and Regional Change in , 1987–1998

Keith Sutton Department of Geography, The University of Manchester, Oxford Road Manchester M13 9PL United Kingdom Downloaded from http://meridian.allenpress.com/awg/article-pdf/4/3/206/1449372/arwg_4_3_a6t23106917322q6.pdf by guest on 30 September 2021 Recent demographic data for Algeria show dra- ceux de 1969 et de 1977, surtout pour les jeunes matic decline to a situation approaching femmes. Plusieurs raisons peuvent être avancées demographic transition. These low birth rates pour expliquer ce changement du taux de fécon- and rates would not have been dité. Les effets du recul de l’âge du mariage et predicted in the mid-1980s. Age-specific fertility l’utilisation croissante du planning familial ont data reveal the key to this change, as 1990s rates été démontrés. L’influence moins nette des révo- are much lower than 1969 and 1977 rates, espe- lutions urbaines, éducationnelles et sociales en cially for younger women. Several factors behind Algérie ont eu des impacts, tout comme l’exemple this fertility change are suggested. The rising age des émigrants algériens en Europe qui avaient of marriage and the increased practice of family auparavant montré des taux de fécondité moins planning are demonstrated. The less precise élevés. Un dernier facteur proposé concerne la influence of Algeria’s urban, educational, and crise économique algérienne et celles de l’emploi social revolutions are shown to have had an et du logement. L’année 1998 était celle du qua- impact, as has the example of Algerian emigrants trième recensement. Les aspects spatiaux de la in Europe who had lower fertility levels. A final croissance de la population sont analysés : les suggested factor concerns Algeria’s economic régions de l’intérieur et du Sud présentent une crisis and the related job shortages and housing croissance plus forte et constituent actuellement crisis. 1998 saw Algeria’s fourth population cen- des réservoirs démographiques. Les taux de sus. Spatial aspects of population growth are dis- natalité ont été cartographiés et permettent des cussed with interior and southern regions comparaisons. Les aspects régionaux des migra- showing stronger growth so that they now repre- tions, de la scolarisation des filles et de l’anal- sent demographic reservoirs. Recent birth rates phabétisme sont aussi brièvement examinés et are mapped and bear some comparison. Regional avancés comme des facteurs explicatifs des dif- aspects of migration and of female school atten- férences régionales de l’accroissement de la pop- dance and illiteracy are also briefly examined as ulation. possible explanations of regional patterns of pop- ulation growth. Mots-clés : Algérie, fécondité, transition démo- graphique, résultats du recensement de 1998, Keywords: Algeria, fertility, demographic transi- migrations tion, 1998 census results, migration

Des données démographiques récentes pour Regional Fertility Decline in the Maghreb l’Algérie montrent un déclin spectaculaire de la fécondité, présentant maintenant une situation de By 2000 Algeria’s population had reached 30 transition démographique. Les bas taux de natal- million, about 2.5 times what it had been at ité n’auraient pas pu être prédits vers le milieu the time of the first post-independence cen- des années 1980. Les données concernant la sus in 1966. Moreover, Algeria had achieved fécondité par âge révèlent ce changement : les this population size ahead of the neighbour- taux des années 1990 sont beaucoup plus bas que ing and sometimes rival state of .

The Arab World Geographer/Le Géographe du monde arabe Vol 4, No 3 (2001) 206-219 © 2001 by AWG The Arab World Geographer, Toronto, 207 Keith Sutton

However, both countries had experienced a annual population growth rates of just two remarkable decade of fertility decline during per thousand for Spain and and five per the 1990s that, arguably, amounted to a thousand for France (World Bank 2000) and demographic transition, or at least to a signif- the prediction of both an ageing and a declin- icant step towards such a transition (Sutton ing population for the European Union, the 1999). As recently as the mid-1980s fertility Maghreb’s recently lower but positive popu- changes in the Maghreb, along with other lation growth rates could again represent a Islamic Middle Eastern countries, had been potential source of in-migrant labour in the relatively moderate, so much so such that future. Downloaded from http://meridian.allenpress.com/awg/article-pdf/4/3/206/1449372/arwg_4_3_a6t23106917322q6.pdf by guest on 30 September 2021 Clarke (1985) could question whether demo- graphic transition was to be regarded as any- 1990s Demographic Transition in Algeria thing other than “limited.” Islamic religious-cultural factors were offered as With specific reference to Algeria’s recent explanations. Within a few years, however, demographic trends, Figure 1 illustrates the Courbage (1994; 1996a) and Fargues (1996) country’s dramatic fertility decline from were able to demonstrate a clear demograph- levels of natural increase in the ic transition. Table 1, based on Institut 1960s to a situation well on the way to demo- national des études démographiques (INED) graphic transition by the year 2000. Natural data that some would regard as conservative, increase rates halved from about 30 per thou- shows strong fertility decline across the four sand through to the 1980s to just below 15 Maghreb countries. Despite a sharp drop in per thousand by the century’s end. Birth rates mortality rates as well, a significant decline declined even more steeply from around 50 in natural increase rates resulted, especially per thousand in 1967 to 19.76 per thousand in in and Morocco. As will be demon- 2000. Low and declining death rates reflect strated later, Algeria’s Office national des both social progress and a youthful popula- statistiques has published figures which sug- tion. Figure 1 shows particularly marked fer- gest that Algeria’s fertility transition is much tility decline in the mid-to-late 1980s and closer to that of neighbouring countries than again in the mid-1990s. INED’s data indicate. A range of population and demographic Fertility decline in North Africa also indicators with Tunisian and Moroccan data means a revision in the perception of the for comparison (Table 2) illustrates Algeria’s Mediterranean as the dividing line in popula- progress in terms of fertility decline and tion terms between the stagnating population socio-economic development. Algeria’s fer- growth rates of Europe and the supposed tility rate, average life expectancy, literacy “galloping demographies” of the countries of rate, and level of are all well the South. Kliot (1997) stressed this North- ahead of what still prevails in most Third South contrast in the Mediterranean region’s World countries, suggesting that the country , albeit with an appreciation of should be classified as having an intermedi- recently converging fertility rates. George ate level of development. Population growth (1995) even wrote of “un véritable raz-de- rates and total fertility rates declined marked- marée démographique” (or tidal wave) with ly for all three Maghreb countries between reference to North African fertility. A drop in the 1980s and the late 1990s. Life expectancy Maghreb natural increase rates from 25–34 and rates improved, though per thousand in 1983 to 15–25 per thousand the latter showed scope for further improve- in 1999 (Table 1) ought to lessen concern ment. The reasons for lower fertility include about continuing rapid demographic growth contraceptive use, growing levels of urban- in Europe’s south-of-the-Mediterranean ization, and lower illiteracy levels, though neighbours. Indeed, with 1990–99 average again with much scope for improvement in

The Arab World Geographer/Le Géographe du monde arabe 4, no 3 (2001) Demographic Transition and Regional Population Change in Algeria 208 Downloaded from http://meridian.allenpress.com/awg/article-pdf/4/3/206/1449372/arwg_4_3_a6t23106917322q6.pdf by guest on 30 September 2021

FIGURE 1 Algeria—Demographic trends, 1967–2000. female literacy. The often better figures for of registered births, which numbered 845 381 Morocco and Tunisia undoubtedly reflect in 1985, dropping to 711 000 by 1995, with earlier attempts by the governments of those just 600 000 births in 2000 (ONS 1996; countries to control fertility, especially in the 2001a; 2001b). These successively smaller case of Tunisia, something of a family plan- cohort groups are now having a major impact ning pioneer in the Islamic and African on Algeria’s formerly broad-based popula- worlds. tion pyramid. It is important to stress that Algeria’s An alternative to the use of birth rates as fertility decline appears to have been contin- an index of fertility decline is examining total uing apace in the mid-to-late 1990s. Having fertility rates. Estimates and calculations of dropped markedly from 28.24 per thousand the fertility rate vary, but all authorities do in 1994 to 25.33 per thousand in 1995 and to register a marked decline from very high fer- 22.91 per thousand in 1996, Algeria’s birth tility levels in the recent past (Table 3). High rate continued to drop each year down to fertility indices of 8.36 in 1960 (Courbage 19.76 per thousand in 2000 (Table 3) (ONS 1999, 356), 7.9 in 1970 (Hemal and Haffad 2001a; 2001b). The resulting natural increase 1999, 65) and 7.0 in 1983 (Clarke 1985) had rate of 14.3 per thousand in 2000, or a 49- been halved to 3.57 by 1995 (Courbage year population doubling time, would not 1999) and even to 3.14 in 1996 (ONS 1998, have been predicted back in the mid-1980s, 34). These compare with 1990–95 fertility when a more typical Third rates of 1.78 for France and 1.26 for Italy growth rate of3%perannum and a doubling (GERM 1996). Courbage (1999, 353–56) time of just 23 years was giving much cause calculated that by 1995 Algeria’s fertility for concern. Another, perhaps more dramat- transition from its 1960 peak was 76.5 % ic, way of illustrating Algeria’s recent fertili- complete and population projections by the ty decline is to contrast the absolute numbers World Bank predicted that Algeria would

The Arab World Geographer/Le Géographe du monde arabe 4, no 3 (2001) 209 Keith Sutton

TABLE 1 Population and Demographic Data for the Maghreb Countries, 1983 and 1999. 1999 1983 Population Birth Death Natural Birth Death Natural (million) Rate Rate Increase Rate Rate Increase ‰‰ ‰ ‰ ‰ ‰ Algeria 30.8 30 6 24 46 14 32 Tunisia 9.5 22 7 15 35 10 25

Morocco 28.2 23 6 17 44 13 31 Downloaded from http://meridian.allenpress.com/awg/article-pdf/4/3/206/1449372/arwg_4_3_a6t23106917322q6.pdf by guest on 30 September 2021 Libya 5.0 28 3 25 47 13 34 Sources: Clarke 1985; INED 1999.

TABLE 2 Recent Population and Demographic Indicators for Algeria, Morocco, and Tunisia.

ALGERIA MOROCCO TUNISIA Population growth rate (% p.a. average) 1980–1990 2.9 2.2 2.4 1990–1999 2.2 1.8 1.6 Fertility rate (births per woman) 1980 6.7 5.4 5.2 1998 3.5 3.0 2.2 *Synthetic index of fertility, 1999 (Children/woman) 4.1 3.1 2.8 Life expectancy at birth —males 69 72 65 —females 69 70 74 *Infant 1999 ‰ 44 37 35 *Projected population in year 2025 (millions) 47 39 13 Urban population 1999 % of total 60 55 65 Adult illiteracy rate (% of population 15 years & Above) —male 24 46 21 —female 40 66 42 Contraceptive prevalence rate, 1990–98 (% of women, 15–49 years) 51 59 60 (Data for most recent year available) Sources: World Bank 2000, 274–87; INED. achieve transition to a replacement level fer- 2.5, while women who had never gone to tility rate of 2.11 by 2025–2030 (GERM school had one of 5.6. By the late 1990s 1996), as would Tunisia and Morocco. Algerian women were getting married later Algeria’s fertility rate varies markedly than their mothers (25.7 years old in 1994 by age, by rural-urban location, and by edu- compared with 18.3 years old in 1966) and cation. According to 1992 data, the urban fer- were having decidedly fewer children, as tility rate was 3.6 when the rural was 5.3. they were much more in control of their fer- Similarly, Algerian women who had had a tility (Hemal and Haffad 1999, 66–67). secondary education had a fertility index of The major evidence for Algeria’s fertility

The Arab World Geographer/Le Géographe du monde arabe 4, no 3 (2001) Demographic Transition and Regional Population Change in Algeria 210

TABLE 3 Recent decline in Algeria’s fertility rates. 1970 1977 1983 1990 1994 1996 2000 ‰ 49.0 46.0 46 30.9 28.2 22.9 19.8 Death Rate ‰ 17.0 14.0 14 6.0 6.6 6.4 5.5 Rate of Natural Increase ‰ 32 32 32 24.9 21.7 16.9 14.3 Fertility Index

(Children/Woman) 7.9 7.4 7.0 3.97 3.14 n.d. Downloaded from http://meridian.allenpress.com/awg/article-pdf/4/3/206/1449372/arwg_4_3_a6t23106917322q6.pdf by guest on 30 September 2021 Infant Mortality Rate ‰ 141.5 114.0 57.8 54.2 54.6 51.1 Total Population (millions) 13.1 16.95 23.04 28.5 30.38 (1987) Sources: ONS 1975; ONS 1979; Clarke 1985; ONS 1998; ONS 2001a; ONS 2001b.

TABLE 4 Fertility rates per thousand by age group of mother—Algeria

Year Age Groups 15–19 20–24 25–29 30–34 35–39 40–44 45–49 1969 112 324 356 329 263 148 37 1977 97.02 284.6 341.6 336.3 266.8 128.7 27.3 1983 47.83 241.62 325.53 296.57 230.08 104.85 20.39 1988 33 175 260 250 202 104 18 1992 26 151 208 214 174 81 18 1996 19 109 150 154 125 58 13 Sources: ONS 1994; 1995; 1996; 1998;Hemal and Haffad 1999. decline is revealed in Table 4 which charts Factors behind Algeria’s fertility decline changes in age-specific fertility between 1969 and 1996. Note that 1990s rates were much Several explanatory factors for this1980s and lower than 1969 and 1977 rates for all age 1990s fertility decline have already been men- groups of Algerian women, and especially for tioned briefly. The conventional factors such the age cohorts 15 to 19 and 20 to 24 years of as the higher age of marriage, the extension of age. Perhaps more than the birth rate and pop- , and the greater participation ulation growth rate data, these strong and per- of women in the workforce and in secondary sistent declines in age-specific fertility rates and higher education have contributed to lower through the 1980s and 1990s have convinced fertility. These will be reviewed before consid- observers like Courbage at INED that UN pro- ering two further factors suggested by jections have been too cautious and that demo- Courbage (1994; 1999): first, the role of graphic transition is now firmly established in Algerian emigration to Europe and the related Algeria, and indeed across the Maghreb. feedback of lower fertility practices to friends Indeed, Courbage (1997) revised downwards and relatives remaining in Algeria; and second, the UN 1992 projection of 51.8 million the role of economic factors, especially the Algerians by the year 2025 to 44.8 millions. post-1986 drop in oil and gas revenues and the associated economic uncertainties.

The Arab World Geographer/Le Géographe du monde arabe 4, no 3 (2001) 211 Keith Sutton

TABLE 5 Increase in Family Planning The rising age of marriage in Algeria (years) Furthermore, these Algerians marrying late Men Women are more aware of family planning practices 1966 23.8 18.3 and are employing them to reduce the num- 1977 25.3 20.9 ber of children they have or to space out 1984 27.4 22.1 births. The use of modern contraceptive prac- 1992 30.2 25.8 tice in the Maghreb was low outside of 1998 31.3 27.6 Tunisia until the 1980s, but by the mid-1990s Downloaded from http://meridian.allenpress.com/awg/article-pdf/4/3/206/1449372/arwg_4_3_a6t23106917322q6.pdf by guest on 30 September 2021 Sources: ONS 1998, 32; 1999, 9; Hemal over half of the region’s women were practis- and Haffad 1999, 67. ing family planning. In Algeria the rate of use of contraception was as low as 2% of women Rising Age of Marriage in 1962, rising to 8% in 1968, 25% in 1984, 35.3% in 1986, 50.7% in 1992, and 57% in Although marriage levels remain high in 1995 (Yaakub 1996; Farid 1996; Hemal and Algeria, the average age of marriage has risen Haffad 1999). This significant growth, which considerably for both men and women (Table was even higher amongst urban and educated 5). From an average age in the late teens for women, and its timing are linked to the even- women and early twenties for men in 1966, the tual introduction of an official family plan- average age at time of marriage has risen to ning program by the Algerian Government in 27.6 years for women and 31.3 years for men, 1983, several years after its Maghreb neigh- according to a 10 % sample from the 1998 cen- bours. The program focussed on centres d’e- sus. A 1992 enquiry revealed that the average spacement des naissances, which increased age of marriage of women in the 45–49 years in number from 745 in 1984 to 1 400 in 1986, cohort group was as low as 16.3 years com- 1 872 in 1987, and 1 955 in 1988. Ladjali pared with 21.9 years for the younger 25–29 (1985) showed that the late 1970s were sig- years cohort group. Further, this 1992 survey nificant years in the popularizing of family showed that only 3.4 % of women aged 15–19 planning through the expanding network of years were married, compared with 28.1 % centres d’espacement des naissances.She married for the 20–24 years cohort and still calculated that already by 1979 48 898 births only 60.9 % for the 25–29 years cohort. Only a year were being avoided. Arguably these 27.5 % of men were married before the age of were the family planning foundations on 30 years, according to this 1992 survey (ONS which Algeria’s 1990s fertility reductions 1998, 31; Courbage 1999, 360). This trend in were based. Before 1990 contraception was delayed marriage was noticed for the 1980s by mainly used by women in the 35–39 years Iles (1991, 104), who argued that “one princi- age group to keep family size down. By the pal reason is the chronic housing shortage.” 1990s, Hemal and Haffad (1999, 71) record- For women Iles has suggested that greater and ed 27% of the 15–19 years group and 42% of longer participation in education had also con- the 20–24 years cohort were utilizing contra- tributed. However, unlike the absolute number ception, mainly the pill (73.2%). Courbage of births, the absolute number of marriages (1995) noted that by 1992 family planning registered increased steadily through the had spread significantly to rural areas, with 1990s, from 149 345 in 1990 to 177 500 mar- 44% of rural married women and of illiterate riages in 2000 (ONS 2001a; 2001b). The women planning their families. Even abor- decline in fertility for the younger age groups tion was practised despite its interdiction by shown in age-specific data obviously results the Islamic Shariah. In 1990–92 for each 100 from these drops and delays in nuptiality. live births there were 10.5 abortions, with only a slightly higher rate in urban areas

The Arab World Geographer/Le Géographe du monde arabe 4, no 3 (2001) Demographic Transition and Regional Population Change in Algeria 212 compared with rural. towns was approaching the replacement Iles (1991, 114), using 1986 survey data, level. In Algeria the participation of women claimed that the east-west gap in family plan- in the labour force remained low in 1996 at ning practice was greater than the urban-rural 11.1% of the employed population (ONS gap. He stated that 40 % of women in 1998, 47). What probably contributed as Algeria’s western regions used modern con- much to fertility reduction was the shortage traceptive techniques, compared with 24 % of dwellings in Algeria’s , which served in eastern regions. to delay the age of marriage and then perhaps inhibit couples from starting families while Downloaded from http://meridian.allenpress.com/awg/article-pdf/4/3/206/1449372/arwg_4_3_a6t23106917322q6.pdf by guest on 30 September 2021 Urban, Educational, and Social still living with in-laws. Revolutions Example of Algerian Emigrants in Europe In arguing the case for demographic transi- tion in Algeria and elsewhere in the A couple of less conventional factors con- Maghreb, Courbage (1994, 53) considered tributing to fertility decline were suggested that three revolutions—urban, educational, by Courbage. These were the diffusion of and social s—accounted for the fertility pro-fertility-control attitudes and practices decline discussed so far. Fertility decline is from Algerian emigrants to Europe and the attributed to the increased presence of mid-1980s oil prices slump and associated women in towns, classrooms, factories, and economic decline. Of the first factor, offices. An urban place of residence usually Courbage wrote, means greater participation by women in education. By 1992 53% of Algerian women “the accelerated pace of demographic transition aged 15–49 years were deemed to be literate. in the Maghreb may be the unexpected result of Educated women marry later: in 1992 at 30.0 the Maghrebin presence in Europe” (1994, 60). years of age compared with 25.5 years of age for primary-educated women and only 23.7 He demonstrated that fertility amongst years of age for uneducated women (Hemal immigrant Algerian and other Maghrebin and Haffad 1999, 68). Educated women are women in France fairly quickly diverged from more aware of and inclined to use family that of their country of origin. By 1981 fertili- planning. By the 1995–96 school year 88% ty rates for Algerian women in France had of youngsters between six and 15 years of dropped to 4.35 compared with 6.39 in age were in education (ONS 1997). The role Algeria. By 1992 the fertility rate for Algerian of educating girls as a factor in fertility con- women in France had dropped further to 3.27 trol is strongly suggested by the results of a compared with 4.40 in Algeria (Courbage 1992 survey, which produced fertility rates 1996b; Hemal and Haffad 1999, 65). Can a of 5.6 and 2.5 respectively for never-educat- link be hypothesized between these demo- ed and secondary-school-educated women. graphic transitions amongst Algerians on both Even a primary school education lowered the shores of the Mediterranean? Within a genera- fertility rate to 3.3 (Hemal and Haffad 1999, tion a Maghreb immigrant society was created 67). in France whose values were no longer strictly The participation of women in the urban, congruent with those of its founders. though not rural, labour force means that Courbage (1994) stressed the substantial net- pregnancy and child-care have to compete works of exchange between Maghreb immi- with paid work and so pose an opportunity grants and residents still in the Maghreb and cost. Courbage (1994) provided some com- suggested three possible mechanisms that pelling data from Morocco that suggested could have influenced fertility in the Maghreb. that fertility amongst working women in First, money remittances change consumption

The Arab World Geographer/Le Géographe du monde arabe 4, no 3 (2001) 213 Keith Sutton Downloaded from http://meridian.allenpress.com/awg/article-pdf/4/3/206/1449372/arwg_4_3_a6t23106917322q6.pdf by guest on 30 September 2021

FIGURE 2 Algeria—Average Annual Population Change, 1987–98. patterns. A more consumerist culture in driving force of fertility transition” (Courbage Algeria may have heightened the desire to 1999, 361). In his wide-ranging study of Arab acquire further consumer goods, a desire world demographic trends, Courbage which may have competed for limited family observed that the years 1985–90 saw the onset resources with the desire for an additional of fertility transition in most Arab countries, child. Aspirations and values may then have including Algeria. He argued that these were changed, with family values becoming less largely countries previously awash with oil pro-natalist. Second, a holidaying migrant has revenues and that “high fertility is consolidat- significant influence at the local level. When ed when the individual is looked after from back in Algeria, such migrants may have cradle to grave” (Courbage 1999, 362), as was encouraged non-traditional behaviour, such as the case for Algeria’s hydrocarbons-based giving more formal education to girls, economy during the 1970s. After oil prices arguably the keystone for future demographic plummeted in 1986 and oil revenues were trends. Third, remittances to Algerian house- halved, the economic downturn soon had a holds and the associated desire for more con- demographic impact. “The opportunity costs sumer goods may have engendered of a birth became a factor in family strategies rural-urban movement, which in turn prompt- even amongst illiterate groups” (Courbage ed increased female participation in education 1999, 363). As more women had to seek to and in the formal labour force. enter the labour force, female employment became a more significant determinant of fer- Role of Algeria’s Economic Crisis tility. Within the space of a few years after the mid-1980s the Arab Middle East and Algeria A further argument is that Algeria’s emer- began to experience the demographic trends gence from an economy based primarily on experienced in Morocco from the mid-1970s income from hydrocarbon exports was “the on following the sudden reduction in income

The Arab World Geographer/Le Géographe du monde arabe 4, no 3 (2001) Demographic Transition and Regional Population Change in Algeria 214 from phosphate exports. rienced low population growth, some adja- Hemal and Haffad (1999, 68–69) also cent wilayate within commuting distance saw Algeria’s economic crisis as a factor for experienced more average growth rates, fertility decline, together with the related such as Tipaza (2.34%) and Boumerdes job shortages and the housing crisis. The (2.10%) near Algiers, El Tarf (2.27%) near negative social and economic effects of Annaba, and El Mila (2.57%) near Algeria’s structural adjustment program, Constantine. In summary, there was a north- imposed by the IMF, have arguably had a south cleavage between a more rapidly demographic impact even, and perhaps growing southern Algeria and a more slow- Downloaded from http://meridian.allenpress.com/awg/article-pdf/4/3/206/1449372/arwg_4_3_a6t23106917322q6.pdf by guest on 30 September 2021 especially, amongst the poor. Algeria’s ly growing northern Algeria. Within the transition to a market economy has appar- Tell (or northern) part of the country a fur- ently contributed to demographic transition. ther contrast was found between faster growing in the east and a less Spatial Aspects of Algeria’s Population dynamic western Algeria. Change, 1987-1998 These spatial disparities represent some interesting differences from the 1977–87 pop- Algeria’s fourth population census since ulation changes discussed and mapped in independence was carried out in June 1998. Sutton and Nacer (1990). High population It recorded a residential population of 29 growth continued in the Saharan and Saharan 276 767 persons, which represented a Atlas wilayate such as Tamanrasset, Illizi, 2.15% average annual increase over the Tindouf, Djelfa, El Oued, and El Bayadh 1987 census population. This rate of between 1987 and 1998. The wilaya of Bechar increase was considerably lower than that changed from above-average to below-aver- for the two previous inter-census periods— age population growth. Lower than average 3.06% per annum for the 1977–87 period population growth continued in 1987–98 in and 3.21% per annum for the 1966–77 peri- the northern coastal wilayate focussed on the od. When mapped (Figure 2) by wilayate main cities such as Algiers, Annaba, and (administrative districts) population growth Constantine, and in the group of western rates across Algeria show considerable spa- wilayate focussed on Oran and Tlemcen. This tial variation. High and above-average pop- coastal zone of low population growth spread ulation growth prevailed in most of Saharan to become more continuous, incorporating Algeria, such as Tamanrasset and Illizizi Saida, Relizane, Bouira, and Blida, for exam- wilayate, and in many interior wilayate in ple, wilayate that had shown greater popula- the Saharan Atlas and eastern High Plateaux tion dynamism between 1977 and 1987. In parts of the country, for example, in Djelfa contrast, the High Plateaux belt was no longer and Khenchela wilayate. Exceptions to this a solid zone of above-average growth. higher growth trend were the more urban- Compared with during the previous inter-cen- ized wilayate of Bechar and Ghardaia. sus period, wilayate such as Tiaret, Setif, and Below-average population growth prevailed Batna showed less strong population growth. in 1987–98 in the northern parts of Algeria However, certain eastern wilayate such as and especially in the five metropolis wilay- Mila and Oum-el Bouaghi moved into the ate of Oran, Algiers, Blida, Constantine, above-average population growth category. and Annaba. All of the western Oranais These spatial variations in population growth wilayate shared this low population growth. and especially the coast-interior or north- Exceptionally low growth of below 1.5% south contrast in mean occurred in the four northern wilayate of that the interior and southern wilayate Relizane, Tissemsilt, Medea, and Tizi remained as reservoirs of dynamic vitality Ouzou. While the main urban centres expe- during this 1990s decade of demographic tran-

The Arab World Geographer/Le Géographe du monde arabe 4, no 3 (2001) 215 Keith Sutton Downloaded from http://meridian.allenpress.com/awg/article-pdf/4/3/206/1449372/arwg_4_3_a6t23106917322q6.pdf by guest on 30 September 2021

FIGURE 3 Algeria—Births per thousand, 1998. Source: ONS 2001a. sition for Algeria as a whole. 9–17 per thousand in 1998. The fact that the interior and southern As was the case for population growth regions of the country were a demographic between 1987 and 1998, the map of 1998 reservoir is also apparent in Figure 3, which birth rates (Figure 3) demonstrates marked shows regional variations in 1998 crude birth spatial variation between lower birth rates in rates. The data for Figure 3 were calculated coastal and urban wilayate and higher birth from a comparison of 1998 data on registered rates in interior and especially Saharan and births (ONS 2001a) with the June 1998 cen- Saharan Atlas wilayate, which were almost sus results (ONS 1998b). Again it is instruc- consistently at the top end of the range. tive to compare Figure 3 with an equivalent Further north most of the High Plateaux map of 1987 birth rates published in Sutton wilayate also had above-average birth rates. and Nacer (1990). The keys to the two fig- The coastal wilayate along and north of the ures reflect the inter-census demographic Cheliff River, north of Tiaret, remained con- transition. In just eleven years birth rates sistent in having higher birth rates in both dropped considerably. Middle ranking wilay- 1987 and in 1998, for example, Relizane, ate now had birth rates of 19.0 per thousand Chlef, and Tissemsilt wilayate. Relatively to 21.55 per thousand, compared with 34–37 lower rankings by 1998 were achieved for per thousand in 1987. The upper range wilay- wilayate such as Ain Defla, Medea, Bouira, ate dropped from birth rates of 38.8–41.1 per and Jijel. Relatively higher rankings by 1998, thousand down to birth rates of 25.0–28.4 per however, typified wilayate such as Tindouf, thousand. The lowest ranked group of wilay- Saida, and Naama. More interesting are the ate declined more markedly from birth rates higher birth rates rankings found in 1998 for of 23.9–30.2 per thousand in 1987 to just the most urban wilayate based on the metrop-

The Arab World Geographer/Le Géographe du monde arabe 4, no 3 (2001) Demographic Transition and Regional Population Change in Algeria 216 olises of Algiers, Constantine, Annaba, and or below-average population change relative Oran. Indeed, there seems to have been to natural population growth. However, the something of a revival of demographic vitali- low population growth in the metropolitan ty in these major urban centres compared wilayate of Oran, Algiers, etc. between 1987 with 1987. However, this was not reflected and 1998 and earlier between 1977 and 1987 yet in the overall population increases for does not support the expectation of strong 1987–98. rural-urban migration. Some migration data Lower fertility is attributed to female have recently emerged from the 1998 census education and improved female literacy. (ONS 2001c). These seem imperfect in that Downloaded from http://meridian.allenpress.com/awg/article-pdf/4/3/206/1449372/arwg_4_3_a6t23106917322q6.pdf by guest on 30 September 2021 Sample data from the 1998 census can be they only include 1998 migrants who were used to investigate links between progress in recorded previously in the 1987 census. female education and regional variations in Younger migrants who were born after 1987 birth rates. However, there seems to be a lack but who then migrated with their parents do of correlation between birth rates and levels not seem to have been included. This limited of female school attendance and of female set of migration data suggests that the wilay- illiteracy. The group of six wilayate with the ate with high net in-migration between 1987 lowest birth rates had an average female and 1998 included Algiers and its adjacent school attendance rate of 82.5% and a female wilayate of Boumerdes, Tipaza, and Blida; illiteracy rate of 40.8%, rates similar to the Oran, and nearby Ain Temouchent; and the rates averaged across the ten largely southern southern wilayate of Ouargla and Djelfa. wilayate with the highest birth rates (school Regions of strong net out-migration included attendance rate of 78.2% and female illitera- Adrar, Chlef, Bouira, Jijel, Medea, cy rate of 42.6%) (ONS1999). This probably Tissemsilt, Ain Defla, and Relizane. As well resulted from the fact that many Saharan as the two main metropolises other cities wilayate were quite highly urbanized and that such as Sétif, Constantine, and Annaba also this led to high school attendance rates. had quite high figures for long-term in- Indeed, female school attendance is now migration (ONS 2001c, iv–v). As observed quite high everywhere in Algeria. Also the earlier, these migration patterns do not fit in main urban metropolitan wilayate had rela- well with the population change patterns, tively high birth rates, especially in the cases 1987–98, shown in Figure 2. of Annaba and Constantine, and this served It is necessary to emphasize how difficult to weaken any correlation. Some indication it is to explain regional patterns of population of rurality and the dominance of agricultural growth and regional birth rates. This is because employment may well correlate better with the anticipated explanatory factors such as birth rates, as rural women’s illiteracy tends female illiteracy and female education rates do to be higher and rural school attendance lev- not appear to correlate. Also urbanization, els lower. often regarded as a factor for lower fertility in So regional variations in fertility, as the past and in other countries, fails to fit com- expressed by 1998 birth rates, do offer a par- pletely. Wilayate dominated by cities, such as tial explanation for regional variations in Algiers and Oran wilayate, show higher birth population change between 1987 and 1998. rates than other northern wilayate. Another explanation might be spatial varia- Furthermore, Saharan wilayate with higher tions in migration, with the main urban cen- population growth and high birth rates are also tres expected to be poles of inward highly urbanized, adding further complexity to rural-urban migration. Until 1998 Algerian the situation. censuses consistently failed to generate What can be suggested from the popula- migration data, and inter-regional population tion-change and birth-rate data mapped in movement can only be inferred from above- Figures 2 and 3 is that interior and Saharan

The Arab World Geographer/Le Géographe du monde arabe 4, no 3 (2001) 217 Keith Sutton Downloaded from http://meridian.allenpress.com/awg/article-pdf/4/3/206/1449372/arwg_4_3_a6t23106917322q6.pdf by guest on 30 September 2021

FIGURE 4 Population pyramids for Algeria and for the wilayate of Tizi Ouzou, Annaba, Illizi, and El Bayadh. Source: ONS 1999.

Algeria contain regional pools of demographic signs that smaller cohorts are starting to narrow vitality and potential population growth. This the base of regional population pyramids. is ably demonstrated by another section of the These contrasting population pyramids 10% sample analysis of the 1998 census (ONS (Figure 4) allow the patterns of regional varia- 1999), which offers population structures or tion in population change (Figure 2) and in population pyramids for individual s.Atthe birth rates (Figure 3) to be interpreted as a late- national level a decade or more of successively 1990s snap-shot of the diffusion wave of suc- smaller birth cohorts has produced an increas- cessive stages of demographic transition ingly narrow base to the . working their way through the Algerian settle- Indeed, progress towards a less-Third-World, ment system. Northern and urban regions of “beehive” pyramid is well underway. As the country are now in the later stages, as is Figure 4 demonstrates, this is also the case for reflected in their beehive-shaped population several northern wilayate, such as Bouira, Tizi pyramids. More southerly, more rural, and Ouzou, Algiers, and Annaba. More dynamic more peripheral regions of Algeria are still in broad-based population pyramids prevail in an early stage of transition, as is evidenced by El-Bayadh, Illizi, Tindouf, Djelfa, and their higher birth rates and their more dynamic, Tamanrasset wilayate. These represent broad-based population pyramids. However, Algeria’s reservoirs of demographic vitality, the overall national situation is one of strong though further diffusion of demographic tran- demographic transition. sition to such regions can be expected, espe- One consequence of Algeria’s 1990s cially as most other wilayate already show demographic transition is that earlier projec-

The Arab World Geographer/Le Géographe du monde arabe 4, no 3 (2001) Demographic Transition and Regional Population Change in Algeria 218 tions for both national and wilaya population continues to be rapid across Algeria and ear- figures (Brahimi 1994) have now been shown lier expectations that dynamic population to have been too high. Three projections using growth would result in development prob- three different hypotheses were attempted for lems have had to be revised. each year from 1990 to 2020 and for each wilaya, using the 1987 population census data together with the fertility data then available. References For 1998 the three hypotheses projected Brahimi, R. 1994. Demographie: Projections de national populations of 29.7 million, 30.1 mil- population 1990–2020. Collections statis- Downloaded from http://meridian.allenpress.com/awg/article-pdf/4/3/206/1449372/arwg_4_3_a6t23106917322q6.pdf by guest on 30 September 2021 lion, and 30.4 million, all higher than the enu- tiques. no. 66. Algiers: ONS. merated population of 29.276 million Clarke, J. I. 1985. Islamic populations: Limited recorded in the 1998 census. If one considers demographic transition. Geography just the lowest projection, the ONS demogra- 70:118–28. phers underestimated the potential growth of Courbage, Y. 1994. Demographic transition Algiers wilaya, while they greatly overesti- among the Maghreb peoples of North Africa mated the growth of neighbouring Boumerdes and in the emigrant community abroad. In and Tipasa wilayate. They accurately predict- Europe and the Mediterranean,ed.P. ed the growth of Oran wilaya but overestimat- Ludlow, 47–88. London: Brassey’s. ed that of Constantine and Annaba wilayate. ———1995. Avant la tourmente: La situation Elsewhere, they underestimated the future démographique de l’Algérie en 1992. population in strongly growing wilayate such Population et sociétés. Bulletin de l’INED. as Tamanrasset, Djelfa, and Laghouat and no. 307:1–4. overestimated that in what proved to be more ———1996a. The end of the demographic explo- slowly growing wilayate such as Medea, sion in the Mediterranean. Population Relizane, Tizi Ouzou, and Tlemcen. This 8:258–70. shortfall in population growth predictions ———1996b. La population maghrébine à l’é- serves to demonstrate both the complexity and tranger: Dynamique démographique, carac- the contrary nature of Algeria’s regional popu- téristiques socio-économiques. In Séminaire lation geography as well as the difficulties of sur les migrations internationales 1996, planning for population change in a period of Ministère Chargé de la population, Centre demographic transition. d’etudes et de recherches démographiques, 100–108. Rabat: Royaume de Maroc. Conclusion ———1997. La démographie en rive sud de la Méditerranée au XXIe. siècle: Changement It can be argued that the patterns of popula- de perspectives. Espaces, Populations, tion change for 1987–98 represent a society Sociétés 1997(1):11–26. undergoing transformation. The evident ———1999. Economic and political issues of national trend of demographic transition has fertility transition in the Arab world: partly diffused from the urban and northern Answers and open questions. Population regions of the country to the more rural, and Environment 20:353–80. peripheral, and especially southern wilayate. Fargues, P. 1996. Un siècle de transition démo- Birth rates in 1998 and the contrasting popu- graphique en Afrique méditerranéenne, lation pyramids displayed in Figure 4 reflect 1885–1985. Population no. 41:205–32. the contrast alluded to above. What have Farid, S. 1996. Transitions in demographic and been interpreted as demographic reservoirs health patterns in the Arab region. In may well be already shrinking as fertility Proceedings of the Arab Regional transition extends into more peripheral Population Conference, Cairo 1996, Vol. 1, regions of the country. Demographic change 435–468. Liege/Cairo: IUSSP/CDC Press.

The Arab World Geographer/Le Géographe du monde arabe 4, no 3 (2001) 219 Keith Sutton

George, P. 1995. Le basculement démographique ———.1998a. Annuaire statistique de l’Algérie: des pays riverains de la Méditerranée occi- Résultats 1995/1996. no. 18. Algiers: ONS. dentale. Méditerranée 81(1–2):3–4. ———.1998b. 4eme. Recensement Général de la Groupement d’études et de Recherches sur la Population et de l’Habitat, 1998: Résultats Méditerranée (GERM). 1996. L’Annuaire de préliminaires. Données statistiques. no. 270. la Méditerranée. Edition 1996. Rabat: Algiers: ONS. GERM/PUBLISUD. ———.1999. IV Recensement Général de la Hemal, A. and Haffad, T. 1999. La transition de Population et de l’Habitat, 1998: Principaux la fecondite et politique de population en résultats (issus du sondage au 1/10). Downloaded from http://meridian.allenpress.com/awg/article-pdf/4/3/206/1449372/arwg_4_3_a6t23106917322q6.pdf by guest on 30 September 2021 Algérie. Revue Sciences Humaines Collections statistiques. No. 80. Algiers: (Université Mentouri, Constantine) ONS. 12:63–73. ———.2001a. World Wide Web Iles, A. 1991. The Algerian population. Between http://www.ons.dz/Demogr/Demog- archaic values and modernity. In Algeria: 1999.htm. Access date: September. The challenge of modernity, ed. A. Kenz, ———.2001b. World Wide Web 117–46. London: Codesria Book Series. http://www.ons.dz/Demogr/pop-titres.htm. INED. 1999. World Wide Web Assess date: September. http://www.ined.fr/population-en- ———.2001c. Les Migrations Internes Inter- chiffres/monde/tableaux1999/afrisept99.htm communales. Données statistiques. no. 331. . Paris: Institut National Etudes Algiers: ONS. Démographiques. Access date: September. Sari, D. 1997. Transition démographique et Kliot, N. 1997. Politics and society in the développement en Algérie. In Mediterranean Basin. In The Développement et transition démographique Mediterranean:. Environment and society, en Afrique. Actes du colloque international ed. R. King, L. Proudfoot, and B. Smith, de démographie. Série Colloques: 7ème. 108–15. London: Arnold. Vol. VIII, ed. K. Taamallah, 217–33. Tunis: Ladjali, M. 1985. L’espacement des naissances Faculté des Sciences Humaines et Sociales dans le Tiers-Monde: L’expérience algéri- de Tunis. enne. Algiers: Office des Publications Sutton, K. 1999. Demographic transition in the Universitaires. Maghreb. Geography 84(2):111–18. ONS (Office National des Statistiques). 1975. Sutton, K. and Nacer, M. 1990. Population Annuaire statistique de l’Algérie 1974. changes in Algeria, 1977–87. Geography Algiers: ONS. 75:335–47. ———.1979. Annuaire statistique de l’Algérie World Bank. 2000. World Development Report 1977–78. Algiers: ONS. 2000–2001. Oxford: Oxford UP. ———.1994. Annuaire statistique de l’Algérie Yaakub, A.-I. 1996. La fécondité dans le monde 1992. no. 16. Algiers: ONS. arabe: Tendances et facteurs d’évolution ———.1995. Démographie algérienne 1995. (1960–95). In Proceedings of the Arab Algiers: ONS. Regional Population Conference, Cairo ———.1996. Annuaire statistique de l’Algérie: 1996, Vol. 3, 164–185. Liège/Cairo: Résultats 1993/1994. no. 17. Algiers: ONS. IUSSP/CDC Press. ———.1997. L’Algérie en Quelques Chiffres 1997. Algiers: ONS. Submitted 10/01; revised 02/02; accepted 02/02

The Arab World Geographer/Le Géographe du monde arabe 4, no 3 (2001)