in Table 25 from 2.6 percent between 1976 and 1981 to 2.8 percent between 1985 arLd 1989, or an average of 2.7 percent over the period. 1/ This conclusion is no longer valid for certain densely populated regions in the Groundnut Basin. The introduction of light mechanization has reduced farmers' dependence on nav&tanes and sourghas (temporary labor) during the summer months; it also made village heads more prldent about giving out land to newcomers.  - 62 - Table 25: POPULATION AND LABOR FORCE (ACTIVE POPULATION AGED 6 YEARS AND OVER), SELECTED YEARS, 1976-1989 Year /1 Population Projected Annual Labor Force Growth Participation Aged 6 Years Rate Rate /2 and Over (x) '000 M ('000) (1) (2) (3) (4) 1976 2.6 5,115.7 /3 32.7 1,673.8 1977 2.6 5,248.7 32.5 1,707.3 1981 2.7 5,816.2 31.8 1,849.9 1985 2.7 6,470.3 31.1 2,012.8 1989 2.8 7,226.0 30.5 2,205.4 Percent: Average annual rate of growth 1976 - 1981 2.7 2.1 l July of each year. /2 Labor force divided by population. /3 Projected to July 1976 from April 1976 census estimate of 5,085,388 at 2.6% per year. Source: Preliminary estimates by Ministry of Planning, Division of Human Resources, October 1976. Col. (2): Projected from April 1976 at rates shown in col. (1). Col. (3): Tableau I of October 1975 document of the Division des Ressources Humaines, Projections de la population active (1971 - 2000) divided by tableau I of the August 1975 document, Projections demographigues sur la base de 1'enguete 1970-1971. Col. (4): Col. (2) x col. (3).  - 63 - 3.15 The labor force participation rate is expected to decline from 32.7 percent in 1976 to 31.8 percent in 1981 to 30.5 percent in 1989. The labor force is defined to include those in the working ages who are working and those who are capable of working and are looking for work (the un- employed). In the case of Senegal, the working ages are taken as 6 years and over, which means that children who are actually working are included in the labor force but not those who are not working. This accounts for the remark- ably low overall rates of labor force participation shown. An increase in the rate of school attendance will by itself reduce the labor force participation rate. 3.16 The breakdowns of the projected population totals by sex and by zone (urban and rural) are shown in Table 26. The expected rate of growth of the urban populatioa over the next three plan periods to 1989 is 4.1 percent per year; for the rural population, it is 1.9 percent per year. The urbanization rate (urban population divided by total population) is projected to rise from 32.5 percent in 1976 to 34.9 percent in 1981 to 38.8 percent in 1989. 3.17 The sex and zonal breakdowns of the labor force are shown in Table 27. The labor force participation rate for males is projected to decline from 48.4 percent in 1976 to 47.1 percent in 1981 and to 44.9 percent in 1989; for females, from 17.5 percent in 1976 to 17.0 percent in 1981 to 16.6 percent in 1989. The urban participation rate, which is already lower than the rural participation rate because of higher rates of school enrollment in urban areas among the younger ages, is projected to decline less rapidly than the rural participation rate.  - 64 - Table 26: POPULATION BY SEX AND BY ZONE (URBAN AND RURAL) SELECTED YEARS 1976-1989 /1 Total Population Average annual Urbanization growth By_Sex ('000) BY Zone ('000) rate (%) rate (% '000 Males Females Urban Rural (5) i (2) (1) (2) (3) (4) (5) (6) (7) 1976 2.6 5,115.7 2,519.5 2,596.2 1,664.1 3,451.6 32.5 1977 2.6 5,248.7 2,585.0 2,663.7 1,731.5 3,517.2 33.0 1981 2.6 5,816.2 2,864.5 2.951.7 2,027.4 3,788.8 34.9 1985 2.7 6,470.3 3,186.6 3,283.7 2,384.8 4,085.5 36.9 1989 2.8 7,226.1 3,558.9 3,667.2 2,803.5 4,422.6 38.8 Average annual 2.7 2.7 2.7 4.1 1.9 rate of growth, 1976-1989 (%) /L July of each year. Source: Preliminary estimates by the Ministere du Plan, Division des Ressources Humaines, October 1976. Column (2) - Projected from April 1976 at rates shown in Column (1). Columns (3) through (6) - Using the proportions for each year as in Division des Ressources Humaines: Proiections d6mographigues sur la base de V'enquete 1970-71 (Dakar, August 1975), Tableaux I and II.  - 65 - Table 27: LABCiR FORCE AGED 6 YEARS AND OVER BY SEX AND BY ZONE (URBAN AND RURAL), SELECTED YEARS 1976-1989 Year /1 Labor By Sex _ By Zone Force Males Females Urban Rural Number /2 (thousands): 1976 1,673.8 1,218.5 455.3 458.5 1,215.3 1977 1,707.3 1,243.9 463.4 476.6 1,230.7 1981 1,849.9 1,349.3 500.6 553.8 1,296.1 1985 2,012.8 1,465.2 547.6 641.6 1,371.2 1989 2,205.4 1,597.2 608.2 743.8 1,461.6 Average Annual Rate of Growth 2.1 2.1 2.2 3.8 1.4 (percent) Participation Rate /3 (percent): 1976 32.7 48.4 17.5 27.6 35.2 1977 32.5 48.1 17.4 27.5 35.0 1981 31.8 47.1 17.0 27.3 34.2 1985 31.1 46.0 16.9 26.9 33.6 1989 30.5 44.9 16.6 26.5 33.0 /1 July of each year. /2 Using participat:ion rates below. /3 Labor force divi.ded by total population, as derived from Division des Ressources Humaines: Proiections de la population active (1971-2000) (Dakar, October 1975), tableaux I, II, and III; and Projections demographigues sur la base de l'enquete 1970-1971 (Dakar, August 1975), tableaux I, II and III. Source: Preliminary estimates by Ministere du Plan, Division des Ressources Humaines, October 1976.  - 66 - Urban Employment 3.18 Of the estimated population of Senegal of 5.1 million in July 1976, 32.5 percent (1.6 million) was urban. For the 5.8 million population projected for 1981, the proportion that may be expected is 34.9 percent (2.0 million). The labor force participation rate for the urban population (including in the labor force persons aged 6 years and over) is estimated at 27.6 percent for 1976 and 27.3 percent for 1981. The urban labor force for July 1976 is thus estimated at 458,500 and projected at 553,800 for 1981. 3.19 Of the urban labor force in 1976, 30 percent were employed in the modern sector: in private sector enterprises, in public sector establish- ments and in the Government. An estimate of the breakdown of urban employment in the modern sector in 1976 is shown in Table 28 separately for Senegal and for Cap Vert. 3.20 The size of the urban informal sector can be estimated only very approximately: only a small part of it has been enumerated in any systematic way, while the frontier between informal sector activity and open unemployment is difficult to delineate. The numbers in the enumerated categories within the informal sector are limited: 34,000 artisans counted (undoubtedly under- counted) in a 1972 census carried out by the Direction de la Statistique; 5,000 domestic servants declared to the social security agency (perhaps one quarter of the total number of persons working as domestic servants); and 15,000 petty traders reportedly identified in a survey carried out in Cap Vert in the mid-1970s. These 54,000 represent only one fifth of the 227,000 in the urban informal sector estimated in Table 29 as the residual element in the urban labor force on the assumption of an open unemployment rate of 20 per- cent. On this latter assumption, the urban informal sector absorbs 50 percent of the urban labor force.  - 67 - Table 28: URBAN EMPLOYMENT, SENEGAL AND CAP VERT, 1976 Senegal Cap Ver _- Private Public Govern- Total Private Public: Govern- Total Branch of Activity sector sector ment sector sector ment estab- estab- lishments lishments Agriculture, fishing 7,70D 2,300 - 10,000 4,200 400 - 4,600 Mining and quarrying 1,400 - - 1,400 900 - - 900 Manufacturing 20,500 - - 20,500 18,200 - - 18,200 Electricity, water, gas 3,200 - - 3,200 2,000 - - 2,000 Construction 9,500 - - 9,500 8,900 - - 8.900 Commerce, hotels, restaurants 17,003 - - 17,000 15,900 - - 15.900 Transport and communications 6,70) 6,300 - 13,000 6,500 2,900 - 9,400 Banking and insurance 2,50) - - 2,500 2,400 - - 2,400 Other services and Government 9,10) 9,400 45,900 64,400 8,300 8,000 21,000 37,300 Total Modern Sector Employment 77.60) 18,000 45,900 141,500 67,300 11300 21 000 99 600 Registered artisans 34,00') - - 34,000 22,000 - - 22,000 Household personnel 20,00) - - 20,000 16,000 - - 16,000 Grand Totail 131,600 _18 00 45,900 195,500 105,300 11 300 21,000 137 600 Source: Vth Plan.  I . C' I GC |OC. UC C j oC' oooo o I I 1_C 000 0000 0o 100.0000 0 0 _' I CC l C C' .00'CC C C' C'CI C R. 00 00 000 000 0 0Do S O I| * 00 00 000G880 0000 0 0o S. O '' '1 ' D_'O' ..'Or0,.0 G J.*0C 0s S ', C S