7
Internal Migration, Urbanization, and Population Distribution
John O.Oucho and William T.S.Gould
INTRODUCTION
In the three decades since the main period of independence in Africa, population distribution and redistribution through migration have remained important and widely recognized features of the population dynamics of the continent. Despite the continuing importance of the phenomenon, its status in the late 1980s and into the 1990s has largely remained as it was described by Prothero in 1968: the “Cinderella” of population studies. It is still not completely accepted as part of the inner family in demography (largely because of its “inferior” data and its variable and technically “soft” techniques); it is still starved for resources in comparison with the attention given to fertility and mortality data collection and analysis; and it still plays
John O.Oucho is an associate professor of demography and director of the Population Studies and Research Institute in Nairobi, Kenya. The chapter was written during his tenure as a demographer/instructor at the Regional Institue for Population Studies, University of Ghana, Legon, Ghana. William T.S.Gould is reader in geography, University of Liverpool, United Kingdom. The authors are grateful to Janet L.Ewing of the National Research Council library for her assistance in compiling substantive literature on the subject of this chapter; to Population Action International, formerly Population Crisis Committee, for providing some data on urbanization; to Mike S.Omogi and Cudjoie Dovlo of the Regional Institute for Population Studies, University of Ghana, for computing several tables and illustrative material; and to Claire Sullivan of the Department of Geography, University of Liverpool, for assistance in text processing.
a marginal role in national population policies. Progress in migration studies in Africa has been substantial in quantity, addressing a wide range of empirical, theoretical, and policy issues, and these have generated a large literature. However, they have not coalesced into any consensus on approaches or theoretical baselines. There have been contributions to major and long-standing theoretical debates, such as on the existence of a mobility transition to mirror the demographic transition (Zelinsky, 1971) and whether or not migration is a force for development at both source and destination (Gould, 1988; Oucho, 1990a), but the research agenda has moved away from questions associated with general global models to those arising explicitly out of the African experience. The systems framework of Mabogunje (1970) is perhaps the most widely cited model of this type.
Even in the area of models of African experience, however, there has been a general weakening of theoretical work in the face of a growing complexity of what is known about the migration experience throughout sub-Saharan Africa. To some extent it could be argued that migration studies lost their way in the 1980s, overshadowed by major developments in fertility and mortality studies. Yet migration remains important and needs to be considered not only in its own right, but also in the context of asserting its importance alongside fertility and mortality as a component of population dynamics. It has recently been argued, for example, that in Ghana, “migration may worsen the population pressure by undermining traditional demographic controls, and this supports and even increases high fertility rates” (Cleveland, 1991:238; see also Diop, 1985, on Senegal). Mabogunje (1990), on the other hand, implies that this loosening of demographic controls would lead to increased nucleation in family relationships that would, in turn, lead to reduced fertility. Much work remains to be done in this area.
This chapter describes the major characteristics, trends, and differentials, as well as the determinants of internal migration, urbanization, and population distribution, in sub-Saharan Africa by using available data and estimates for at least the last two decades (1970–1980 and 1980–1990) and projections for 1990–2000 and into the twenty-first century. The United Nations classification of sub-Saharan Africa into four subregions—eastern, middle, southern, and western—is used throughout. Because there are many countries and because data vary greatly in quantity and quality, it is inappropriate to give an exhaustive treatment to all countries in equal detail or to focus on all aspects of migration, urbanization, and population distribution. Rather, the discussion is directed to the three major features in separate sections. The first of these examines the typology and patterns of internal migration, migration differentials, and determinants of migration; the second reviews the magnitude, trends, and determinants of urbanization;
and the third discusses aspects of distribution, redistribution, and population density.
INTERNAL MIGRATION
Conceptual Issues and the State of the Art
Migration is the movement of people in space, often involving a change in the usual place of residence; internal migration is such a movement within national boundaries (International Union for the Scientific Study of Population, 1982:92–93). Because migration is a continuous, often repeated process rather than a single event, it is difficult to measure. Furthermore, because it is studied by researchers in all the social sciences, it lacks a standard data source or uniform approach.
The typology of African population mobility described in Table 7–1 differentiates the main types of movement in space (in a fourfold classification of rural and urban sources and destinations) and in time. The principal distinction made is between circulation (i.e., involving repetitive, nonpermanent moves—daily commuting and other short-term mobility have been excluded in this case) and definitive migration (Gould and Prothero, 1975). Circulation is subdivided in the table into three categories according to the length of the period of absence. Periodic movements are mostly short term. Seasonal movements, prominent in interior West Africa and among pastoralists, have a regular annual rhythm. Long-term circulation involves an absence of more than 1 year, but an expectation of return. Definitive migration, by contrast, is essentially a creation of the data collection methodology, when the individual migrant is recorded as being at a different place from one recorded at an earlier time (whether in a previous enumeration or as a result of some retrospective question, such as place of birth or place of previous residence). In practice it is very difficult to establish permanency, for the exact timing or direction of subsequent moves cannot be known—although probabilities of further movement may be estimated. Definitive migration may be further subdivided into irregular movements, where neither the timing nor the destination of the next move is known (characteristically in the case of refugees), and permanent movement, where the moves are considered by those involved to imply a permanent commitment to the new area of residence. This differentiation is not made in Table 7–1.
Many of these issues have been addressed in the rapidly growing literature on African migration—a literature that has been composed largely of empirical studies, often addressing only implicitly the larger theoretical questions on the causes and implications of the moves; many of these questions remain unresolved (Gould, 1992a). Important collections of empirical work from the early 1980s include the analysis of West African censuses of the
TABLE 7–1 Typology of Internal Migration with African Examples
1970 round (Zachariah and Condé, 1981) and two volumes derived from the work of the Commission on Population Geography of the International Geographical Union on population distribution and associated policies (Clarke and Kosinski, 1982; Clarke et al., 1985). In the mid-1980s, the United Nations Regional Institute for Population Studies in Ghana produced a wideranging collection (1987), and there were several formal and informal sessions on aspects of migration at the IUSSP African Population Conference in Dakar in 1988 (International Union for the Scientific Study of Population, 1988). In 1990, in association with the Nairobi Conference of the Union for African Population Studies (UAPS), three volumes including papers by many African scholars and a large bibliography, were published (Union for African Population Studies, 1990a,b,c). Table 7–2 summarizes the bibliographical material under various headings and provides a snapshot of areas of predominant interest and strength.
Data and Data Collection Methods
One critical technical issue that is immediately apparent in this literature concerns the long-standing problem of data for migration analysis, in the sense of both poor quality data and a reduced availability and use even of traditional sources. The recent literature offers a very distinct shift from systematic national analysis based on census sources to more specific and localized survey-based studies. In the African census rounds of the 1960s and 1970s, there was a great improvement in migration questions, with a shift from questions about ethnicity to questions about birthplace. Extensive use was made of these data in the form of place-of-birth/place-of-residence matrices (Masser and Gould, 1975). In addition, many African countries included a time-specific question in the 1980-round of censuses (e.g., Where were you living one year ago?). In theory, these changes and additions represented an important improvement in the quality of migration data, but in practice the results have been most disappointing and the resulting tabulations little used by analysts. The official report of the Kenya census of 1979 (Kenya, 1982:64), for example, concluded that
the data on place of residence in 1978 was bedevilled by the biases that are liable to afflict all questions involving dating and reference periods in Africa…. It cannot be recommended for inclusion in future censuses in Kenya.
Nevertheless, a 1-year retrospective question was included in the 1989 Kenya schedule, but it will probably also produce unusable data. It is too early to offer systematic consideration of the results of the 1990 census round, but it is unlikely to generate many new insights into migration.
Because census analysis based on place-of-birth data no longer adds
TABLE 7–2 Literature Sources on Migration in Africa by Theme
Theme |
Number of Studies Cited |
Major Issues Addressed, Papers Presented |
Data, methodology |
128 (8.5%) |
Data sources, scope, and limitations Conceptual issues and problems |
Labor migration |
202 (13.4%) |
Internal: rural-rural to wage sector and rural-urban migration International: direction, types (especially brain drain), and effects |
Resettlement/spatial distribution |
153 (10.2%) |
Postindependence resettlement process, types, results, and problems Demographic and socioeconomic challenges of resettled areas |
Urban system/ urbanization |
219 (14.5%) |
Migrants’ adjustment in urban milieu Relationship between migration and urbanization |
Linkages to migration system |
100 (6.6%) |
Networks and linkages of internal and international migrants with areas of origins Indices of linkages: visits, remittances, sociocultural ties |
Female migration |
37 (2.5%) |
Lack of studies on female migration and gender roles |
Refugees |
103 (6.8%) |
Sources and destinations of refugees Causes and consequences of refugeeism and displacements |
Nomadism |
25 (1.7%) |
Nomadism in Sahelian countries and Kenya Process, determinants, and consequences of nomadism |
Migration and regional integration |
34 (2.3%) |
International migration in the context of regional integration/cooperation institutions in African national subregional economies |
Migration and basic needs |
53 (3.5%) |
Migration and provision of food, shelter, education, health, etc. |
General interrelationships and policies |
453 (30.1%) |
General internal and international migration studies Migration-influencing and migration-responsive policies |
All themes |
1,507 |
|
SOURCE: Compiled from Union for African Population Studies (1990c). |
significantly to what is known of patterns and differentials in movement, this source has been rather neglected in the search for information that will address the most interesting policy and research issues (e.g., circulation and return migration, household migration strategies, women as independent or family migrants). These data are much more likely to come from surveys, which have mostly been small scale. In only two cases—the National Migration Survey in Botswana (1978–1979) and the National Retrospective Survey in Burkina Faso (1974–1975) —have there been major innovations on the national scale directed to improvements in migration data. The Botswana multiround survey was based on a 3 percent national sample and four rounds of survey within 18 months, so that it was able to identify seasonal mobility, rural-urban interactions, the mobility of individuals in the context of the household, etc. The three-volume report of the Botswana migration survey offers a glimpse into what is possible with more innovative data collection methodologies (Botswana, 1982). The Burkina Faso study used a retrospective approach to the collection of individual migration-history data (Piché, 1990). It was able, for example, to record subsequent migration probabilities of various subgroups in the population by number of previous moves. Whereas 77 percent of men who moved once had a second move recorded in the 5-year period, the proportion was only 32 percent for women (Piché, 1990:307). However, the experience of these two innovative cases has not been repeated in the 1980s in any other African country.
Thus, for most countries, there has been an increase in the quantity, but not the quality, of material available, and there has simultaneously been a widening range of approaches. However, these have not been accompanied by equivalent improvements in or agreement about the most appropriate techniques of analysis. In particular, migration model building, so favored by economists and others in the 1970s, is no longer common. None of the migration papers at the International Union for the Scientific Study of Population (1988) African Population Conference or in the Union for African Population Studies (1990a,b) papers is based on spatial models of migration, such as the gravity model, or econometric models derivative of the Todaro model that was so widely discussed a decade earlier. An exception in the 1980s, however, is Mazur (1984) on labor migration in Mali. In a more strictly demographic perspective, however, some further possibilities do exist for exploration of multiregional methods for migration estimates (Bah, 1990).
The following section explores major trends and differentials in sub-Saharan Africa. It first adopts an essentially spatial perspective in summarizing the geographical patterning of movement, emphasizing the mix of rural and urban sources and destinations, and then explores migration selectivity, in particular through sex, educational, and occupational differentials. Explanations for these patterns and differentials are sought at different spa-
rial scales, from the factors operating at a global scale to those operating at the household scale.
Spatial Patterns
Rural-Rural Migration
Given that most people continue to live in rural areas, and that there is in all countries continuous and complex movement within rural societies, even at subsistence levels of development, intrarural movements continue to be the most common of the four major directional types of movement high-lighted in Table 7–1. They are of many types and include movements of nomads as well as those of agriculturalists. They may be seasonal, as in movements between the dry savannas and better-watered areas, or more long term into the commercial rural sector. They may be permanent moves for agricultural colonization or into formal resettlement schemes.
Nomadism is a feature of Sahelian Africa in the Horn of Africa (including northern Kenya, northern mainland Tanzania, and northeastern Uganda) and in southwestern Africa (Botswana and Namibia). Recent studies of nomadic pastoralists have emphasized the increasing impetus on the part of governments for sedentarization, as in Sudan and in the Sahel in general. Sedentarization involves the permanent settlement of once seasonally mobile communities. It requires year-round provision of water and pasture for animals, and cultivation has been increasingly incorporated into these economies with implications for the sustainability of the environment as well as for mobility. The Turkana, a typical example of nomads in Kenya, have their lives hanging in the balance as a result of the encroachment by modern life styles and the vagaries of climate, as well as other factors (Odegi-Awuondo, 1990; see also Ayiemba, 1990). Such is also the case elsewhere in sub-Saharan Africa, for example, in Somalia (Maro, 1990), Niger (Wright, 1990), and Mauritania (Traoré, 1990). In each of these cases, government schemes have sought to mitigate the disastrous effects of drought and, in some cases, civil war by facilitating the restocking of herds and flocks in association with settlement projects that ensure permanent water and pasture. However, the number of animals has tended quickly to outstrip the carrying capacity of the local environment, however enhanced, and environmental deterioration has often been the result.
More generally, migration within rural areas involves farmers moving spontaneously in search of new land or in formally organized resettlement programs. The significance of spontaneous migration is probably falling as suitable land is increasingly in short supply. However, spontaneous migration is still important in the general drift southward in West Africa and in movements to marginal lands, to the dry margins as in Kenya (Dietz, 1986),
and to lower altitudes and therefore more disease-prone margins as in Ethiopia (Woldemeskel, 1989).
Much more widely discussed and much more obviously within the ambit of government policies are movements into government rural development schemes (Maro, 1990). Land transfer from the colonial to independent sub-Saharan African governments at or immediately after independence facilitated resettlement of the former “squatters” on the foreign-owned farms. Resettlement has also involved landless citizens and, in the case of Zimbabwe after 1980, combatants in the independence struggle. The Kenyan land settlement program in the highlands had largely ended by the 1980s (Leo, 1985), although it continued into the semiarid marginal lands. Resettlement in revolutionary Ethiopia (Wood, 1982, 1985), the ujamaa settlements in Tanzania (Thomas, 1982), and the “regrouping” of population in Botswana (Silitshena, 1982) are examples of politically inspired local redistribution and settlement. In the most recent experience in Zimbabwe, there remain deep-seated political and economic conflicts over the extent, type, and speed of resettlement. At independence in 1980, the government proposed a resettlement program of 18,000 households on former European-owned land. In 1981, this target was tripled to 54,000, and it increased again to 162,000 in 1982 with a completion goal of 1984. However, by 1989, only 52,000 households had been resettled, some 32 percent of the target, and most of these were in the poorer areas of the country and on individual farms rather than in cooperative schemes (Palmer, 1990).
Rural-Urban Migration
Although rural-urban migrants are not the largest group of internal migrants in sub-Saharan African countries, rural-urban movement, whether circulation and for a temporary sojourn in town or for permanent urban residence, is by far the most significant form of movement for the long-term trend of spatial redistribution, and as Table 7–2 suggests, it has attracted much study. To many governments, planners, and policymakers in sub-Saharan Africa, rural-urban migration is seen as the general case that all internal migration embodies. They have tended to overemphasize the importance of migration to the primate cities. Findings on urbanization as a migration phenomenon are discussed in detail below. Suffice it to say at this stage that the attraction of urban areas is largely, but not entirely, economic (Adepoju, 1990) and that rural-urban income and quality of life differentials remain large. The availability of jobs is critical, and rural-urban labor migration is dominant. However, the better availability of superior health care and educational opportunities (Gould, 1990), as well as housing (Ohadike and Teklu, 1990), can be additional attractions.
Urban-Rural and Interurban Migration
Two forms of migration are discussed together and rather briefly because of their relative unimportance, at least numerically, in sub-Saharan Africa. Urban-rural circulation consists of both periodic return migration synchronized with the peak agricultural seasons (notably weeding and harvesting) and labor migration to rural agro-industrial or mining complexes. Urban-rural migration is characterized by irregular “repatriation” not only of unemployed persons but also of criminals, and the relatively permanent return migration of both retirees and unsuccessful urban migrants. The Botswana National Migration Survey of 1978–1979 was able to show that 36 percent of all people surveyed in the four largest towns were rural-urban migrants; it was also able to show that 6 percent of those recorded had left town as urban-rural migrants (Case, 1982:117). Some urban-rural migration is already gathering momentum as sub-Saharan African governments continue to lower workers’ minimum retirement ages from 60 to 55 years and, in some cases, from 55 to 50 or even 45 years, and also as a result of retrenchment in public sector employment due to structural adjustment programs. Peil et al. (1988), for example, write of the Nigerian experience of workers “going home” after a career in urban employment.
Interurban movements are still minimal in sub-Saharan African countries, except in a large country such as Nigeria where they occur vertically within the urban hierarchy and horizontally among urban centers of the same size order. A survey of migrants in the mid-1970s to three of the largest towns in Nigeria—Benin, Kano, and Ibadan—recorded 21 percent coming from large cities (more than 100,000 inhabitants) and 18 percent from medium-sized cities (20,000–99,999). Another 18 percent of those recorded were intraurban migrants, from elsewhere in these three surveyed cities, but 43 percent were from rural areas or small towns of less than 20,000 (Lacey, 1985:697; see also Adepoju, 1983). The migrants include the self-employed, prospecting for profitable ventures and often moving frequently. Stepwise movement of migrants from smaller to larger urban centers includes public as well private sector workers transferred from one urban center to another.
Return Migration
Given their strong attachment to home areas and transient residence at the career destination, whether rural or urban, African workers tend to visit or migrate back to their homes after a period away.1 Return migration
remains seriously understudied in sub-Saharan Africa, yet it has important demographic and developmental implications in the migrants’ home areas. Persistence of return migration is an African phenomenon that continental literature surveys (Oucho, 1985b, 1990a) as well as some local studies, for example, in Zambia (Chilivumbo, 1985) and Kenya (Oucho, 1988), have confirmed. Oucho (1990a) reports that in all parts of sub-Saharan Africa, more than two-thirds of migrants visit home at least once a year. Related to the issue of return migration is the remittance that migrants send or bring back to their home of origin. These remittances help to ensure that the migrants will be accepted back into the home should they need or want to return at some point in the future. In southwest Nigeria about 60 percent of migrant heads of households in rural areas were remitting to their home areas at least once a year. In Kenya, in the early 1980s, more than 70 percent of urban households were remitting income at least once a year (Oucho, 1990a:121).
Migration Selectivity and Differentials
An important feature of any form of voluntary internal migration is the selectivity of migrants by demographic and socioeconomic characteristics from the general population. In this section, attention is directed to the most discussed differentials that are of particular interest to population analysts: age, sex, education, and occupational status.
Age Selectivity
Age differentials are well documented. Table 7–3 reports the peak age of internal migration based on census and survey data for 1964–1984 in selected countries. The majority of the studies identified in Table 7–3 indicate 20–24 as the modal age group. As a result of rapidly accelerating numbers of primary and secondary school graduates, for example, in Kenya (Gould, 1985), who are younger than their counterparts of two or three decades ago, the peak age at outmigration is probably falling, but migration still affects people throughout their economically active lives. The decline in the average age of migrants, however, has created a widespread desire on the part of governments to stem rural-urban migration of young, able-bodied, better-educated, and more development-conscious people, and to encourage them to be more active in development in the rural areas.
Sex Differentials
There was a very marked sex differential in both forced and voluntary labor recruitment during the colonial period. Males were recruited for ardu-
ous tasks in the agricultural wage sector as well as in the mines. They also occupied blue-collar and white-collar jobs in urban areas, but were often prohibited from bringing their wives or families. The use of sex ratios to distinguish between outmigration and inmigration areas is made possible by the continuing sex differentials in migration. Estimates of sex ratios from censuses and surveys reveal high sex ratios (i.e., more men than women) in destination areas and low sex ratios in areas of origin. With the major exception of Ethiopia, sex ratios are generally higher in urban areas than for nations as a whole, but that differential is being reduced in most cases (Table 7–4). Institutional constraints in rural areas (notably gender bias in allocation of land in resettlement schemes) and in urban areas (notably the use of educational qualifications for job selection) have kept women out of the migration system. The National Migration Survey of Botswana in 1978– 1979 showed that although equal numbers of males and females were recorded as intrarural (intervillage) movers in the multiround sample, many more males (69 percent) than females (29 percent) migrated to towns, mines, and commercial farms (Case, 1982:144). By the 1980s, however, conditions in rural and urban areas had changed, and more women were migrating, either with their husbands in family or associative migrations, or as individual women on their own account. One survey in Dakar, Senegal, for example, has disaggregated migrants by sex and cause of migration, and found that only one-third of women migrants were involved in household migration primarily, but nearly 30 percent of women were looking for formal employment (Badiane, 1990:249). The variability of national experiences, however, is again considerable. A survey in Nairobi, Kenya, at about the same time as the study in Senegal found that the majority of women migrants were involved in household-related moves (Omogi, 1992).
Zachariah and Condé (1981) concluded that for West Africa the sex composition of migration streams varied considerably in the 1960s and 1970s, but that the proportion of women in the migration streams was increasing in that period. This conclusion was likely to be as true for other major regions of the continent. The UAPS-commissioned overview of female migration in sub-Saharan Africa reaches five major conclusions: (1) female migrations are on the increase; (2) women migrants are motivated by much the same economic and sociopsychological reasons as male migrants; (3) women are more likely to migrate from areas where there has been a slackening of control on gender-ascribed roles; (4) women’s life cycles affect the nature of the migrations they are involved in; but (5) “in spite of economic gains to a significant proportion, women’s status is not much enhanced after migration” (Makinwa-Adebusoye, 1990:208). Overall, autonomous migration is now more important than associational migration of women.
TABLE 7–3 Peak Age of Internal Migrants in Selected Sub-Saharan African Countries
|
Peak Age Group of Migrants |
|
|||
Region and Country |
Year of Census/Survey |
Migration Type |
Author and Date of Publication |
Years |
Percentage of Total Migrants |
Western |
|
||||
Burkina Faso |
|
Rural-urban |
Coulibaly et al. (1980) |
20–24 |
40.0 |
Ghana |
1964–1965 |
Rural-urban |
Caldwell (1969) |
15–19 |
11.0 (M) 10.0 (F) |
|
1970a |
Rural-urban |
|
||
1970a |
Rural-urban |
Gaisie and de Graft-Johnson (1976) |
15–19 |
N.S. |
|
1972–1973 |
Internal |
Nabila (1979) |
20–24 |
34.5 (M) |
|
Liberia |
1974a |
Internal |
Campbell (1987) |
25–34 |
58.8 (M) 33.4 (F) |
Mali |
1978–1979 |
Internal |
Mazur (1984) |
20–24 |
35.2 |
|
Labor |
|
38.8 |
20–24 |
|
Nigeria |
1954–1965 |
Rural-urban |
Ejiogu (1968) |
15–24 |
34.2 (M) |
|
25–34 |
28.7 (F) |
|||
|
1976–1977 |
Rural-urban |
Makinwa (1981) |
20–29 |
47.5 |
1979 |
Rural out-migration |
Adepoju (1986) |
15–29 |
39.5 |
|
Eastern |
|
||||
Ethiopia |
1984a |
Rural-urban (to Addis Ababa) |
Ethiopia (1987) |
15–49 |
70.4 |
TABLE 7–4 Urban Sex Ratios, Sub-Saharan Africa, 1963–1985
|
Sex Ratio: Males per 1,000 Females |
|||
Region and Country |
Year |
National |
Urban |
Adjusted Urbana |
Western |
|
|||
Cameroon |
1976 |
960 |
1,077 |
1,122 |
Côte d’Ivoire |
1975 |
1,074 |
1,177 |
1,096 |
Ghana |
1960 |
1,022 |
1,062 |
1,039 |
|
1970 |
985 |
996 |
1,011 |
1984 |
973 |
949 |
976 |
|
Nigeria |
1963 |
1,020 |
1,149 |
1,127 |
Middle |
|
|||
Zaire |
1984 |
988 |
992 |
1,004 |
Eastern |
|
|||
Ethiopia |
1968 |
1,025 |
903 |
881 |
|
1984 |
994 |
867 |
872 |
Kenya |
1969 |
1,004 |
1,386 |
1,380 |
|
1979 |
985 |
1,216 |
1,234 |
Mozambique |
1980 |
945 |
1,097 |
1,161 |
Tanzania |
1967 |
955 |
1,180 |
1,236b |
|
1973 |
969 |
1,078 |
1,112 |
1978 |
962 |
1,075 |
1,117 |
|
Uganda |
1969 |
1,019 |
1,191 |
1,169 |
Zimbabwe |
1969 |
1,012 |
1,412 |
1,395 |
|
1982 |
960 |
1,140 |
1,188 |
Southern |
|
|||
South Africa |
1951 |
1,031 |
1,192 |
1,156 |
|
1960 |
1,010 |
1,150 |
1,138 |
1970 |
973 |
1,119 |
1,151 |
|
1980 |
1,035 |
1,068 |
1,032 |
|
1985 |
975 |
1,007 |
1,032c |
|
Northern |
|
|||
Sudan |
1973 |
1,023 |
1,131 |
1,105 |
|
1983 |
1,031 |
1,133 |
1,098 |
NOTE: Ratios are calculated from national censuses and estimates for the 1950s, 1960s, and 1970s compiled by the United Nations. aThe adjusted urban sex ratio is the urban figure divided by the national figure, and multiplied by 1,000; slight divergences are due to rounding. bTanganyika only. cExcluding Bophuthatswana, Ciskei, Transkei, and Venda. SOURCE: Gilbert and Gugler (1992:76,77). By permission of Oxford University Press. |
Educational Differentials
Many studies have established the positive relationship between migration and education (Gould, 1982). As early as the 1960s, Caldwell showed in Ghana that “what education does, more than anything else, is to promote long-term rural-urban migration” (Caldwell, 1969:62). In the colonial period the majority of the educated—almost invariably males—migrated to urban areas, and the uneducated gravitated toward areas of mining and agricultural wage employment. Lipton (1980:6) concluded that the movement was “educated to the big city, illiterate to rural areas.” Because fewer females than males in any age cohort had received formal education, their propensity to migrate was low, but over the last few years the picture has changed dramatically with rising female enrollment ratios in the majority of countries.
Generally migrants have higher educational attainment than nonmigrants. Even in Ghana where great strides were made in educational expansion, migrants recorded higher levels of school attainment than nonmigrants in the 1960 and 1970 censuses (Zachariah and Condé, 1981:70–71). This differential is consistent with rural areas having poorer prospects because they continue to lose the educated to urban areas. In Tanzania, it was found that “when educated workers are in surplus relative to the number of skilled jobs, urban expected income for the educated declines from the skilled to the unskilled wage” (Barnum and Sabot, 1976:37–38), but that because the unskilled wage still exceeds rural income, educated workers continue to migrate to urban areas. Increasingly, however, the educated are becoming involved in rural-rural migration streams to jobs that they had previously scorned, or else they are migrating into low-income or informal-sector activities in urban areas. School leavers may be prepared to take up any job in order to survive, as in Kenya in the tea and sugar industries (Odallo, 1985; Oucho 1985a).
Occupational Characteristics
Rural-urban migration studies in Africa and elsewhere have confirmed that relatively wealthy households are more able than the poorer ones to sponsor outmigration of some of their members. Thus, household income is positively related to outmigration, as it is to the expected income at destination. Skilled persons, or those with some education and potential for skill training, tend to prefer rural-urban migration, but the unskilled and poorly educated are found in rural-rural migration streams. In West and Central Africa, an increasing number of rural-urban and urban-urban flows consists of self-employed persons or of persons in formal employment with some informal sources of income.
Data from two West African countries illustrate occupational differentials of migrants. In Ghana, in 1970, substantial proportions of intraregional and interregional migrants of both sexes—63 and 44 percent, respectively, were employed in agriculture, followed by commerce as a distant second. In neighboring Togo, in 1970, 45 percent of intraregional male migrants and 42 percent of interregional male migrants were classified as production workers, transport equipment operators, and laborers; corresponding proportions for migrants in agro-industry including hunting and fishing, were 38 and 41 percent, respectively (Zachariah and Condé, 1981:76–77). Both cases reflect the importance of unskilled migrants as well as the self-employed, and the fact that migrants move into the informal as well as the formal sector (Lubell, 1991).
Determinants of Migration
Nearly all migration studies recognize that economic motives are necessary but not sufficient to explain population movements, and much of the recent work has sought to identify causation on different scales—from the macroscale, based on broad and generalized inferences from structuralist analyses, to the microscale, based on behavioral studies. However, no single study can claim an exhaustive inventory of all possible causes. This section identifies five major types of explanation, each of which has important current relevance to the policy context in which the components of population change are assessed.
External Factors
The global economic and political order increasingly dictates internal conditions and policies of African governments, whether indirectly through prices of exports, for example, or directly through structural adjustment policies. These, in turn, may affect patterns and types of internal movements. However, the process is not one in which crude economic determinism provides the outcome. One general aim of structural adjustment policies has been to narrow the income gap between rural and urban areas by offering greater incentives through higher prices to rural producers and by squeezing urban income earners. The normal migration expectation from such policies would be to reduce the rate of rural-urban movement. However, as Jamal and Weeks (1988) have amply demonstrated, rural-urban migration has not stopped, despite the narrowing of the income gap between rural and urban areas. It is clear that much more attention needs to be directed to these effects, particularly to supply-side rather than demand-side factors in rural-urban migration.
Government Policies and Programs
Administrative and structural arrangements have a direct and pervasive influence on migration. Development plans designed by individual governments often incorporate policies that explicitly address migration though, normally, consideration of migration is much less strongly emphasized than fertility or mortality. These include rural development programs and resource development programs for river or lake basins (e.g., the Niger River Authority, the Lake Victoria Development Authority in Kenya). These stimulate outmigration from some areas and inmigration to others (Maro, 1990). The experience of the majority of countries underscores the relationship between migration and development at national, regional, community, and household levels, because migration is both a cause and a consequence of development. Specifically, migration has been seen as both a function and a cause of uneven development in Zambia (Chilivumbo, 1985). Several papers in the UAPS volumes (Union for African Population Studies, 1990a,b,c) and most in the collection Population and Development Projects in Africa (Clarke et al., 1985) directly address the role of government in the migration process. The most obvious general conclusion of these studies is that governments have been able to exercise some control, which is most usefully directed to particular projects. However, governments generally have been able to exercise relatively little control, whether directly or indirectly—certainly less than they would have wished.
There are a number of countries in which government policies have caused migration very directly. A rice-growing effort in northern Cameroon has modified demographic behavior including migration patterns (Audibert, 1985). Similarly, the Cross River plantation projects in Nigeria have stimulated inmigration (Uyanga, 1985). In Ghana, north-south population movements to the cocoa and mining areas, as well as the coastal towns, and migration from the eastern part of the country to the same areas are stimulated and sustained by resource endowments and labor demand (Addo, 1987). Adepoju (1987) has shown how the Fourth National Development plan of Nigeria spelled out both migration-influencing and migration-responsive policies. Ejionye (1988) further suggests that Nigeria should guide migration and regional planning, adopt a meaningful industrial policy, and set up a national planned population movement commission or perhaps incorporate migration in a more comprehensive population policy. In Togo, migration is considered a rational response to collective needs related to systems of production, notably the need for diversifying cash income (Vignikin, 1986).
Community-Level Variables
Outmigration and inmigration are both affected by factors operating at the level of the community (e.g., transportation systems, community facili-
ties (mainly social), institutional factors, agriculture, researchers’ impressions of the economy, and modernization). One example comes from a study of migration from the Tshopo area to the city of Kisangani in Zaire, which found that push factors, such as the abuse of power by village elders, witchcraft, and the lack of infrastructure at the village of origin have a stronger influence than the city’s pull (Streiffeler and Mbaya, 1986). In Mazur’s (1984:248) village-level analysis of Mali, village characteristics supersede or interact with particular characteristics of individuals and households to determine labor allocation and migration.
Moreover, in sub-Saharan Africa, there exist strong sociocultural bonds and networks among migrants, as well as sustained linkages between migrants and stayers, that determine and sustain migration (Oucho, 1990a). The presence of relatives at a destination is sometimes more important than economic motives because prospective migrants must start from a base before being self-reliant. So, chain migration may be sustained within ethnic or kinship structures.
Household Decisions and Considerations
Some migration-influencing or migration-responsive policies are aimed at relocation of families (e.g., land settlement) and, accordingly, involve household decisions and all or part of a household moving in the process. In the case of rural-urban migration it has been recognized that the migration process may create the phenomenon of “one family, two households” (Weisner, 1972). This situation of a family geographically separated into two households as a result of migration is familiar throughout sub-Saharan Africa, and explains the compatibility of urban employment and job search with the rural linkages of migrants in towns (Oucho, 1985a, 1988, 1990a). These and other studies suggest that individual migration behavior may be due more to family or household obligations and considerations than to individual economic decisions. The Burkina Faso National Retrospective Survey in 1974–1975 showed a positive relationship between the proportion of households with absentees and the size of the household. For example, whereas slightly more than 15 percent of households that have one member residing at home have other members who are absent, more than one-third of households with five residents have absentee members, and more than half of the households with ten residents have absentee members (Piché, 1990). The dynamics of the household and the new home economics, currently major issues for fertility studies in sub-Saharan Africa, are also highly relevant to migration analysis.
Environmental and Resource Factors
The environment has a significant effect on population movements. The case of nomadic pastoralists and their seasonal or shorter-term mobility in search of pasture for their animals is the most obvious example, but where there has been sedentarization of pastoralists the environment needs to be managed to ensure survival. More generally, however, migrants move from hostile to less hostile environments or into environments recently freed of environmentally conditioned disease infestation (e.g., malaria; Prothero, 1965) or onchocerciasis, as in the current major program in the West African savanna. Labor migration may take place to extremely inhospitable areas, where settled agriculture is almost impossible, for the exploitation of mineral deposits (e.g., Mauritania or the Orapa diamond mine in Botswana). More typically, however, rural-rural resettlement has been into benign agricultural environments, such as the resettlement lands of Kenya and Zimbabwe or riverine areas in West Africa.
The environmental factor in migration has not been given much prominence in recent years, as is evident in the lack of concern shown in the Nairobi conference papers and migration literature (see Table 2). However, the African environment and its management in the face of global and more local environmental change have emerged as key factors in development, and there is often a presumption in the nonmigration literature that these factors will lead to major permanent migrations within, and even out of, the continent. It is clear from the discussion in this chapter that permanent migration is not the only response to change, and that various forms of circulation and other nonpermanent moves are the likely migration response to environmental change in the short and medium term. Mortimore’s (1988) study of farmers’ responses to drought in northern Nigeria illustrates the range of responses, including both circulation and permanent migration, that allow a population to be resilient in the face of environmental deterioration. His approach may be a model for a great deal of migration analysis on this theme that may be needed in the coming years.
URBANIZATION
Conceptual and Measurement Issues
In sub-Saharan Africa, rapid urbanization has preceded industrialization; indeed, the African experience seems to imply that it is completely independent of it. Growth of the urban population can be looked at in two ways: on its own, in which it is described as urban growth, and as a proportion of the national population, in which the term urbanization is used. Urban growth, strictly speaking, refers to the growth rate of urban
areas themselves (annual net additions to urban population divided by the size of the urban population), and urbanization is the process of growth in the proportion of the national population living in urban areas (Preston, 1982:650).
The urban population is typically measured on the basis of “localities,” which are spatially defined population conglomerations. Localities are defined as urban or rural generally on the basis of a size criterion, though other criteria such as the nature of the locality are sometimes used. Changes in definitions can lead to large changes in urban population, without any corresponding difference in residence or function.
Cross-national comparisons are difficult because the population threshold of urban localities differs from country to country such that even simple comparison of urban populations can be problematic. In sub-Saharan Africa the urban population threshold ranges from settlements of 20,000 to as few as 500 inhabitants, as in South Africa and Zimbabwe, which have townships of 500 or more inhabitants. In Kenya and Zaire, an urban area is a settlement with 2,000 or more inhabitants; in Ghana, the threshold is 5,000 inhabitants (Ajaegbu, 1979:87). According to the United Nations classification, any settlement with a population of 2,000 or more is considered urban, but the United Nations Economic Commission for Africa (1975) recommends the following classification of urban localities, which is used in the discussion below: rural locality, less than 20,000 and therefore including many small towns; urban locality, 20,000–99,999; city, 100,000– 499,999; big city, 500,000 or more.
Urbanization Levels and Trends
Sub-Saharan Africa is characterized both by a small proportion of its population in urban areas and, at the same time, by rapid urban growth.
Urban Population Size
In 1970, 32 million people in sub-Saharan Africa lived in urban areas with populations of more than 20,000:32.5 percent in urban localities, 42.3 percent in cities, and 25.2 percent in big cities (Ajaegbu, 1979:88). Thus, about three-quarters of the urban population lived in urban centers of 20,000 to 500,000 people. Of the 339 urban centers then recording populations of 20,000 or more, 184, or 54.3 percent, were in the size class of 20,000– 50,000, mainly in western Africa. Only two, Johannesburg in South Africa and Kinshasa in Zaire, had assumed the status of cities with populations of more than a million. The number of cities with more than 500,000 inhabitants rapidly increased from 3 in 1960 to 28 by 1980. By 1990, only four sub-Saharan African urban agglomerations—Johannesburg (4.6 million), Lagos
(4.0 million), Kinshasa (3.2 million), and Cape Town (2.4 million) —ranked among the 100 largest metropolitan areas of the world (Population Crisis Committee, 1990). The least urbanized areas of sub-Saharan Africa include the Sahel zone from Senegal to the Horn of Africa and south from there through the East African highlands and Zambia into Namibia (Escallier, 1988).
Though the total urban population in all four regions has increased substantially (see Figure 7–1), the average growth rates have not; in fact, according to United Nations estimates and projections, growth rates throughout Africa are expected to decrease (see Figure 7–2).
Urban primacy, the dominance of a single city, generally the national capital, as the main administrative, economic, and financial center, is the rule rather than the exception (Adepoju, 1988:126). Overall, primate cities in sub-Saharan Africa account for 30–40 percent of the national urban populations (Hill, 1990). In some countries, however, there is a dual dominance—Yaoundé and Douala in Cameroon, Brazzaville and Pointe Noire in Congo, and Harare and Bulawayo in Zimbabwe. Nigeria is an outstanding exception because there are several large cities, with Lagos, Kano, and Ibadan all having more than 1.5 million people, and Port Harcourt, Kaduna,
and Ilorin about 1.0 million in 1990. In other cases, though, the dominance of the primate capital is not only large, but seems to be increasing. In Table 7–5 the size of the four largest cities at a range of dates is given for five countries, with indices of primacy.2 Nigeria is clearly the most diversified and has stable indices of primacy over time. Ethiopia, Mali, and Senegal, by contrast, all have capital cities that are at least three times the size of the second largest city, and the indices of primacy have increased over time. Zambia has a lower level of primacy, but the primacy indices are rising (Antoine and Savané, 1990).
Many countries have also experienced the growth of smaller centers, often as a result of administrative decentralization, as in the state capitals in Nigeria and, to a lesser extent, the regional capitals in Tanzania. In Kenya, concern for the structure of the urban hierarchy and the need to develop a mature pattern of lower-order centers have been central to national development plans since the mid-1970s. Although Nairobi continues to dominate, the number of smaller centers and the proportion of the urban population
living in them have risen consistently. Nairobi contained 43, 40, and 47 percent of the urban population recorded in the 1948, 1962, and 1969 censuses, respectively, but its proportion fell to 36 percent in 1979, and urban areas with fewer than 10,000 people had 12 percent of the national urban population by that date; see Table 7–6 (Obudho, 1990).
Rate of Urban Growth
There is a diversity of regional experience of rates of urban growth in the period since independence (Figure 7–2). The overall rate of urban population growth rose slightly, from 4.6 percent a year for 1965–1970 to 5.0 percent a year for 1985–1990, and it is expected to remain at about that level or slightly less to the end of the century. Projections of declining growth rates of urban populations in western and eastern Africa defy the popular notion of expected accelerated rates about which national governments often warn their citizens, and which the international media tend to assume.
Proportion of Urban Population
Trends in the urban proportion show striking subregional differentials, as can be seen in Figure 7–3. Between 1965 and 1985, the proportion of the total population living in urban areas in southern Africa rose from 43 to more than 50 percent; in middle Africa, from 21 to almost 35 percent; and in western Africa, from 17 to 29 percent. The least urbanized subregion, eastern Africa, has experienced increased urbanization, though the urban proportion has stayed less than 20 percent, rising from less than 10 percent in 1965. By the year 2000, the projected proportions are expected to exceed one-third of the national populations in all regions except eastern Africa. Four eastern African countries (and all but three in middle Africa), Namibia and South Africa (whose urban populations had already exceeded one-half by 1985), and four western African countries are expected to be 50 percent urban by the turn of the century. However, the extent to which the proportion of the population in urban areas can be expected to rise is likely to be highly variable, dependent on local cultural and land use considerations that affect the persistence of circulation of urban migrants.
Components of Urban Population Growth
Demographic Components
Although rural-urban migration has generally been identified by national governments and scholars of migration as a principal cause of urban-
TABLE 7–5 Population (in thousands) of the Four Largest Cities and Urban Rank Indices, Selected Countries
|
Nigeria |
Zambia |
||||
1963 |
1972 |
1984 |
1963 |
1969 |
1980 |
|
Population |
|
|||||
First city |
665 |
1,569 |
4,486 |
123 |
262 |
536 |
Second city |
627 |
1,479 |
4,230 |
123 |
200 |
264 |
Third city |
295 |
578 |
1,654 |
93 |
160 |
250 |
Fourth city |
209 |
409 |
1,157 |
81 |
107 |
139 |
Rank Indices |
|
|||||
First and second cities |
1.06 |
1.06 |
1.06 |
1.00 |
1.31 |
2.03 |
First through fourth cities |
0.59 |
0.64 |
0.64 |
0.41 |
0.56 |
0.82 |
SOURCE: Antoine and Savané (1990:65). |
ization in sub-Saharan Africa, it is by no means the only one. Migration combines with natural increase in urban areas, as well as administrative reclassification of former rural territories into urban territories.3 These three components of urbanization, the first two demographic and the third nondemographic, may be influenced by urban and industrialization policies and by national development policies (Makannah, 1990).
In the first postindependence decade (1960–1970), a period during which national governments relaxed the restrictive rural-urban migration laws of the colonial period, 12 of the 16 cities of more than 100,000 grew mainly as a result of inmigration. Migration was particularly significant to Conakry, Freetown, and Lagos in western Africa, and to Kinshasa and Luanda in middle Africa. The greater contribution of migration persisted into the decade, 1980–1990, but is projected to decrease in the last decade of the century in a majority of cities. Makannah (1990:16) has estimated for 14 cities that between 1960–1970 and 1970–1985 there were declining trends in
Ethiopia |
Mali |
Senegal |
|||||
1967 |
1984 |
1960 |
1976 |
1987 |
1950 |
1976 |
1988 |
644 |
1,413 |
128 |
419 |
646 |
375 |
813 |
1,309 |
179 |
275 |
28 |
65 |
89 |
70 |
115 |
175 |
51 |
98 |
28 |
53 |
74 |
70 |
104 |
152 |
43 |
76 |
20 |
49 |
73 |
50 |
92 |
137 |
3.60 |
5.14 |
4.57 |
6.45 |
7.26 |
5.36 |
7.07 |
7.48 |
2.36 |
3.15 |
1.68 |
2.51 |
2.73 |
1.97 |
1.61 |
2.82 |
the contribution of migration to their growth (Figure 7–4). These data show considerable consistency in the relationship between the declining overall rate of growth and the declining contribution of migration to that growth. In all subregions the contribution of migration to urban growth has been declining, despite the fact that natural growth rates in towns have been lower than in rural areas.
POPULATION DISTRIBUTION AND DENSITY
Conceptual and Measurement Issues
Both migration and urbanization involve the redistribution of population. There are parts of sub-Saharan Africa that contain large numbers of population and others that are virtually uninhabited, but as a result of the mobility described in previous sections there is continuous redistribution. Yet the distribution at least of the rural population remains seriously constrained by environmental considerations and carrying capacities at prevailing levels of agricultural technology and organization. The concept of carrying capacity is a controversial one, as are implications of population pressure (Mabogunje, 1990), but both of these relate to the ratio of popula-
TABLE 7–6 Urban Centers by Size of Urban Population: Kenya, 1948, 1969, and 1979
tion to land resources, and therefore to the population distribution and density. Both crude and more refined indices of population density need to be considered. Crude or arithmetic population density (population per unit of area) in subnational units often reported in censuses needs to be supplemented by physiological (population per area of arable land) or nutritional (potential calorie production per unit of land) density. Hance (1972) cautioned about the “crudeness of crude density” approach, which on a continental scale gives the impression of an empty continent yearning for a bigger population. It is only on a smaller scale, however, that the effect of population densities and issues arising out of the relationships between rising population densities and environmental sustainability can be usefully approached.
Population Density
Table 7–7 presents national estimates of crude population density for the majority of sub-Saharan Africa in 1975 and 1985. These are national averages and therefore need to be treated with some caution because they obscure the local variations that in most countries are substantial. In a
country such as Kenya, for example, 90 percent of the population lives on only 20 percent of the land area, and locally the population density is a function of both economic and environmental conditions (Bernard, 1982; Gould, 1992b). The physiological densities, measured by population per square kilometer of arable land, are also shown on Table 7–7;4 these take environmental conditions into account.
It should be noted at this point that increases in population density on the national scale are due primarily to population growth rather than to redistribution. Even in countries with very considerable population move-
ment, as described in this chapter, rising densities occur in most rural areas, even those with net outmigration, because the extent of outmigration seldom exceeds the rate of natural growth. Only in a few areas (e.g., parts of the Niger Delta in Nigeria, coastal Sierra Leone) have absolute population declines been recorded in areas of settled agriculture (Gould, 1992a:303).
As can be seen from Table 7–7, the population densities of countries in sub-Saharan Africa vary widely. In 1985, they ranged from 2 people per square kilometer in Botswana and Mauritania to 232 people per square kilometer in Rwanda. A cursory look at the data indicates that, on average, population densities tend to be the lowest in middle Africa, where they range from 4 to 21.
Between 1975 and 1985, the crude population density increased in every country for which data are presented. These data indicate that the percentage increase was greatest in Somalia, Gabon, Botswana, and Mauritania, where it increased by 100 percent. The smallest percentage increase took place in Madagascar, 21.4 percent.
As mentioned above, these crude measures conceal local variations in density that are related, in part, to environmental conditions. Physiological density provides one way to control for environmental differences. In 1985, the physiological densities ranged from 79 in Botswana to 1,675 in Liberia. Again, by using this criterion, middle Africa is somewhat less dense than the rest of Africa, though there is considerable overlap. Like crude densities, most physiological densities increased between 1975 and 1985. The largest percentage increases were in Mali, Mauritania, and Niger, whereas there were actually declines in physiological density in Angola, Gabon, Botswana, and Senegal, probably due, for the most part, to a reestimation of the amount of land that is arable.
Population Redistribution
As cited in the earlier discussion of migration, a large volume of literature on population redistribution in sub-Saharan African countries has appeared during the postindependence era. Overall settlement concentration, through urbanization as well as villagization programs in rural areas of Ethiopia, Tanzania, Mozambique, Zimbabwe, and Botswana, has been characteristic on the national scale. The literature addresses diverse forms of redistribution that have taken place, including agricultural resettlement that characterized these countries in the process of land transfers from foreign settlers to the indigenous people, as well as redistribution of population from areas of population pressure to either the newly acquired lands or other less densely settled parts.
Population redistribution in sub-Saharan Africa may be classified into five main types: (1) resource- or development-induced redistribution; (2)
TABLE 7–7 Arithmetic and Physiological Population Densities per Square Kilometer in Sub-Saharan Africa by Subregion and Country, 1975 and 1985
|
Crude |
Physiological |
Change, 1975–1985 (%) |
|||
Region and Country |
1975 |
1985 |
1975 |
1985 |
Crude |
Physiological |
Western |
|
|||||
Benin |
28 |
36 |
106 |
291 |
28.6 |
174.5 |
Burkina Faso |
22 |
29 |
108 |
301 |
31.8 |
178.7 |
Côte d’Ivoire |
21 |
32 |
83 |
355 |
52.4 |
327.7 |
Ghana |
41 |
54 |
940 |
1,146 |
31.7 |
21.9 |
Guinea |
18 |
25 |
108 |
405 |
38.9 |
275.0 |
Guinea-Bissau |
15 |
25 |
208 |
307 |
66.7 |
47.6 |
Liberia |
15 |
20 |
1,221 |
1,675 |
33.3 |
37.2 |
Mali |
5 |
7 |
49 |
390 |
40.0 |
695.9 |
Mauritania |
1 |
2 |
132 |
929 |
100.0 |
603.8 |
Niger |
4 |
5 |
31 |
164 |
25.0 |
429.0 |
Nigeria |
68 |
103 |
276 |
334 |
51.5 |
21.0 |
Senegal |
21 |
33 |
173 |
123 |
57.1 |
–28.9 |
Sierra Leone |
41 |
51 |
74 |
223 |
24.4 |
201.4 |
Togo |
40 |
52 |
100 |
218 |
30.0 |
118.0 |
Middle |
|
|||||
Angola |
5 |
7 |
496 |
297 |
40.0 |
–40.1 |
Cameroon |
13 |
21 |
95 |
167 |
61.5 |
75.8 |
Central African Republic |
3 |
4 |
31 |
136 |
33.3 |
338.7 |
Chad |
3 |
4 |
58 |
159 |
33.3 |
174.1 |
Congo |
4 |
5 |
211 |
264 |
25.0 |
25.1 |
Gabon |
2 |
4 |
421 |
340 |
100.0 |
–19.2 |
Zaire |
11 |
13 |
342 |
512 |
18.2 |
49.7 |
government-induced population transfers and resettlement; (3) spatial redistribution due to migration and urbanization; (4) relocation resulting from environmental hazards and catastrophes; and (5) spontaneous settlement/ resettlement (Clarke and Kosinski, 1982; Clarke et al., 1985; Maro, 1990). The first, third, and fifth categories are found in almost all sub-Saharan African countries, which suggests the importance of resource and development endowments in attracting large numbers of redistributed populations. Government policies based on bargains between the departed colonial powers and the incoming independent government, as in Kenya and Zimbabwe, triggered redistribution not only of the landless population but also of progressive farmers lured by commercial agriculture. In Ethiopia and Tanzania, socialist political ideology resulted in large-scale population resettlement and villagization in the 1970s and 1980s. These political policies have been relaxed only recently in light of new realities in these countries’ political economies. It should be cautioned, however, that the five types of population redistribution are not mutually exclusive, and most if not all can be found in any one country. Combinations of the five exist, triggered by varying factors and having different effects on national populations and in different regions.
In the longer term, though, into the twenty-first century, there remains some uncertainty over the extent of possible population redistribution in sub-Saharan Africa. Almost certainly there is likely to be further concentration in the better-favored areas over areas likely to be affected by environmental decline, but the extent of such migration will vary from country to country, and probably from region to region within each country.
CONCLUSION
Internal migration is an important component of the demography of sub-Saharan Africa. Unfortunately, its study has not kept pace with the study of the two other demographic components, fertility and mortality. Though questions dealing with place of birth and place of residence in the past have been asked on recent censuses, the data have been largely useless because of problems with dating and reference periods. Attempts have been made in some countries, namely, Botswana and Burkina Faso, to collect reliable survey data; however, in general, data quality is not good.
Despite the shortage of adequate data, some observations about migration within sub-Saharan African countries can be made. Although the focus of policymakers and the research community on rural-urban migration may seem to imply that this type of migration is the most prevalent, it is not; in fact, intrarural migration, which includes nomadism and movements due to resettlement programs, is the most common. Migration from urban areas to either rural or other urban areas is the least common.
Those who decide to migrate differ from those who do not in several ways. Migrants tend to be young adult males who have a higher level of educational attainment than those who do not migrate. Wealth is also related to migration: Being better able to sponsor migrants, more wealthy families tend to be more likely to have family members who migrate.
Factors external to families and individuals are also important in creating an atmosphere conducive to migration. The international community, through export pricing and structural adjustment activities, and domestic governments, through programs that directly or indirectly address migration, may influence internal migration flows. In addition, environmental and health factors may cause some people to move within their country.
Urbanization is becoming increasingly important within sub-Saharan Africa. Unlike much of the industrialized world, where urbanization followed industrialization, urbanization and industrialization in sub-Saharan Africa have largely taken place independently. The proportion of the population living in urban areas has increased in all regions of the continent. By the turn of the century, it is expected that aside from the countries of eastern Africa, one-third of the population in all sub-Saharan African countries will live in urban areas.
Internal migration in sub-Saharan Africa has ramifications for population density. Although it does not alter the density of an entire country, it does affect the distribution of population within a country’s borders and, therefore, the density of regions within a country.
Because sub-Saharan African countries continue to be greatly affected by spatial dynamics of population, it is expedient that researchers, planners, and policymakers collaborate in research undertakings that are immediately responsive to development planning and policies.
REFERENCES
Addo, N.O. 1987 Population, migration and employment: The case of Ghana. Pp. 283–352 in H. Jemai, ed., Population et Développement en Afrique. Dakar: CODESRIA.
Adepoju, A. 1983 Studies on the Dynamics and Consequences of Migration to Medium-Size Towns in Nigeria: Research and Policy Prospects. Reports and Papers in the Social Sciences, No. 53. Paris: UNESCO.
1986 Rural Migration and Development in Nigeria. Department of Demography and Social Statistics, University of Ife, Ila-Ife, Nigeria.
1987 An overview of migration-influencing policies and programmes in Africa. Pp. 396–407 in Internal Migration and Regional Development in Africa. Accra: Regional Institute for Population Studies.
1988 Migration and urbanization: issues and policies. Pp. 123–138 in E.van de Walle, M.D.Sala-Diakanda, and P.O.Ohadike, eds., The State of African Demography. Liège: International Union for the Scientific Study of Population.
1990 State of the art review of migration in Africa. Pp. 3–41 in Conference on the Role of Migration in African Development: Issues and Policies for the ‘90s. Vol. I: Commissioned Papers. Dakar: Union for African Population Studies.
Ajaegbu, H.I. 1979 Urbanisation in Africa. Pp. 87–96 in R.K.Udo, ed., Population Education Sourcebook for Sub-Saharan Africa. Nairobi: Heinemann Educational Books.
Antoine, P., and L.Savané 1990 Urbanisation et migration en Afrique. Pp. 55–81 in Conference on the Role of Migration in African Development: Issues and Policies for the ’90s. Vol. I: Commissioned Papers. Dakar: Union for African Population Studies.
Audibert, M. 1985 The primary consequences of a large agricultural project (SEMRY II) on the demographic situation in its area of influence. Science and Technology Review: Social Sciences Series 3(3–4):64–82.
Ayiemba, E.H.O. 1990 Environmental perception as a stimulant to rural migration in a semi-arid ecosystem: A case study of Samburu District in Kenya. Pp. 115–128 in Conference on the Role of Migration in African Development: Issues and Policies for the ’90s. Vol. II: Spontaneous Papers. Dakar: Union for African Population Studies.
Badiane, W. 1990 Migration et mobilité professionelle des jeunes travailleurs de la Medina. Pp. 247– 261 in Conference on the Role of Migration in African Development: Issues and Policies for the ’90s. Vol. II: Spontaneous Papers . Dakar: Union for African Population Studies.
Bah, S.M. 1990 The application of multi-regional demography to the estimation of migration. Pp. 844–858 in Conference on the Role of Migration in African Development: Issues and Policies for the ’90s. Vol. II: Spontaneous Papers. Dakar: Union for African Population Studies.
Barnum, H.N., and R.H.Sabot 1976 Migration, Education and Urban Surplus Labour. The Case of Tanzania. Development Centre Studies, Employment Series #13. Paris: Organisation for Economic Co-operation and Development.
Bernard, F.E. 1982 Rural population pressure and redistribution in Kenya. Pp. 150–156 in J.I.Clarke and L.A.Kosinski, eds., Redistribution of Population in Africa. London: Heinemann.
Botswana 1982 Migration in Botswana: Patterns, Causes and Consequences, 3 Vols. Gaborone: Central Statistics Office.
Caldwell, J.C. 1969 African Rural-Urban Migration: The Movement to Ghana’s Towns. Canberra: Australian National University Press.
Campbell, E.K. 1987 Internal migration and regional development in Liberia. Pp. 226–253 in Internal Migration and Regional Development in Africa. Accra: Regional Institute for Population Studies.
Case, J. 1982 Migration flows, sizes and directions. Pp. 92–148 in Migration in Botswana: Patterns, Causes and Consequences, Vol. 2. Gaborone: Botswana Central Statistics Office.
Chilivumbo, A. 1985 Migration and Uneven Rural Development in Africa: The Case of Zambia. Lanham, Md.: University Press of America.
Clarke, J.I., and L.A.Kosinski, eds. 1982 Redistribution of Population in Africa. London: Heinemann.
Clarke, J.I., M.Khogali, and L.A.Kosinski, eds. 1985 Population and Development Projects in Africa. Oxford: Oxford University Press.
Cleveland, D. 1991 Migration in West Africa: A savanna village perspective. Africa 61(2):222–246.
Coulibaly, S., J.Gregory, and V.Piché 1980 Importance et Ambivalence de la Migration Voltaïque. Ottawa, Canada: International Development Research Center.
Cobbe, J. 1990 Botswana. Pp. 17–29 in C.B.Nam, W.J.Serow, and D.F.Sly, eds., International Handbook on Internal Migration. New York: Greenwood Press.
Dietz, T. 1986 Migration to and from dry areas in Kenya. Tijd. v.Econ. en Soc. Geografie 77(1):18–26.
Diop, F.P. 1985 Internal migrations, nuptiality and fertility (translated). Pp. 73–104 in Y.Charbit, L.Gueye, and S.Ndiaye, eds., Nuptialité et fécondité au Senegal. Paris: Presses Universitaires de France.
Ejiogu, C.N. 1968 African rural-urban migrants in the main migrant areas of the Lagos Federal Territory. Pp. 320–330 in J.C.Caldwell and C.Okonjo, eds., The Population of Tropical Africa. London: Longman.
Ejionye, U.A. 1988 The socio-economic impact and population policy implications of the changing patterns of population movement in Nigeria. African Population Conference Dakar 1988. Liège: International Union for the Scientific Study of Population.
Escallier, R. 1988 La croissance des populations urbaines en Afrique. Dueles element d’introduction. Espace, Populations, Sociétés 2:177–182.
Ethiopia 1987 Population and Housing Census of Ethiopia, 1984: Analytical Report on Results for Addis Ababa, Vol. 1. Addis Ababa: Office of the Population and Housing Census Commission.
Food and Agricultural Organization 1976 Production Yearbook. Rome: Food and Agricultural Organization.
1986 Production Yearbook. Rome: Food and Agricultural Organization.
Gaisie, S.K., and K.T.de Graft-Johnson 1976 The Population of Ghana. 1974 World Population Year, CIRCRED Series. Paris: Committee for the International Coordination of Research in Demography.
Gilbert, A., and J.Gugler 1992 Cities, Poverty and Development. Urbanization in the Third World. Oxford: Oxford University Press.
Gould, W.T.S. 1982 Education and internal migration. International Journal of Educational Development 1(3):103–112.
1985 Migration and development in Western Kenya, 1971–1982: A retrospective study of primary school leavers. Africa 55(3):262–235.
1988 Urban-rural return migration in Western Province, Kenya. Pp. 41–55 in African Population Conference, Dakar, 1988. Conference papers, Vol. 2, 4.1. Liège: International Union for the Scientific Study of Population.
1990 Migration and basic needs in Africa. Pp. 142–155 in Conference on the Role of Migration in African Development: Issues and Policies for the ’90s. Vol. I: Commissioned Papers. Dakar: Union for African Population Studies.
1992a Population mobility. Pp. 284–314 in M.B.Gleave, ed., Tropical African Development: Geographical Perspectives. London: Longman.
1992b Population growth, environmental stability and migration in Africa: The case of Western Province, Kenya. Paper presented at the Annual Conference of the British Society for Population Studies on Population and Environment, Oxford, September.
Gould, W.T.S., and R.M.Prothero 1975 Space and time in African population mobility. Pp. 39–49 in L.A.Kosinski and R.M.Prothero, eds., People on the Move. London: Methuen.
Hance, W.A. 1972 The crudeness of densities. Pp. 36–40 in S.H.Ominde and C.N.Ejiogu, eds., Population Growth and Economic Development in Africa. London: Heinemann.
Hill, A.L.L. 1990 Population conditions in mainland sub-Saharan Africa. Pp. 3–27 in G.T.F.Acsadi, G.Johnson-Acsadi, and R.A.Bulatao, eds., Population Growth and Reproduction in Sub-Saharan Africa: Technical Analyses of Fertility and Its Consequences. Washington, D.C.: The World Bank.
International Union for the Scientific Study of Population 1982 Multilingual Demographic Dictionary. Dolhain, Belgium: Ordina Editions.
1988 African Population Conference, Dakar, 1988. Conference papers, 3 volumes. Liège: International Union for the Scientific Study of Population.
Jamal, V., and J.Weeks 1988 The vanishing rural-urban gap in Sub-Saharan Africa. International Labour Review 127(3):271–292
Kenya 1982 1979 Population Census. Vol. II: Analytical Report. Nairobi: Central Bureau of Statistics.
Lacey, L. 1985 Interurban flows of population and occupational skills to three cities in Nigeria. International Migration Review 19(4):686–707.
Leo, C. 1985 Land and Class in Kenya. Toronto: University of Toronto Press.
Lesotho 1982 Migration and Manpower in Lesotho. Report of a National Sample Survey, 1978/ 79. Maseru: Bureau of Statistics.
Lipton, M. 1980 Migration from rural areas of poor countries: The impact on rural productivity and income distribution. World Development 8(1):1–24.
Lubell, H. 1991 The Informal Sector in the 1980s and 1990s. Paris: Organization for Economic Co-operation and Development.
Mabogunje, A.L. 1970 Systems approach to the theory of rural-urban migration. Geographical Analysis 2(1):1–18
1990 Agrarian responses to out-migration in sub-Saharan Africa. Pp. 324–342 in G.
McNicoll and M.Cain, eds., Rural Development and Population. New York: The Population Council.
Makannah, T.J. 1990 Policy measures for stemming urban in-migration. Pp. 82–95 in Conference on the Role of Migration in African Development: Issues and Policies for the ’90s. Vol. I: Commissioned Papers. Dakar: Union for African Population Studies.
Makinwa, P.K. 1981 Internal Migration and Rural Development in Nigeria: Lessons from Bendel State. Ibadan: Heinemann Educational Books.
Makinwa-Adebusoye, P. 1990 Female migration in Africa. Pp. 198–211 in Conference on the Role of Migration in African Development: Issues and Policies for the ’90s. Vol. I: Commissioned Papers. Dakar: Union for African Population Studies.
Maro, P.S. 1990 Resettlement in Africa: policies, achievement and prospects. Pp. 55–81 in Conference on the Role of Migration in African Development: Issues and Policies for the ’90s. Vol. I: Commissioned Papers. Dakar: Union for African Population Studies.
Masser, I., and W.T.S.Gould 1975 Inter-regional Migration in Tropical Africa. Special publication #8. London: Institute of British Geographers.
Mazur, R.E. 1984 Rural out-migration and labor allocation in Mali. In C.Goldscheider, ed., Rural Migration in Developing Nations: Comparative Studies of Korea, Sri Lanka and Mali. Boulder, Colo.: Westview Press.
Migot-Adholla, S.E. 1975 Migration and rural differentiation in Kenya. Unpublished Ph.D. dissertation, University of California, Los Angeles.
Mortimore, M. 1988 Adapting to Drought: Farmers, Famine and Desertification in West Africa. Cambridge: Cambridge University Press.
Nabila, J.S. 1979 The process of the decision to migrate in Ghana: An analysis of the migratory patterns of the Frafra. Pp. 228–243 in R.K.Udo, ed., Population Education Sourcebook for Sub-Saharan Africa. Nairobi: Heinemann Educational Books.
Obudho, R.A. 1990 The changing nature of Kenya’s urban demography. Pp. 214–234 in Conference on the Role of Migration in African Development: Issues and Policies for the ’90s. Vol. II: Spontaneous Papers. Dakar: Union for African Population Studies.
Odallo, D.O. 1985 Rural-rural migrant selectivity in the face of agro-industrialization (case study of migration in Mumias and Nzoia sugar schemes). Unpublished M.A. thesis, University of Nairobi.
Odegi-Awuondo, C. 1990 Life in the Balance: Ecological Sociology of Turkana Nomads. Nairobi: ACTS Press.
Ohadike, P.O., and T.Teklu 1990 Migration and shelter in a suburban African setting. Pp. 156–180 in Conference on the Role of Migration in African Development: Issues and Policies for the
’90s. Vol. I: Commissioned Papers. Dakar: Union for African Population Studies.
Omogi, M.S. 1992 Typology and characteristics of female migrants in Nairobi, Kenya: A case study of Kariobangi. Unpublished M.Phil, thesis, United Nations Regional Institute for Population Studies, University of Ghana.
Oucho, J.O. 1981 Rural-rural migration and population change. A study of the Kericho tea estates complex in Kenya. Unpublished Ph.D. thesis, University of Nairobi.
1985a Some demographic measures of rural-rural migrants in Kenya based on survey data. Genus 41(1):77–95.
1985b Co-existence of rural-urban migration and urban-rural links in sub-Saharan Africa. Development Policy and Administration Review 11(2):33–60.
1988 The rural bias of first generation rural-urban migrants: Evidence from Kenya migration studies. African Population Studies 1:61–78.
1990a Migrant linkages in Africa. Pp. 109–141 in Conference on the Role of Migration in African Development: Issues and Policies for the ’90s. Vol. I: Commissioned Papers. Dakar: Union for African Population Studies.
1990b Kenya. Pp. 275–285 in C.B.Nam, W.J.Serow, and D.F.Sly, eds., International Handbook on Internal Migration. New York: Greenwood Press.
Palmer, R. 1990 Land reform in Zimbabwe 1980–1990. African Affairs 89(2):163–182.
Peil, M., S.K.Ekpenyong, and O.Y.Oyeneye 1988 Going home: Migration careers of Southern Nigerians. International Migration Review 22(4):563–585.
Piché, V. 1990 Potentialités et originalité d’un enquête retrospective pour l’étude de la migration. Pp. 283–313 in Conference on the Role of Migration in African Development: Issues and Policies for the ’90s. Vol. I: Commissioned Papers. Dakar: Union for African Population Studies.
Population Crisis Committee 1990 Cities: Life in the World’s Largest Metropolitan Areas. Washington, D.C.: Population Crisis Committee.
Preston, S.A. 1982 Urbanization: Developing countries. Pp. 650–655 in J.A.Ross, ed., International Encyclopedia of Population, Vol. 2. New York: The Free Press.
Prothero, R.M. 1965 Migrants and Malaria. London: Longman.
1968 Migration in tropical Africa. Pp. 250–263 in J.C.Caldwell and C.Okonjo, eds., The Population of Tropical Africa. London: Longman.
Regional Institute for Population Studies (RIPS) 1987 Internal Migration and Regional Development in Africa. Accra: Regional Institute for Population Studies.
Rempel, H., and M.P.Todaro 1972 Rural-to-urban labour migration in Kenya. Pp. 214–231 in S.H.Ominde and C.N. Ejiogu, eds., Population Growth and Economic Development in Africa. London: Heinemann.
Silitshena, R.M.K. 1982 The regrouping policy in the North-East District of Botswana. Pp. 202–208 in J.I. Clarke and L.A.Kosinski, eds., Redistribution of Population in Africa. London: Heinemann.
Streiffeler, F., and M.Mbaya 1986 Zaire: Village, Ville et Campagne. Paris: Harmattan.
Thomas, I. 1982 Villagization in Tanzania: Planning potential and practical problems. Pp. 182– 191 in J.I.Clarke and L.A.Kosinski, eds., Redistribution of Population in Africa. London: Heinemann.
Traoré, S. 1990 Nomadisme et sedentarisation. L’example de la Mauritanie. Pp. 646–662 in Conference on the Role of Migration in African Development: Issues and Policies for the ’90s. Vol. II: Spontaneous Papers. Dakar: Union for African Population Studies.
Union for African Population Studies 1990a Conference on the Role of Migration in African Development: Issues and Policies for the ’90s. Vol. I: Commissioned Papers. Dakar: Union for African Population Studies
1990b Conference on the Role of Migration in African Development: Issues and Policies for the ’90s. Vol. II: Spontaneous Papers. Dakar: Union for African Population Studies
1990c Literature Review of Migration Studies in Africa. Dakar: Union for African Population Studies.
United Nations 1989 Prospects of World Urbanization 1988. New York: United Nations Department of Economic and Social Affairs.
1991 World Population Prospects 1990. New York: United Nations. United Nations Economic Commission for Africa
1975 Demographic Handbook for Africa. Addis Ababa.
1978 Demographic Handbook for Africa. Addis Ababa.
1988 Demographic Handbook for Africa. Addis Ababa.
Uyanga, J. 1985 The demographic impacts of the Cross-River plantation projects in Nigeria. Pp. 99–132 in R.E.Bilsborrow and P.E.Delarg, eds., Impact of Rural Development Projects on Demographic Behaviour. New York: United Nations Fund for Population Activities
Vignikin, K. 1986 Means of production and migration in South-east Togo (translated). Pp. 299–320 in J.Gregory, M.Therèse Duquette-Ahado, and K.Vignikin, eds., Migrations Togolaises: Bilan et Perspectives. Lomé, Togo: Université du Benin, Unité de Recherche Démographique .
Weisner, T.S. 1972 One family, two households: Rural-urban ties in Kenya. Unpublished Ph.D. dissertation, Department of Social Anthropology, Harvard University, Cambridge.
Woldemeskel, G. 1989 The consequences of resettlement in Ethiopia. African Affairs 88(3):359–374.
Wood, A.P. 1982 Spontaneous agricultural resettlement in Ethiopia 1950–74. Pp. 157–164 in J.I. Clarke and L.A.Kosinski, eds., Redistribution of Population in Africa. London: Heinemann.
1985 Population redistribution and agricultural settlement in Ethiopia 1950–1980. Pp. 84–111 in J.I.Clarke, M.Khogali, and L.A.Kosinski, eds., Population and Development Projects in Africa. Oxford: Oxford University Press.
Wright, I. 1990 Nomadisme—Pastoralisme et développment economique: cas du Niger. Pp. 630– 644 in Conference on the Role of Migration in African Development: Issues and Policies for the ’90s. Vol. II: Spontaneous Papers. Dakar: Union for African Population Studies.
Zachariah, K.C., and J.Condé 1981 Migration in West Africa: Demographic Aspects. New York: Oxford University Press.
Zambia 1985 1980 Census of Population and Housing. Vol. I: General Population and Migration Tables. Lusaka: Central Statistical Office.
Zelinsky, W. 1971 The hypothesis of the mobility transition. Geographical Review 61(3):219–249.