Linda G.Martin, Kenneth H.Hill, and Karen A.Foote
The first systematic attempt to describe the population dynamics of sub-Saharan Africa dates back to Kuczynski’s Demographic Survey of the British Colonial Empire, published in two volumes in 1948 and 1949.1 Despite its wealth of fascinating anecdotal information, Kuczynski’s broad conclusion was that very little of a substantive nature could be said with confidence on the basis of the data then available. By the early 1960s, and the next attempt at a broad review, both data bases and methodological procedures had improved substantially. The Office of Population Research at Princeton University was able to compile estimates of demographic parameters for national and subnational populations encompassing more than half the population of the region (Brass et al., 1968).
The most striking feature of the demographic estimates published by the Princeton project was the wide variation in rates within and between countries. Coale (1968), in a paper prepared for the First African Population Conference, reported birth rate estimates that ranged from 30 to 60 births per 1,000 population, and total fertility estimates ranging from 3.5 to more than 8.0 children per woman. Mortality estimates were regarded as
Linda G.Martin is director and Karen A.Foote is a research associate for the Committee on Population at the National Research Council. Kenneth H.Hill is professor in the Department of Population Dynamics at Johns Hopkins University.
less well founded, but the crude death rate was estimated to range from 15 to 40 deaths per 1,000, the infant mortality rate from less than 100 to more than 300 deaths per 1,000 live births, and the expectation of life at birth from less than 30 to more than 45 years.
These early analyses of African data were able to draw approximate conclusions about levels of fertility and mortality, but the nature of the data and methods available precluded anything more concrete than speculation about changes over time. However, Caldwell (1968:11) wrote that “the conviction is widespread that death rates are falling, perhaps rapidly,” though also noting “that as yet no major fall in fertility has been experienced…” (p. 14). Coale (1968:186) speculated that “tropical Africa is in the early phase of rapidly accelerating population growth.”
We now know that Caldwell and Coale were correct with respect to both mortality decline and rapidly accelerating population growth. In 1960, Harold MacMillan, the British prime minister who presided over dismantling the British colonial empire in Africa, described the “wind of change” of national political consciousness that was blowing through the continent. A wind of demographic change was also beginning to blow. However, it was not until the late 1970s that good-quality demographic data became available to quantify trends. Until that time, censuses were few, far between, and of limited content, and surveys were usually neither nationally representative nor comparable across countries.
The World Fertility Survey (WFS) Program, which included surveys in nine sub-Saharan African countries in the late 1970s and early 1980s, in combination with the 1970 and 1980 rounds of censuses, provided much needed new information to policymakers and stimulated new research (see, for example, International Union for the Scientific Study of Population, 1988; van de Walle et al., 1988). However, that new research, combined with reversals of economic progress in many parts of Africa in the 1980s, generally did not offer much hope for rapid change in the underlying parameters of the population dynamics of sub-Saharan Africa.
In 1989, the World Bank (1989:40) concluded:
Significant improvement in living standards cannot be achieved over the long term unless population growth is slowed. On current trends Africa will increasingly be unable to feed its children or find jobs for its school leavers.
Estimates and projections made by the United Nations (1991) indicate that the population of the region increased from about 225 million in 1960 to 527 million in 1990. Over the past 30 years, fertility appears to have changed little, the birth rate remaining stable at 47 to 49 per 1,000. Mortality fell, though not as fast as in some other areas of the world: The crude death rate fell from about 24 per 1,000 in the early 1960s to 16 per 1,000 in
the late 1980s, with infant mortality falling from 164 per 1,000 live births to 109, and life expectancy at birth increasing from 41 to 53 years. The age distribution has remained youthful: About 45 percent of the population was under the age of 15, and 3 percent was aged 65 or over during 1960 to 1990.
The United Nations’ medium projection indicates that the population of the region will increase enormously, to 1.4 billion by 2025, with no substantial change in the youthful age structure until after 2015. By comparison, between 1990 and 2025, the population in Latin America is projected to increase from 448 to 757 million and that in Asia from 3.1 to 4.9 billion.
The African population projection assumes that fertility decline will begin and mortality decline will continue in the coming decades. However, until the late 1980s there was little evidence of any change in fertility. Since then, many changes have occurred in sub-Saharan Africa. Although population growth rates remain high, signs of reductions in fertility are appearing in several populations once regarded as having little or no prospect of lower levels of reproduction in the short term. Child mortality continues to decline in many areas, but the decline may have stagnated or even reversed in some countries, and the AIDS epidemic threatens to increase mortality among young adults in some African countries. Cities are growing rapidly, and flows of refugees due to war and political instability have been great.
At the same time, the quantity, quality, and availability of data on demographic processes in Africa have improved. This improvement has resulted from the Demographic and Health Survey (DHS) Program, which included surveys in 13 sub-Saharan African populations as of 1991, as well as from the release of additional results from the 1980 round of censuses and early results from the censuses of the 1990 round. National survey programs have often added useful information as well. Moreover, analytical procedures for dealing with incomplete or inaccurate data improved in both flexibility and scope during the 1980s, and there has been a growth in the number of African demographers and research institutions.
Accordingly, the Committee on Population of the National Research Council in 1989 established a Panel on the Population Dynamics of Sub-Saharan Africa with the charge of describing and explaining demographic changes in the region. The panel decided to conduct six formal studies, two in-depth studies of population and socioeconomic change in individual countries— Kenya and Senegal—and four analyses and cross-national comparisons of substantive population issues of particular concern in sub-Saharan Africa. The topics selected for these substantive studies were the social dynamics of adolescent fertility, factors affecting contraceptive use, demographic effects of economic reversals, and mortality effects of child survival and general health programs. The panel formed a working group of experts to carry out each of the substantive and country studies.
The panel also thought that it would be useful to both scholars and policymakers to have an up-to-date review of the levels and trends of demographic phenomena that would serve as a backdrop for the studies by the working groups. The topics selected for the review were fertility, the proximate determinants of fertility, nuptiality, child mortality, adult mortality, population distribution and internal migration, and international migration. Given the growing importance of AIDS, the panel also decided to include a review of models that have been designed to estimate the demographic implications of AIDS. A series of papers on the above topics were commissioned from experts, who were asked not to try to explain the changes, because explanation was the focus of the six working groups; rather they were asked to document actual levels, trends, and differentials to the degree that the data would allow. The eight papers appear as Chapters 2–9 in this volume.2
In Chapter 2, Barney Cohen reviews levels, differentials, and trends in fertility for more than 30 countries from 1960 to 1992. He finds evidence of fertility decline in Botswana, Kenya, and Zimbabwe, confirming the basic results of the DHS. What is new here though is his finding that the fertility decline appears to have occurred across cohorts of women at all parities, rather than just among women at middle and higher parities, as might have been expected on the basis of experience in other parts of the world. He also presents evidence that fertility may have begun to fall in parts of Nigeria and possibly in Senegal.
In Chapter 3, Carole Jolly and Jay Gribble use the proximate determinants framework and DHS data for 12 sub-Saharan African populations to estimate the contributions to fertility limitation of marriage patterns, contraceptive use, postpartum infecundability, and primary sterility. Besides analyzing national patterns, they also investigate patterns by age, education, and residence, and analyze how the roles of proximate determinants changed between the dates of the WFS and the DHS for the four countries that participated in both programs. Postpartum infecundability continues to play the most important role in reducing fertility. Marriage patterns also reduce fertility substantially, especially in urban areas and among women with more education. Contraceptive use remains low, except in Botswana, Kenya, and Zimbabwe. A comparison of the proximate determinants based on
WFS and DHS data indicates that the fertility decline in Kenya has been due primarily to a change in contraceptive use.
Etienne van de Walle tackles the challenge of teasing out trends in marriage in Chapter 4. Different data collection strategies in surveys and censuses, combined with the processual nature of union formation in sub-Saharan Africa, make it difficult to document trends with the data at hand. Van de Walle concludes that age at marriage appears to have been increasing, with women in countries in eastern and southern Africa generally having average ages of marriage of 20 years or higher, but women in western and middle Africa having relatively low ages at marriage. He also reviews the evidence for a connection between changes in marriage patterns and fertility patterns, and finds that the direct effect of later marriage on fertility may be diluted in some African populations by the incidence of premarital sexual activity.
In Chapter 5, Althea Hill uses data from surveys and censuses to estimate the trends in child mortality in the 1970s and 1980s. The news is generally good, with earlier declines continuing into the 1980s, although there are unhappy exceptions especially in countries that have suffered from political instability. In the past, mortality tended to be higher in the northern and western parts of sub-Saharan Africa, as opposed to the southern and eastern parts, but as of the late-1980s, this differential appears to have weakened.
Ian Timæus describes levels, trends, and patterns of adult mortality in Chapter 6. Once again, the results are mixed. Adult mortality in sub-Saharan Africa remains high by world standards, but the evidence suggests that there was considerable decline during the 1960s and 1970s, especially in western Africa. He notes that in the countries that have been most successful in reducing overall mortality, adults have tended to benefit more than one would have expected on the basis of mortality declines in other parts of the world. Furthermore, the limited data available suggest that there are considerable differentials in adult mortality within African populations by area, urban and rural residence, and socioeconomic status.
In Chapter 7, John Oucho and William Gould review patterns of internal migration, urbanization, and population distribution in sub-Saharan Africa. They find that rural-to-rural migration remains more important than rural-to-urban migration, although urbanization is occurring. It is expected that by the year 2000, one-third or more of the population will live in urban areas in all the subregions except East Africa.
Sharon Russell highlights in Chapter 8 the fact that the number of migrants in sub-Saharan Africa is much larger than might be expected on the basis of population size alone. She finds that movements of refugees have been most important in East Africa, whereas migration for employ-
ment has dominated elsewhere. Overall, West Africa has the highest concentration of international migrants.
Finally, in Chapter 9, Mike Stoto reviews models of the demographic effect of AIDS in Africa and finds that the predicted effects range from minor to substantial. Despite this variability, he is able to conclude that the most likely effects will be on mortality of children and young adults. However, little effect on broad population age distributions is expected. Less certain are the overall implications for population growth; growth rates are likely to be lower as a result of AIDS, but it is not clear how large the effects will be over the next few decades.
Together, these eight chapters document considerable change in the population dynamics of sub-Saharan Africa. Much remains uncertain about the future, especially given the spread of AIDS. Continuing emphasis on data collection, particularly on the conduct of regular censuses and sample surveys on specific topics, is essential. Data from the latter tend to be especially useful in explaining demographic change. Continuing support for data analysis and for demographic research centers in the region must also be given high priority. There is little doubt that a second “wind of demographic change,” the second stage of the demographic transition, is starting in some countries of Africa where fertility is beginning to fall. This process needs to be monitored and data and analyses provided to policy makers. We believe that the six reports of the working groups of the Panel on the Population Dynamics of Sub-Saharan Africa, along with the chapters in this volume, represent a useful step forward, but it is only one among the many necessary steps needed for substantial progress in understanding population change in Africa.
Brass, W., A.J.Coale, P.Demeny, D.F.Heisel, F.Lorimer, A.Romaniuk, and E.van de Walle 1968 The Demography of Tropical Africa. Princeton, N.J.: Princeton University Press.
Caldwell, J.C. 1968 Introduction. Pp. 3–27 in J.C.Caldwell and C.Okonjo, eds., The Population of Tropical Africa . London: Longmans, Green, and Co.
Coale, A.J. 1968 Estimates of fertility and mortality in tropical Africa. Pp. 179–186 in J.C.Caldwell and C.Okonjo, eds., The Population of Tropical Africa. London: Longmans, Green, and Co.
International Union for the Scientific Study of Population 1988 African Population Conference, Dakar 1988 (three volumes). Liège: IUSSP.
Kuczynski, R.R. 1948 Demographic Survey of the British Colonial Empire, Vol. 1 (West Africa). Oxford: Oxford University Press.
1949 Demographic Survey of the British Colonial Empire, Vol. 2 (East Africa). Oxford: Oxford University Press.
Macmillan, H. 1960 Speech to the South African Houses of Parliament, Cape Town, South Africa, February 3. In Oxford Dictionary of Quotations (1979:326). New York: Oxford University Press.
United Nations 1991 World Population Prospects 1990. New York: United Nations.
van de Walle, E., P.Ohadike, and M.D.Sala-Diakanda, eds. 1988 The State of African Demography. Liège: International Union for the Scientific Study of Population.
World Bank 1989 Sub-Saharan Africa: From Crisis to Sustainable Growth. Washington, D.C.: The World Bank.