Senegal. Philippe Antoine and Gora Mboup

Transcription

1 . Senegal 17 Phlppe Antone and Gora Mboup! Senegal had about 7.7 mllon nhabtants n 1992 (6.88 mllon at the last census n 1988); more than 40 percent of the populaton lve n ctes, wth one-half n Dakar, the captal. Essentally Sahelan, the country s n the mddle of an economc and socal crss. ts economy depends very much on the exportaton of raw products such as peanuts and phosphate. Urbanzaton does not rely on ndustralzaton. Agrculture, domnated by the producton of peanuts, was partcularly affected by the drought of recent years and by the fluctuaton of the world prces of peanuts; therefore, ncome n the rural areas has become scarce. Wthout the dversfcaton of crops, food producton s nsuffcent to feed the populaton. Rce s a food of frst necessty n Senegal and remans mostly mported. The local producton of rce fals to meet the needs of the populaton. The phenomenon of rural exodus s stll gong on, and the rural populaton contnues to mgrate forced by the deteroraton of ther natural, economc, and socal envronment. Data from Dakar, organzed by the Offce de la Recherche Scentfque et Technque d outre-mer (ORSTOM) and the nsttut Fondamental d Afrque Nore (FAN), provde an mportant quanttatve nqury. Ths nqury was mplemented on a stratfed sample of resdences (2,100 households spread out n the whole cty) to obtan a comprehensve pcture of the mgratory system and of the economc actvtes n the cty. The nqury has provded the gatherng of mgratory, professonal, and famly bographes of 1,550 ndvduals (men and Translated from the orgnal French text by Sarah G. Brownng and John Gron.

2 Ar 280 Selected Countres Senegal 28 1 Y Fgure 17.1 Major Ctes n the Republc of Senegal MAURTANA M\ GAMBA KQM \ Source: Adapted from Mnstre de l Econome et de Fnances, Drecton de la Statslque, Dvson de Enquetes et de la Wmographc and Demographc and Health Surveys, lrd/westnghouse, 1988: Enqutte Demographque et de Sante au StMgal women) to underscore nterrelatons between movng, the acquston of employment, and the creaton of a home (Antone et al., 1992). URBANZATON N SENEGAL The Steps of Urbanzaton Colonzaton gave urbanzaton the character t stll has today. The large ctes of today were founded by the former colonal authortes n chosen stes, wth functons lnked to ts needs, and the localzaton of large urban centers had a great mpact: Dakar consttutes the perfect prototype (Fgure 17.1). Urbanzaton n Senegal went through several phases lnked to colonal hstory (Champaud, et al., 1985). Frst of all, the towns on the rver Senegal, the ports of call, developed along ths artery of communcaton and have carred on partcularly the rubber trade (Sant-Lous founded n 1659, Podor, Matam, Bakel). ~ The next phase was marked by the colonzaton of the whole country, the de- velopng of the culture of peanuts, and the expanson of towns near ralroads (between 1885 and 1924). Peanut tradng led to a pyramd-lke urban network composed of multple collectng centers and three port outlets: Dakar, Rufsque, and Kaolack. ndependence n 1960 ended the peanut trade, roads supplanted ralways, and urban functons were modfed. The urban system lost ts balance, and through the years, a whole seres of factors combned to make Dakar the man urban development pole of the whole country. Dakar s hstory shows a progressve monopolzng of actvtes spread out n varous parts of the century; the Dakar harbor supplanted those of Sant-Lous and Rufsque. n 1930, Kaolack s port actvtes were transferred to Dakar. n 1958, Sant-Lous was strpped of ts functon as Senegal s captal to the beneft of Dakar. Lttle by lttle, startng n 1960, Dakar began to monopolze the country s ndustral actvtes, ncludng the peanut ol factores and the wholesale busnesses (large busness companes shut the doors of ther branches located n the nland towns). The Growth of Dakar Dakar s the oldest French-speakng cty of black Afrca. For a long tme (seventeenth and eghteenth century or durng the whole perod of the slave trade) Europeans stayed on the sle of Goree, facng Cape Verde. n 1857, the French took control of the Senegal coast, and Dakar was establshed as an urban dstrct, ndependent of Goree. n 1866, Goree had 3,400 nhabtants and Dakar only a few hundred. By 1891, Dakar already had 18,OOO nhabtants and Goree hardly n 1885, the ralroad from Sant-Lous to Dakar was completed. The great constructon works at the Dakar harbor and publc buldngs were completed durng the perod, and Dakar became the captal of the French Westem Afrca federaton. From ths tme on colonal authortes decded to separate European dstrcts from Afrcan descent. n 1915, the natve dstrct of the Medna, next to the Plateau, was created. Later, the town expanson took place by consecutve operatons of housng developments, especally at the end of World War, when an attempt was made to mplement a new urban polcy favorng the ntegraton of the communtes. Dakar s demographc growth rate was 5.8 percent annually between 1921 and 1951; t ncreased substantally between 1951 and 1961 to 8 percent. Captal of the French West Afrcan natons untl 1958, Dakar lost ts poltcal role n Frenchspeakng Afrca at the tme of ts ndependence. But the transfer of Senegal s captal from Sant-Lous to Dakar has allowed Dakar to renforce ts promnence among the other towns of Senegal. Between 1955 and 1961 ts growth reached a rate of 9 percent. As Dakar contnued to grow, t generated ts own replca: Pkne (accordng to M. Vemere s expresson). n 1955 the admnstratve quarters n Dakar began expandng, resultng n people movng to Pkne. Although Pkne was not as l

3 282 Selected Countres Senegal 283 a well equpped as Dakar and most of ts habtat was rregular n land values, but ts buldngs had defnte qualtes. n spte of a slowng down n the demographc growth4 percent annually between 1976 and 1988 (Mbodj, 1989), Dakar stll has an ncrease of 55,000 people per year, the sze of an average regonal captal. Ths concentraton of populaton goes together wth a varety of actvtes such as admnstraton, servces, and ndustres. The cty contrbutes 67 percent of the ndustral producton and 73 percent of the natonal added value. The Urban Network Dakar monopolzes numerous functons, and ts relatve weght n the urban populaton ncreases. n 1955, the Senegal urban populaton was 545,000 people n 24 ctes, 16 of them wth less than 10,000 nhabtants (Manet, 1988). At present. out of the 37 dstrcts classfed as ctes, the number wth over 100,OOO nhabtants went from 4 n 1976 to 7 n 1988, ncludng Dakar and Pkne. Two dstrcts have less than 5,000 nhabtants; by contrast, 16 dstrcts have more than 5,000 nhabtants. Accordng to the crtera, all dstrcts or ctes of more than 5,000 nhabtants. the urban populaton vares between 2,650,000 and 2,890,000 nhabtants (Ba and Sam. 1990), whch demonstrates that 40 percent of Senegal s populaton lve n ctes. On the whole, ntemal mgratons have lttle effect on some nteror ctes. For example, Sant-Lous (115,372 nhabtants n 1988 and 2 percent annual growth between 1976 and 1988). Longa (2 percent), and some smaller towns of the country s nteror regons have lower growth rates. The two man ctes of the nteror, Thhs (175,465 nhabtants n 1988 and 3.6 percent annual growth) and Kaolack (152,007 nhabtants and 3.2 percent annual growth), have slghtly hgher rates. The ctes of the nteror raxely show a substantal demographc growth typcal average ctes that became major centers of ther regons are Fatck (18,416 nhabtants and 5.2 percent growth) and Kolda (34,337 nhabtants and 5.1 percent growth). Mbour (6.3 percent growth) and Zgunchor (124,283 nhabtants and 4.9 percent growth) also represent hgh populaton gans whch rest on real economc dynamsm. The urban herarchy of the ancent nteror town of Rchard Toll changed very lttle. Thanks to sugar cane plantatons and sugar mlls, t grew from 1,000 nhabtants n 1970 to nearly 30,000 n 1988 (and became the 12th largest cty). On the whole, the mbalance of the urban network s ncreasng to the beneft of the captal, whch receves about one Senegalese mgrant out of every fve. THE DEMOGRAPHC COMPONENTS OF URBAN GROWTH The man factors of urban growth are, of course, mortalty, fertlty, and mgraton. The Senegalese fertlty survey (ESF, 1981) and the Senegalese de-! mographc and health survey (EDS, 1988) are sources for mortalty and fertlty estmates; for mgraton data the results come chefly from the 1988 census. Mortalty on the Declne For the last two decades mortalty has declned sgnfcantly, partcularly at the chld level (from brth to age 5). At the natonal level, the nfant death rate (from brth to the ffth brthday), between 1963 and 1985, dropped from 287 per thousand to 191 per thousand, a decrease of 32 percent. Ths declne s relatvely larger after than before the frst brthday: the rsk of dyng from bah to the frst brthday declned from 103 per thousand to 86 per thousand at the natonal level (a decrease of 17 percent), whle the rsk from the frst to the ffth brthday dropped from 190 per thousand to 114 per thousand (a declne of 40 percent), Durng ths perod, chld mortalty remaned hgher n the rural zones than n ctes, partcularly n the captal cty. Chld mortalty n urban areas s now 68 percent as hgh as n rural areas (70 per thousand compared to 102 per thousand) (EDS, 1986). Ths nequalty of mortalty becomes hgher when chldren have reached ther frst brthday. The lkelhood of an urban one-year-old dyng before ts ffth brthday s, ndeed, less than twce that n the rural area (71 thousand versus 164 per thousand). The crude death rate at the natonal level dropped from 27 per thousand n 1960 to 19 per thousand n 1978, then to 16 per thousand n 1981 and n Dakar, where the mortalty level s the lowest, the mortalty rate was estmated to be 12.7 per thousand n 1981 and As an ndcaton of the excellent state of health of the populaton, the declne of nfant and chldhood mortalty reflects the prolferaton of health programs, the mprovement of santary nfrastructures, and the better knowledge people have of hygene condtons and chldren nutrton. n Dakar, there are concentratons of santary unts (more than 50 percent of health centers, centers of chld and maternty care, and hosptals), and an ncreasng propagaton of publc health counselng, whch have contrbuted to lowerng the mortalty rate of nfants and chldren. The most hghly educated women lve n Dakar, and the tranng of young mothers contrbutes sgnfcantly to the lowerng of chld mortaltyelmnatng deaths due to the lack of hygene, encouragng better care for nursng and feedng the newborn, and elmnatng poor nterpretatons of symptoms or use of medcnes. n 1986 only 9 percent of the females years of age had a secondary educaton, whereas 21 percent of those lvng n urban areas studed n secondary schools. Twenty-fve percent of the urban Senegal women had a prmary educaton, compared to only 14 percent n the entre Senegal populaton. n the urban areas slghtly over half of the urban women never attended school, whereas almost 80 percent of them n the RCpublque of Senegal had no formal schoolng (Mnstere de l Econome et des Fnances, 1988). The demographc and health survey (1986) demonstrated that chldren whose

4 + 284 Selected Countres Senegal 285? mothers had an elementary or a secondary school educaton had two or three tmes more chances to reach ther ffth brthday than those whose mothers never attended school (140 or 225 vs. 72 per thousand) (Mnsttre de l Econome et des Fnances, 1988). Deaths due to measles, darrhea, and malara are decreasng rapdly n the ctes. Chldren of women who have a hgh school educaton, or whose husband or partner s workng for the modem sector, are better protected aganst malara, darrhec dseases, and malnutrton than others. n Dakar and other urban areas respectvely, 10 to 15 percent of all chldren suffer severe malnutrton versus 25 percent n rural areas. Chldren lvng n the captal cty are not only better protected aganst these dseases, but are also better taken care of when they are affected by them (Mnsttre de l Econome et des Fnances, 1988). The low level of mortalty of chldren lvng n the captal does not apply to poor famles, although ther mortalty rates are not as hgh as n rural areas. n fact, the EDS data (1986) demonstrate that the chldren of Dakar women whose husbands or partners work for the nformal sector have less chance to survve than those employed n the modem sector. Fertlty For the frst twenty years after ndependence, fertlty remaned hgh n both rural and urban areas. The Senegal fertlty survey (ESF, 1981) shows that Dakar famles have as many chldren as rural famles. The offsprng of women wth completed famles s over seven chldren for each woman. Urbanzaton does result n a decrease n fertlty, but at present the opposte s true, as the socal transformatons takng place n ctes are favorable to an ncrease. Dakar women many young: more than half before the age of 19. The ESF (1981) and EDS (1988) estmate the average age at marage s respectvely 18.3 years old and 18.6 years old (MBDUP, 1992), close to the natonal average. Therefore, Dakar women start ther maternty very early, at an average age of 19. Urbanzaton, whle brngng mprovements n santaton and a better knowledge of hygene, decreases the rsks of mscarrages and nvoluntary abortons and therefore ncreases fertlty. Moreover, t contrbutes to the declne of tradtonal belefs related to sexual taboos and lengthy nursng. ndeed, Dakar women breast-feed and abstan durng the postpartum perod not as long as rural women (respectvely 16.5 months versus 20; 5 months and 6.7 months versus 8.4 months). The reducton n the duraton of postpartum and nursng abstnence s not accompaned by a concomtant ncrease of contracepton. Less than 5 percent of cty women use contraceptve methods. Thus, ntervals between brth are shorter for Dakar women than for those lvng n rural areas. There s an ncrease n potental of chldren, that s to say n natural fertlty, as demonstrated by B. A. Easterln and M. C. Crmmns (1982) and J. Bongaarts (1982). c 1 At the present tme, urbanzaton, nstead of reducng fertlty as t does mortalty, tends to ncrease t or to mantan t at ts hgh level. The lack of synchronzaton between mortalty and fertlty n Dakar shows that the,only urbanzaton crtera s nsuffcent to explan the demographc regme specfc to Senegalese ctes. t s mportant to ntegrate other dmensons such as the level of educaton and the famly s socal poston n the economc lfe and n the modem sector. Fertlty levels depend, ndeed, on the condtons of producton and reproducton specfc to each socal group, accordng to ts degree of nvolvement n the urban actve lfe (Locoh, 1988; Mboup, 1992). A decrease n the fertlty of women wth hgher educaton or a husband employed n the modem sector occurred. However, those consttute a mnorty n the Dakar female populaton (less than 10 percent.) n 1986 Senegal women years old wth no formal educaton had 6.8 chldren, those wth a prmary educaton had 5.7 chldren, whle women wth a secondary educaton had only 3.8 chldren (Mnsttre de l Econome et des Fnances, 1988). The majorty of the women n urban Senegal (79 percent) are ether llterate or have only an elementary level of educaton, and belong to poor famles: ther employment s precarous (nformal sector), ther salary s very low, and they do not have penson or retrement benefts. Ths s the reason that ther behavor vs-hs procreaton has no sgnfcant mpact n lowerng the fertlty of all women of Dakar. For those famles, a chld remans a very mportant economc value (securty n old age and labor). The ESF (1978) and the EDS (1986) demonstrate, as a matter of fact, that cty women seek large famles (more than 6 chldren per woman). Fnally, bearng n mnd that the evoluton n fertlty has not followed the decrease n mortalty, cty women, partcularly those n the captal, have more survvng chldren than ther rural counterparts. The EDS data show that, ndeed, for women of years of age, 89 percent of the urban chldren at brth survve compared to 75 percent n the rural areas; at years of age the gap s even wder: percent. Mgraton Patterns Accordng to the 1988 results (Ba and Sam, 1990), out of a populaton of 6,881,919 ndvduals bom n Senegal and resdents n 1988, there were 999,060 lfetme mgrants who had changed ther regon of resdency. They are mgrants who resde n a dfferent regon from ther brthplace; ths does not take nto consderaton the length of resdency, nor the number of mgratons; fnally, t hdes return mgratons. The areas of Sant-Lous and Louga consttute genune poles of emgraton. These two regons suppled 277,180 lfetme mgrants, that s, 31 percent of mgrants from nsde the country. The Dakar regon, whch consttutes the man pole of destnaton, receved 462,090 mgrants, 46 percent of the total. The Kaolack, Fatck, and Thbs regons are other poles of destnaton,

5 286 Selected Countres but not comparable wth Dakar. f one takes n consderaton the soldes mgratores mgratory balances among the ten regons of Senegal, only two showed a postve balance: Tambacounda wth a slghtly postve balance of +2,740 (ths regon has a low mgraton but t s helped by the cultvaton of cotton); and Dakar, whch has a net gan of 325,580 lfetme mgrants. Of the 889,550 lfetme mgrants who mgrated wthn the country, 462,090 or 52 percent settled n Dakar. The man mgratory flow toward Dakar orgnated from Thks (105,940 persons n Dakar were orgnally from that regon); then from Sant-Lous (86,190), and natves of three regons, Zgunchor (58,440), Dourbel (55,020) and Louga (52,290); and fnally, mgrants from Kaolack (39,630) and Fatck (38,400). The proporton of lfetme mgrants n the departure area was reported. n relatve weght, Zgunchor provded the most mgrants to Dakar. The 1988 results provde the regonal drecton to Dakar, n the last fve years. For the entre Senegal, only three regons have postve balances: Dakar (+ 19,260), Zgunchor (+6,660), and Tambacounda (+ 150). The Zgunchor regon s a paradox nsofar as t supples an mportant share of Dakar mmgrants (15.2 percent) and receves mgrants from neghborng regons. Of the ntemal mgrants to Dakar, the man flux was of natves of ThEs followed by those of Sant-Lous and Zgunchor. Emgrants from these three regons went essentally to Dakar snce 62 percent of emgrants from Sant-Lous, 61 percent of the emgrants from Zgunchor, and 57 percent from Thhs went to the captal. The mgratory flux contnues, but t s not a rural exodus. On the contrary, the dversty of the flux shows (FAN-ORSTOM survey) that a large number of the rural-urban mgrants were orgnally from an urban mleu or else passed through an urban mleu. The mportant urban mgratons to Dakar suggest the evoluton of a Senegalese urban network. APPRASAL OF THE DEMOGRAPHC GROWTH The declne n mortalty, much faster than that for fertlty, results n a natural ncrease of the populaton at the natonal level at 2.9 percent. The latter s hgher when one consders the captal, whch has the lowest mortalty rate (between 10 and 15 per thousand), and a brthrate dentcal wth the natonal average (46 per thousand). The rate of natural ncrease n Dakar s between 3 and 3.6 percent. Accordng to the FAN-ORSTOM study, the yearly mmgraton rate s approxmately 3 percent. The dfference between the annual growth (4 percent) and the natural growth n Dakar (between 3 and 3.6 percent) gves an estmate of the mgratory remander, between 0.4 and 1 percent. Therefore, each year, between 2 and 2.6 percent of the populaton leaves Dakar. These emgrants (approxmately people) are for the most part mgrants returnng to ther homeland; the others are ntematonal mmgrants who are leavng Senegal ether for other Afrcan countres or for Arab countres, Europe, or North Amer- Senegal 287 ca. ntematonal mgratons are becomng more and more dversfed and represent one of the strateges used to escape the crss. Even f the demographc growth of Dakar has slowed down, the cty contnues to welcome a great number of nterurban mgrants from Senegal. The mgratory phenomenon s partcularly evdent wth the actve age-group, and even a low mgratory growth can hde mportant fluxes wth certan age-groups. For nstance, n Dakar, n 1989, 30 percent of the men and 24 percent of the women between the ages of 25 and 29 arrved after Regardless of the generaton, arrvals are concentrated between 15 and 25 years of age. Mgraton affects especally adolescents and young adults, who must face new responsbltes n unknown surroundngs. Mgratons have an mpact on the age and sex structure n Dakar; 48 percent of male mgrants are between 20 and 39 years of age, and 47 percent of female mgrants are between 15 and 34 years. Women mgrate at a younger age than men, but the women represent a slghtly smaller percent at each age than male mgrants. Ths nfuson of younger blood n Dakar, whch shows for 1989 a relatvely regular age profle, contrary to that of 1955 whch showed a narrowng at the age level, and a defnte ncrease at the level, along wth an overrepresentaton of men beyond 30 years of age. Ths reflects the typcal profle of an expandng cty welcomng mgrants. Despte the slowdown n urban growth, the stuaton remans very alannng: urban equpment and employment are not keepng up wth the rhythm of demographc growth. LVNG CONDTONS N DAKAR Vared lvng condtons are found n Dakar; vllas, buldngs from the end of the colonal era, developments whch are more or less socally orented, more tradtonal concessonare evolutve dwellngs, and sheds. The dwellng s an ndcaton of socal dfferentaton. Three large categores of dwellngs can be dentfed n the cty of Dakar. The frst consttutes the dwellngs of hgh and mddle standng: vllas, apartment buldngs, and apartments n housng developments Habtat 2 Loyer Mod6r6 (HLM) and Socf36 mmoblhre du Cap Vert (SCAP). The mddle class remans the prncpal benefcary of urban plannng. The SCAP (Cap Vert Real Estate Corporaton) was created n 1950, and the OHLM (Low ncome Housng Offce) n Snce ther creaton, the real estate corporatons have bult more than 1 1,000 housng unts for SCAP (mostly between 1960 and 1980), and around 8,000 unts for HLM (mostly between 1960 and 1970). The second category ncludes the rather low-ncome style of housng: multstory houses, houses wth terraces, and houses wth conugated ron or tle roofs. The thrd category ncludes the sheds. Several modes of housng constructon are possble. One part of the constructon s a drect consequence of the cleaned lots development project (partcularly

6 288 Selected Countres Senegal 289 the project dealng wth the clean-up of 10,500 land lots n Cambarene whch have not yet all been completed). Constructon can also be prvate, but wth the help of loans (BHS [The Constructon Bank]). Yet the greatest part of today s constructon comes from self-constructon and jobber s projects. Habtatons of the HLM or SCAP type represent only 14 percent of all dwellngs. The second category s the most often seen n the cty of Dakar; t ncludes multstory houses, houses wth terraces (24 percent of avalable dwellngs), and houses wth cormgated ron or tle roofs, whch by themselves represent 53 percent of all dwellngs. Fnally, the sheds are few (8 percent). One fnds more sheds n Dakar (10 percent) than n Pkne (5 percent). From the results of the FAN-ORSTOM study, and comparsons wth the adjusted fgures from the 1955 census, one can descrbe the evoluton of the lvng modes of the dwellngs, and the consequences of the ncrease of the famly sze on resdental strateges. The nature of the dwellngs changed between 1955 and The 1955 census provdes eloquent nformaton on the topology of habtatons and the composton of households. Of all tradtonal constructons, only 13 percent were made of concrete, 54 percent were made of wood, and 33 percent were made of straw. Constructons made of tradtonal materal (straw) or salvaged materals (boards) are beng replaced by concrete. The mprovement of dwellng condtons s above all the result of judcal constrants and the evoluton of urban developments durng the past thrty years. Before ndependence, constructon wth temporary buldng materals was authorzed n the Afrcan Quarters, but the nhabtant only obtaned a fnal property ttle when bult out of permanent materals (Snou, 1990). Rather quckly, n the fftes, authortes were overwhelmed by the arrval of new mgrants, and many shantytowns appeared on the nondeveloped urban frnges. t was at that tme that a new polcy of massve exodus of the llegals toward the perphery began. Ths polcy was ntensfed after ndependence. Although, the 1967 urbanzaton plan dd not have the means necessary for ts mplementaton, t represented a tumng pont. Spontaneous urbanzaton s no longer consdered an urban perverson whch must be elmnated (Snou, 1990). Dwellng norms are beng lowered to allow more people to fnd a mnmal habtaton and to avod the emergence of shantytowns. As far as the latter s concerned, the plan dd succeed because no real shantytowns exst, and there are very few dstrcts of only sheds, snce these are rather scattered n varous dstrcts, sometmes hdden behnd concesson walls. Ths constructon topology s ndependent of the land status of the dwellng: a spontaneous habtaton can be of good qualty, and a shed can occupy a lot whch has been dvded and regstered for a long tme. Probably, 17 percent of all dwellngs come from state constructon companes, 5 to 6 percent are prvate dwellngs subsdzed by the state (loans), 22 percent result from prvate real estate transactons, and 53 to 56 percent of the dwellngs have no offcal orgns (CCCE, 1991). 1 The proporton of homeowners s relatvely hgh n Dakar, where 48 percent of heads of famles own ther dwellngs. The proporton of homeowners ncreases from the center to the perphery, and t goes from 38 percent n the downtown area, where renters are n the majorty, to 49 percent n the nearby suburbs (Yoff, Patte d oe, Camberene, Parcellas) and to 58 percent n the more remote suburbs of Pkne. Among those who call themselves homeowners, half have property ttles, or lodgng permts, whle the other half are lvng wthout property ttles, especally when the lots have been obtaned from trbal chefs. Homeowners have large famles. f we take nto consderaton the populaton as a whole (and not just heads of famles), 67 percent of the populaton lves n famles whose head s a homeowner, and 28 percent are renters. We should emphasze the large number of dependents who lve wth the heads of households. More than 30 percent of the populaton lves wth a parent. The famly composton has undergone major changes, and several assessments can be made of ts evoluton between 1955 and The average famly sze went from 4.2 persons n 1955 to 8.3 n The second assessment deals wth the decrease n the number of women who are heads of households (15 percent n 1989 and 18 percent n 1955). although the trend s reversed n other Afrcan captals. The two most strkng evolutons deal wth the agng of the male heads of households and the ncrease n the number of polynuclear households. n 1955, 21 percent of male heads of households were over 49 years of age, and n 1989 ths number has reached 37 percent. On the other hand, the proporton of polynuclear households has ncreased from 7 to 24 percent of all households. One may wonder about the consequences of the ncrease n sze of households and ts effects on the lvng condtons of the resdental space. As far as the concessons, or lots, are concerned (n general 200 to 400), one or more households often lve together (1.5 households on the average n 1989). There s a densfcaton of the concessons, especally n the downtown dstrcts of Dakar; but the overpopulated concessons do not ncrease; 5 percent of concessons had more than 30 nhabtants n 1955, and 4.9 percent n f the ncrease n the sze of households has resulted n a very relatve densfcaton of the concessons, the densty per room has evolved. Whereas n 1955, 29 percent of all households and 45 percent of the populaton lved wth more than 3 persons per room, the stuaton worsened by 1989 when 44 percent of households and 54 percent of the populaton lved n more than 3 persons per room. Accordngly, the densty jumped from 2.1 to 2.9 persons per room. Obvously, the ncreased sze of households had repercussons on the lvng condtons of the house space avalable to households. URBAN EQUPMENT The access of households to drnkng water (faucet n the home) and electrcty has markedly mproved. n 1955, two major types of habtatons are consdered

7 290 Selected Countres Senegal 29 the European dwellngs (18 percent of the populaton lves n them, of whch only a lttle more than a quarter are Afrcans), and the Afrcan habtaton (housng 82 percent of the populaton). Ths dchotomy no longer has much meanng today. But, n order to allow for comparsons, we have for 1989 (FAN ORSTOM study) grouped, n the same category, hgh-standng habtatons and the constructons completed by real estate corporatons, and on the other hand, habtatons n the concessons. The frst category sheltered 15 percent of the populaton n 1989, and the second category sheltered 85 percent. The frst category, regardless of the tme perod, s connected n 93 percent of cases to the network of runnng water and electrcty. n the 1989 concessons, the stuaton mproved: nearly 40 percent of households have both water and electrcty at ther dsposal; stll 57 percent are wthout water (91 percent n 1955) and must gather t mostly at publc fountans. whle 40 percent (80 percent n 1955) have no electrcty. The fght aganst shantytowns has turned out to be relatvely effectve; the dwellngs made of temporary buldng materals have gven way to dwellngs constructed of concrete. However, that doesn t mean that everyone has a decent place to lve. As far as the water and electrcty supply s concerned, successful efforts have been carred out, but more than half of the populaton of workngclass dstrcts do not have runnng water n ther homes, wth all the consequences for hygene and health whch result from problems lnked to the stockng of water. The cty of Dakar s better equpped than the rest of the country n communty equpment but, nevertheless, suffers from a lack of nfrastructure. The cty budget s approxmately 6,700 cfa francs per nhabtant (1 U.S. dollar = 300 FCFA). Wthout spendng much tme on major equpment wth a natonal orentaton, what does neghborhood equpment consst of? There are three prmary classes for each 1,OOO nhabtants n Dakar, only one n Pkne, and only one health staton for each 23,000 people n Dakar and one for 19,000 n Pkne. The sewer network s nsuffcent, and many dstrcts n Dakar and all of Pkne have no sewer system. Equpment and budgets for ts upkeep reman too small to meet the needs of the populaton of a cty wth more than one and a half mllon people. ECONOMC ACTVTY To analyze employment, t s necessary to dstngush between the modern sector (such as jobs n busnesses wth accountng servces) and the nonstructured sector. Employment statstcs show that the modern sector, exclusvely developed n ctes, has undergone a growth of more than 6 percent durng the frst two decades of the ndependence of Senegal. But snce the eghtes t has declned. The work force has gone from 86,500 n 1971 to 173,000 n 1982, and 169,000 n 1986 (Bocquer, 1991). Government employment represents 40 percent of the modern sector. Durng. the seventes, the admnstraton work force ncreased, and the supervsng rato (number of government workers per 1,O00 habtants) went from 8 to 11. On the other hand, durng the 1980s. the supervsng rato decreased 4 percent (Bocquer, 1991). The decrease n the employment of the modem economy durng the 1980s had an mpact on the prvate sector whch, after an 11 percent growth, went through a slower evoluton of 2 percent. The prvate-sector share went from 18 percent n 1971 to 30 percent n Busness n the modem sector had a spectacular drop: ts share declned from 14 percent n the 1970s to less than 7 percent n the 1980s. ndustral producton, after a 14 percent growth n the 1970s. had a very slow growth rate of 3 percent n the 1980s. n 1974, t represented 26 percent of the modem sector, and n the 1990s. less than 20 percent (Valette, 1990). Ths decrease n the actvty of the modem sector s also felt n the captal where the publc and parapublc sectors account for 5 1,200 salared employees, ncludng 78 percent male, 67 percent n government jobs, and 29 percent workng for prvate busnesses. ts share n salared employment was 44 percent (Zarour, 1989; Bocquer, 1991). The prvate sector represents an mportant part of the modern sector (18 percent n 1971 and 30 percent n 1988). Employment n the prvate sector has jumped from 42,209 n 1960 to 107,164 n 1980, or a 5 percent yearly ncrease. ndustral producton, mostly concentrated n the captal, ncludes products derved from peanuts, phosphates, and fshng, as well as exports from Senegal. The harshness of clmatc condtons lmts peanut producton. The low productvty and compettveness of local ndustral products, the deteroraton of the exchange rate, the technologcal and energy-related producton costs represent as many road blocks aganst nvestments, whch are essentally foregn. Thus, the ndustry employs only 11 percent of salared workers. Ths percentage s hghly nsuffcent to support an urbanzaton that s qualtatvely suffcent. The rates of workers from the ndustral sector and the whole populaton reached ts lowest level n the 1980s (1.7 percent, compared wth 2.4 percent n prevous years). The modern sector represents less than 25 percent of the total employment n the great Dakar area, and s unable to meet the needs of new job seekers. Durng the last decade, the total number of actve workers ncreased 4 percent, whch corresponds to 98,723 new job seekers. The prvate sector has been able to absorb only 5,144 of them, or 5 percent, and the publc sector 17 percent. The remander, or 78 percent, must look to the nformal sector (Bocquer, 1991). Gven the low absorpton capacty of the modern, publc, and prvate sectors, the majorty s forced to work n the actvty sector called nonstructured. n producton actvtes (woodwork, sewng) or servces (mechancs, talors) or commercal actvtes, t s stll dffcult to determne exactly the number of jobs n the nonstructured sector, these usually beng temporary, n producton, qualfcaton, and ncome.

8 Selected Countres Accordng to the FAN-ORSTOM study, apprentces and famly helpers, unpad for the most part, represent 27 percent of all jobs, and 46 percent of the unstructured sector. Self-employed workers represent 45 percent of the unstructured sector. n the subdvsons of producton and servces, the ndependents, the apprentces, and the famly helpers represent 71 percent (90,300) of the men and 25 percent of the women (18,000). Buldng constructon belongs for the most part to the nonstructured sector, and the number workng s wdely underestmated. The FAN-ORSTOM study estmates the ndependents at 6,000, the apprentces at 1,800, and the salared employees n the modem sector at 3,200, a total number of 11,ooO. An mportant segment of the actve populaton n the unstructured sector s engaged n commercal actvtes. Also, there are more women than men there: 40,OOO compared to 31,000. Among employed women (109,600). only 15 percent are salared workers n a busness. Among the employed men, 39 percent work for a busness. Thus, women are not as well represented as men n the modern sector. Not all unemployed actve people n the modem sector succeed n fndng jobs n the nonstructured sector. The general populaton census of 1988 ndcated an unemployment rate of 27 percent for males and 31 percent for females. n Dakar, an employed person supports an average of 3.4 other persons. The hgh level of unemployment s explaned partcularly by massve layoffs and the closng of prvate busnesses. ndustry s the sector of actvty whch accounts for the largest number of unemployed 22 percent and 15 percent n the servces and producton sectors, and 19 percent of women n commercal actvtes. For men, the unemployment rate s approxmately 14 percent n the ndustry sector. Unemployment s worsened by structural adjustment programs (Douf, 1992). n order to reduce ts expendtures and lower the natonal debt, the state, whch s the man employer n the modem sector (40 percent of admnstratve jobs), s forced to lmt new recrutng and lay off government workers. Wth the use of custom and fscal measures, the revson of work laws, the abolton of restrctons on the mportaton of products, and the lberalzaton of prces, the government of Senegal s attemptng to rejuvenate ndustry and make t more compettve nternatonally. The lmts of the modem sector n ts ablty to employ the urban populaton, along wth lesser nvolvements from the state, have resulted n a progressve development of the sector of ndependent workers, famly helpers, and apprentces. These are usually rural mgrants and lad-off workers who have sought a new type of employment. The declne of the ndustral and admnstratve sectors has had a great mpact on temporary jobs generally held by rural mgrants wthout an educaton. The unstructured sector s dealng prmarly wth survval tactcs n whch all famly members are nvolved, partcularly women and chldren. The necessty of the partcpaton of all members n producton leaves unchanged the lfe cultural pattem of poor populatons n the ctes. Polygamy, fertlty, the reproducton Senegal 293 of socal and dentty allances are ntegral parts of the process of survval tactcs (Antone et al., 1992). The partcpaton of members n famly producton takes place along vared pathways: chldren and women head mostly for the unstructured sector. ndependent male workers and apprentces represent 71 percent n the subdvsons of producton and servces, and women 25 percent. The estmated number n ndependent busnesses s 3 1,300 men and 40,000 women. AN MPOVERSHED URBAN POPULATON There s lttle avalable nformaton on revenues n Dakar. The net ndustral ncome per person n Senegal has gone from cfa francs n 1960 to 143,000 cfa francs n 1985 (n 1985 cfa francs). n 1980, the average monthly famly ncome was estmated at 83,300 cfa francs n Dakar, and 51,200 cfa francs n Pkne (Mnsthre de l urbansme, 1986). The offcal mnmum wage n 1969 of 50.6 cfa francs an hour ncreased to 201.O6 cfa francs n 1988, whch s nsuffcent to offset nflaton (the prce ndex beng n 1989, wth a base of 100 n 1967); wth a constant currency value, the offcal 1988 mnmum wage declned 13 percent snce Some workers earn a salary below the offcal mnmum wage. n the FAN-ORSTOM study (1992), seven questons dealt wth the ownershp of domestc goods, ncludng rados and automobles. Despte these structural mperfectons, ths varable reflects a standard of lvng scale gong from O to 5. At both extremes, we fnd those who have no domestc goods, or at the very least a rado; such a stuaton s the norm for more than half of households (52 percent), and ndcates the poverty whch strkes an mportant proporton of households n Dakar. At the other extreme, 4 percent of households own almost all types of domestc goods. Between the two extremes are approxmately 23 percent of the households (category 1 and 2). whch own comparatvely lttle equpment. Those n category 3, average equpment (16 percent), and category 4, rather good equpment (5 percent), own a mnmum level of equpment whch ncludes n most cases rados, refrgerators, televsons, and lvng room fumture. Heads of households who have a satsfactory level of equpment belong for the most part to the categores of management level personnel (48 percent), mportant ndependent busnessmen (7 percent), manual workers (8 percent), and retred people (10 percent). The majorty n several actve categores have no goods. Thus, 83 percent of unsklled laborers, 67 percent of ndependent busnessmen, and 67 percent of the manual workers fall n ths stuaton. Even among the heads of households who are salared workers of servce ndustres, 45 percent have no domestc goods. Eghty-fve percent of the managementlevel personne1 have average or hgher qualty equpment. Except for management personnel, retred ndvduals enjoy a relatvely better stuaton than other professonals.

9 294 Selected Countres Senegal 295 % When equpment s consdered by type of household, one category stands out very clearly: households wth one or more nonrelated ndvduals, most of whom own no domestc goods. The percentage for men s 81, and for women, 76. These households take n other people (mononuclear households wth chldren and other parents), and polynuclear households own most of the goods. Does ths relatve affluence result from sharng common goods wth one another, or do the wealther heads of households assume the responsblty for a greater number of persons? One s nclned to choose the second hypothess, whch corroborates our prevous analyses. Usually, the households wth female heads are poorer than those headed by men, partcularly sngle-parent famles. Most people wth good equpment lve n hgh-standard dwellngs (81 percent) or n good-qualty dwellngs made of concrete (19 percent). The poorest resde ether n nadequate dwellngs or n concrete dwellngs of nferor qualty; also a large proporton (41 percent) of people n hgh-qualty concrete dwellngs are mpovershed. The poor lve n poor dwellngs, but a few households wthout many goods also lve n better dwellngs. Eghty-three percent of the heads of households wth good equpment are owners (83.3 percent). LMTS OF APPLED PROJECTS As a sgn of the rapd growth of ctes, the government has ncluded n ts offcal populaton plannng a natonal and regonal development program. The regonal ntegrated development projects am to centralze modem actvty (Mnst8re de l Econome et des Fnances 1988:lO). Among the stated objectves were the decentralzaton of ndustral and admnstratve nfrastructures to enhance the value of other ctes and decrease the mgratory flow to the captal. The progressve elaboraton of regonal poles of cultural, socal, and economc development wll enhance the producton actvtes, stablze the local populaton, and slow mgraton toward the Dakar area. The objectves also nclude the adopton of ncentve measures for the mplementaton of small and md-sze busnesses nsde the country, and new actvtes n secondary ctes to acheve a balanced dstrbuton of natonal manpower; and the restructuralzaton of urban equpment, of sewer systems and water networks n the spontaneous dwellng dstrct of Dakar and n the suburbs to mprove santary condtons. The mplementaton of these objectves s beng hampered by demographc constrants and problems lnked to the long- and md-term structural adjustment programs Programme d Ájustement B Long et Moyen Terme) (PALMT). The natalty rate remans hgh n Dakar, and ths shows that the redstrbuton polcy would not tself reduce the rapd populaton growth of the captal. Whle contnung ts sewer and hygene polces, the state must nstll new blood n ts polces concernng the natalty rate. The prorty nvestment and acton programs dealng wth the populaton, whch were elaborated n 1989, have a tendency to revtalze the famly plan! Í, 1 l! l programs, by ntegratng them wthn the health programs amed at mothers and chldren. The communcaton, educaton, and nformaton programs (nformaton, Educaton et Communcaton [EC]),concernng populatons are also fnanced to ncrease nformaton concernng famly plannng (Mnstkre de la Sante Publque. 1989). The assocaton between mortalty and natalty s not lnear. The decrease n the mortalty rate does not necessarly ental the adopton of famly plannng whch leads to a declne n natalty rates. Despte the repeal n 1980 of the 1920 law forbddng contracepton publcty and the use of contraceptves, and the ntegraton of programs and modem servces of famly plannng Planfcaton Famlale (PF) wthn the health servces for mothers and chldren Sante Maternelle et nfantle (SM). the use of contraceptves remans low n the country (EDS, 1986). Famly plannng programs are n fact nseparable from the economc condtons whch preval n the country. Among the most mpovershed socal groups, who operate accordng to survval tactcs n whch the partcpaton of large numbers of ndvduals s ndspensable to ncrease famly producton, the mprovement of Lvng condtons consttutes a prerequste to the acceptance of famly plannng. The structural adjustment program resulted n mportant negatve socal mpacts on the job market, whch are lnked to the polcy of deregulaton from the state, to new ndustral and agrcultural polces, and to a clean-up n publc fnances. Despte the creaton of the nserton and renserton delegaton Délégaton B l nserton et B la Rénserton (DRE), whose functon s to drect young and old workers removed from government jobs to other development sectors and regons, unemployment s at the most crtcal level of ts hstory: not only are the young wthout jobs, but ther parents occupy an unstable poston n the job market. Admnstratve measures are rarely sustaned by very concrete fnancal and economc programs: the dstance between laws and regulatons and facts stll remans very mportant. Ths means that polces are rarely translated nto concrete actons. The solutons to revtalze the economy are correctly dentfed, but the organzatonal and fnancal abltes necessary for ther mplementaton are absent. The payment of the natonal debt and the dran of captal consttute an obstacle to the accumulaton of captal on a natonal scale. CONCLUSON The employment crss manfests grave consequences: the modem sector s no longer a career prospect for young people, the unstructured sector has nsuffcent abltes to offer jobs to the unemployed, and Senegal offers few opportuntes of reconverson n agrculture, despte hopes generated by the development of the Senegal Rver valley. Wthn ths context of vanshng employment, the responsbltes of the heads of households ncrease, and are aggravated by low revenues and resources. The Y l

10 * P t 296 Selected Countres heads of households contrbute to the mantenance of the entre household by assumng responsblty for household expendtures. The crss enhances the dependence of young people on ther elders. Besdes the progressve densfcaton of dwellngs, the ncrease n the sze of households results especally n the delayed departure of youths from ther parental homes. The system of famly concessons whch s reproduced n the cty lessens the tensons of the real estate market. But how long wll one segment of the populaton be able to shelter another? Demographc perspectves do not foresee any mportant slowdown of urbanzaton, gven the natalty rate whch constantly remans hgh. The results from famly plannng programs reman medocre. n fact, reproducton s nseparable from the economc condtons of producton. The acceleraton of mgratory movements results n the rapd development of the unstructured sector, whch wll eventually reach ts saturaton pont. The 1990s began n Dakar wth a saturated job market because of a lack of job opportuntes and wth a real estate market poorly adapted to the requrements of the populaton. The nequaltes of access to resources are accentuated even wthn households. The orentatons of the Structural Adjustments Plan of Senegal, compressng the resources from the modem sector, do not herald a postve soluton to the crss. REFERENCES Antone, Phlppe, O. Barbary, Phlppe Bocquer, A. S. Fall, Y. M. Gusse, J. Nantelamo, and A. Dop L nserton Urbane: Le Cas de Dakar. Dakar: FAN- ORSTOM. Ba, A., and. Sam Mgraton et Urbansaton au Sén6gal. A Paraftre au CERPOD. Dakar. Bocquer, Phlppe Un Exemple d Analyse Statstque des Bographes: 1 Entree Dans la Ve Actve B Dakar. n Les Cahers Pratques Socales et Traval en Mleu Urban. No. 14, Department ORSTOM-SUD. Pars Les Mutatons du Marche du Traval a Dakar (SBn6gal) et 1 Access au Premer Emplo. Caher Vlles et Développment. Montreal: Groupe nterunverstare de Montreal. Bongaarts, J The Proxmate Determnants of Natural Martal Fertlty. New York Populaton Councl. CCEE (Casse Centrale de Cooperaton Economque) Evaluaton des Poltques et Programmes Urbans au Sénégal. Dakar. Champaud, J., J. Lombard, and M. Svgnon Vlles Secondares et Mveloppement Regonal au Shegal. Conventon. Dakar. Douf, M La Crse de Adjustement. Poltque Afrcane 45: Easlerln, R. A., and M. C. Crmmns An Exploratory Study of the Synthess Framework of Fertlty Determnants wth World Fertlty Survey Data. World Fcrtlty Survey Scentfc Reports, no. 40. Voorburg: The Netherlands. Locoh. T Structures Famlales et Changements Socaux. n Populaton et Senegal SocPt6s en Afrque au suo du Sahara, edted by D. T. Butn, Pars: L Harmattan. Manet, G Lu dynamque demographque des vlles Senegalases, Departement de gkographe (Multlgraphed), unverste de Dakar. Dakar. Mbodj, F. G nterpretaton des Resultats Pr6lmnares du Recensement Genbral de la Populaton et de l Habtat de 1988 au Sbndgal. Hstorens Géographes du Sén&al4-5: Mboup, G Recherche des Determnants Soco-Economques et Culturels au SBn6gal a Partr de ESF (1978) et EDS (1986). Ph.D. dss., Unversty of Montreal. Mnsere de L Econome et des Fnances, Drecton de la Statstque, Dvson des Enquetes et de la Demographe Enquete Senegalase de Fecondté, Dakar. Mnstbre de L Econome et des Fnances, Drecton de la Statstque, Dvson des Enquetes et de la Demographe, et DHS-nsttute for Resource Development/ Westnghouse Enquête Demographque et de Santé au Senegal, Dakar. Mnst8re de l urbansme, Drecton de la Statstque Enquête Urbansme au Senegal, Dakar. Mnstve de la Sante Publque, Mnstbre du Mveloppement Socal, OMS Centre Pour la Populaton et la Sante Famlale. Unverstd de Colomba. Dakar. Projet de Preventon de la Mortalt6 Maternelle au S6négal. Dakar. Republque du Senegal, Presdence de la Rbpublque-Secretarat Gkntral, Delegaton au Plan et aux Poltques Fxonomques, Drecton de la Prevson et de la Statstque Les Prncpaux Resultats provsores du Recensement de la Populaton et de l Habtat du SPnégal Dakar. Snou, A Dakar, Bulletn d nformatons Archtecturales, nsttut Franças d Archtecture. Supplément au Numero 141. Dakar. Valette, A Emplo et Nouvelle Poltque ndustrelle au SBn6gal. Les Cahers- Pratques et Socales et Traval en Mleu Urban, Numero 13. Pars: ORSTOM. Zarour. C Etude du Secteur nformel de Dakar et Ses Envrons Phase ll. Fnal Report. Senegal: USAD. 297