Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
8 Conclusions In this concluding chapter, we first return to the question motivating this report: Is sub-Saharan Africa on the brink of a contraceptive revolution that signals the onset of fertility decline? We then consider the research needed to answer outstanding questions. FINDINGS Although there is considerable uncertainty about Africa as a whole, the evidence on balance points to an undeniable trend in Zimbabwe, Botswana, and Kenya. The changes observed in these three countries over the past decade indicate that selected parts of Africa have joined other regions of the developing world in a contraceptive revolution. However, in the vast majority of countries within Africa, the preva- lence of use of modern methods of contraception is less than 6 percent, placing them squarely in the "emergent" category with regard to family planning programs (Destler et al., 1990~. In these countries, postpartum nonsusceptibility due to lactational amenorrhea and sexual abstinence is more dominant than modern contraception in restraining fertility. lone can make this statement not only for countries that have had a major demographic survey, but also for those that have not, given that in the majority of the remaining countries there has been little family planning program activity. 212
CONCLUSIONS 213 A number of factors are associated with the increased use of modern contraception. Female education is clearly an important determinant of contraceptive use at the individual, regional, and national level (see Chap- ters 2, 3, and 6~. In this sense, Africa follows a pattern common to other regions of the developing world, although the changes in contraceptive use associated with changes in female schooling are not as large as in much of Latin America. As demonstrated in Chapter 6, current use of modern meth- ods increases above the 10 percent level only in regions that have a mean length of female schooling of four years or more. No doubt in part because of the greater educational opportunities available in urban than in rural areas, urbanity was shown to directly and positively affect contraceptive use at the individual level and negatively affect fertility at the national level (see Chapters 2 and 3~. At the regional level, urbanization was shown to influence the average level of schooling, but otherwise had no direct effect on contraceptive use (see Chapter 6~. The percentage of women in a polygynous union, a proxy for features of social organization that promote high fertility, was also negatively re- lated to contraceptive use at the regional level. The proportion Muslim indirectly reduced contraceptive use by influencing the average level of schooling. Although significant progress has been made in reducing infant and child mortality in Africa, the three higher-use countries are clearly distinc- tive with respect to mortality levels and trends. In our view, steady in- creases in contraceptive use resulting in fertility decline in other countries of Africa are doubtful without continued improvements in mortality. Mor- tality decline has provided parents with greater assurance that their children will survive to adulthood, thus reducing the need to have additional children as insurance against this threat or as compensation for the. actual loss of one or more children. Such improvements in mortality may prove difficult in countries experiencing economic difficulties and cuts in the provision of health services. In addition to these factors associated with contraceptive use, the strength of family planning programs is central to the prospects for fertility decline. Evidence reviewed in Chapter 5 demonstrates that family planning pro- grams in Botswana, Kenya, and Zimbabwe are the most well developed in Sub-Saharan Africa. As shown in Chapter 6, certain regions that would be expected to have a prevalence of modern method use of at least 10 percent (based on levels of female education) did not, and all these were located in countries with weak family planning programs. Whereas much of the ear- lier demographic literature focused on the socioeconomic factors affecting fertility, there has been a growing awareness during the past decade of the important role of the family planning supply environment with respect to meeting the needs of couples motivated to delay or limit births.
214 FACTORS AFFECTING CONTRACEPTIVE USE In the sub-Saharan African countries that have achieved at least moder- ate success in family planning, the public sector provided the national cov- erage of services that resulted in increases in national contraceptive use. However, it is clear from Chapter 5 that private family planning associa- tions have played a pioneering role in legitimizing the use of family plan- ning and implementing many of the early services. Assuming that the use of modern methods continues to increase in the three higher use countries (and begins to take hold in others), we suspect that Africa will follow the pattern of other developing countries in terms of diversification of method mix. The oral pill is the most widely used contra- ceptive according to the World Fertility Survey (WFS) and Demographic and Health Survey (DHS) studies conducted in Africa to date (see Chapter 2~. Yet the experience of other developing regions suggests that method mix will expand as overall prevalence increases.2 Indeed, the 4 to 5 percent prevalence of female sterilization found in Botswana and Kenya suggests that even this method once thought to be totally unacceptable in cultures that placed such a high value on fertility-is gaining in acceptance. Given anecdotal evidence as to the popularity of Depo-Provera (the three-month injection) in those countries where it has been introduced, increased ava~l- ability of this method would be expected to result in greater diversification of the method mix. Remarkable as the increases in contraceptive use have been in the coun- tnes with higlier prevalence, it is important to keep in perspective the fact that in demographic terms, these three countries represent less than 7 per- cent of the population of sub-Saharan Africa. Moreover, even these coun- tries have attained only a moderate level of use in terms of family planning program evolution. If one excludes Mauritius and South Africa (as we have done in this report, except in passing, based on their atypical socioeconomic levels and ethnic compositions), the two African countries (Liberia and northern Sudan) that follow these three countries in terms of contraceptive prevalence have a modern use rate of only 5.5 percent. Our analysis indicates that several factors will be influential in deter- m~ning future contraceptive prevalence levels. Although impressive progress was made dunug the decade of the 1980s in terms of population policy and family planning program implementation, it is our view that the continued 2The majority of developing countries in Asia and Latin America with a contraceptive prevalence of at least 30 percent are not single-method countries (Rutenberg et al., 1991). Historically, as prevalence increases, users demand methods for both spacing (satisfied by reversible methods) and limiting (satisfied by long-term or permanent methods), and method mix is in turn diversified. Moreover, with the U.S. Food and Drug Administration approval of Depo-Provera and NORPLANTR, the range of methods available in sub-Saharan Africa should be greater in the l990s than in the 1980s.
CONCLUSIONS 215 development of strong family planning programs will be needed to meet the demand for family limitation and offer an alternative to high fertility. The evidence presented in Chapter 5, although qualitative in nature, indicates that the three countries that have achieved modern prevalence rates of more than 15 percent have had the best-developed family planning programs. The development of programs in Africa will be helped undoubtably by sustained government commitment. Recent statements from more than 50 African ministers at the Third African Population Conference in Dakar, Senegal, attest to the growing support for family planning (United Nations Population Fund, 1992~. The mechanisms used to deliver family planning services in Africa in the coming years are likely to resemble those now in use both in Africa and in other regions of the developing world: clinic-based facilities, commu- nity-based distribution, and social marketing. However, the patterns of social organization in Africa may provide a unique opportunity to involve local organizations and community networks (women's groups, networks of traders, etc.) in legitimizing the concept of family planning and disseminat- ing information on the methods. In addition, given that men have a domi- nant role in fertility decision making, programs in Africa may derive par- ticular benefit from targeting interventions to this subgroup. In view of economic difficulties in some African countries, donor sup- port for family planning is likely to remain crucial in the coming years. Even where the political will exists, governments stretched thin with pro- viding basic services to growing populations will find it difficult to imple- ment effective family planning programs without continued donor support. Whereas in the past, Africa had been somewhat neglected (relative to Asia and Latin America) by international donors to population activities, this situation shifted dramatically during the 1980s, when Africa became a pri- ority for numerous organizations. The external investment in family plan- ning in the 1980s contributed to the increases in contraceptive use in Botswana, Kenya, and Zimbabwe, and it may begin to affect prevalence rates in the l990s in other countries that have intensified their program efforts (e.g., Niger, Rwanda). Changes in African social structure will certainly affect the future de- mand for children. As shown in Chapter 4, social factors at the community and household levels (particularly the value attached to perpetuation of the lineage, which has served as an organizing cultural principle in many areas of Africa) have exerted pressure on couples to have large families. In many ways, these factors explain "why Africa is different" with regard to the fertility transition. Yet these social structures are not immutable, and changes in the nature of kinship support and of spousal relations would be expected to influence attitudes toward the value of family planning. There is consensus in the literature that pro-natalist social factors are
216 FACTORS AFFECTING CONTRACEPTIVE USE being undermined progressively by economic development and perhaps, in some areas, by economic crisis. Thus, the degree to which African social organization will limit fertility decline in the future is unclear. For ex- ample, there are indications of growing conjugal closeness and shared deci- sion making, possibly resulting from changes in childrearing costs and edu- cational aspirations. Land scarcity due to high population density and rising educational costs in some areas may increase the perceived benefits of fewer children. As discussed in Chapter 3, deteriorating economic conditions in some countries may decrease the prevalence of child fostering as families seek to care for their immediate kin and avoid the costs associated with caring for children of relatives. These same conditions may also increase the likelihood of resource pooling and joint decision making as families strive to meet the sustenance and educational needs of their children. What is clear from our review of African social structure is that there is great variation in how it affects the demand for children. Although scat- tered qualitative evidence indicates that social factors are changing in such a way that might lead to lower fertility desires, there is an insufficient body of knowledge to predict the direction of change for most of Africa. Many of the changing factors we have emphasized are more important in urban areas and among the educated populations that have expressed a desire for family limitation. Although the high-fertility rationale persists in many areas, the examples of Zimbabwe, Botswana, and Kenya demonstrate that parts of sub-Saharan Africa are receptive to contraceptive use. However, dramatic increases in prevalence may not be imminent for other areas, al- though we believe that contraceptive use will take hold eventually. Be- cause of the variation in cultural and socioeconomic structures across Af- rica, we expect increases in contraceptive use to be uneven; increase is initially more likely in areas that are urban, with educated populations, and with access to social services. There are several factors that may curtail the spread of contraceptive use in Africa. First, although there are few hard data to substantiate the point, preliminary reports from countries ravaged by acquired immune defi- ciency syndrome (AIDS) suggest that this epidemic may change perceptions regarding mortality. In an effort to ensure the survival of sufficient num- bers of children and to maintain continuity of the lineage, women may seek to have as many children as soon as possible. Under such circumstances, contraceptive use would be counterproductive. However, educational pro- grams to increase condom use to prevent the spread of AIDS may have the opposite effect on use. Second, a few African countries are experiencing extreme political and social unrest. Civil war and famine have devastated regions of Angola, Ethiopia, Liberia, Mozambique, Somalia, Sudan, Uganda, and Zaire. Under conditions that threaten survival on a daily basis and severely disrupt access
CONCLUSIONS 217 to health, family planning, and other services, there is no reason to expect more widespread use of contraception. Third, further increases in contraceptive use are dependent on contin- ued improvements in female education and the returns to schooling. As discussed in Chapter 3, the willingness of parents to accept the quantity- quality trade-off (having fewer children but investing more heavily in their schooling) is dependent on their perceptions that increased schooling will in fact result in greater economic benefits to both the child and the family in later life. However, studies in selected countries demonstrate that the qual- ity of education has deteriorated, which decreases the return that can be expected on such an investment. The return is also heavily dependent on the supply and demand for labor by educational level. Thus, the evidence for a quantity-quality demographic transition is decidedly mixed, and con- tinued progress in terms of female education cannot be taken for granted. RESEARCH GAPS There are a number of areas in which further research would greatly assist in understanding of the dynamics of contraceptive use in sub-Saharan Africa. We present these topics in an order that mirrors the chapters in this volume. Levels and Trends in Contraceptive Use- Contraceptive Discontinuation There is little reliable information on the average duration of contracep- tive use (i.e., once an acceptor begins a period of contraceptive use, how long does she use the original method or, if switching occurs, any method). As mentioned in Chapter 2, such information can be obtained from two different sources: program records or population-based surveys. In many countries, service statistics if collected at all are unreliable. Even if carefully recorded, data based on service statistics suffer from the problems of sample selectivity and loss to follow-up. Because of these limitations, there is a preference for obtaining continuation data from population-based surveys. The DHS questionnaire for high-prevalence countries contains an instrument for obtaining retrospective data with which to calculate continu- ation rates. But to date, none of the African DHS have employed this questionnaire, given their relatively low levels of prevalence. A more widespread use of these questions in African countries that have at least 15 to 20 percent prevalence of modern contraceptive methods would provide some needed information on discontinuation.
218 Effects of Economic Downturns FACTORS AFFECTING CONTRACEPTIVE USE Socioeconomic Context There are competing hypotheses as to the effects of economic hardship on contraceptive use. On one hand, it has been argued that low levels of socioeconomic development (which are generally accompanied by low lev- els of female education, high infant mortality, and large percentages of the population living in rural areas) work to sustain the demand for a large number of children. On the other, it has been hypothesized that the current economic crisis in many parts of Africa may cause Africans to respond by altering their attitudes regarding family size and increasing their receptivity to family planning. The theories underlying these competing positions are discussed in detail in Chapter 3, but there has been very little empirical work on the effects of economic downturns.3 Effects of Child Mortality and AIDS on Demand for Children and Attitudes Toward Family Planning It is generally accepted that high levels of child mortality tend to sus- tain the demand for a large number of children among parents seeking to insure themselves against possible future loss or compensation for deaths that have already occurred. The populations of countries ravaged by the AIDS epidemic are now painfully aware that children born to mothers in- fected with the human immunodef~ciency virus ~V) may be infected themselves. Whereas medical specialists advise against pregnancy for women who are infected with HIV (to avoid hastening the onset of symptoms), this advice may be meaningless to women who measure their own personal worth by their contribution to continuing the lineage. More data of a qualitative nature are needed to understand the motives of women in this situation and how AIDS affects their attitudes toward family planning. Costs of Investments in Children, Including Education Given the importance of the quantity-quality trade-off to the question of impending fertility decline in Africa, it is surprising that there are not more data on the actual costs of investments in children in the African context. The few studies presented in this report suggest that in certain 3An exception is the report of the Working Group on the Effects of Economic and Social Reversals (1993), which estimates the effects of economic reversals on child mortality, mar- riage, and fertility, with special attention to first and second births, in seven African countries.
CONCLUSIONS 219 settings (e.g., Kenya), men have begun to entertain the prospect of fewer children because of the increased costs of schooling. However, these stud- ies have tended to be qualitative in nature, such that the results are limited in geographical scope and do not produce quantitative estimates of invest- ment per child. To investigate this issue more rigorously, it is essential to have better data on the actual costs of schooling in different countries, as well as the perceptions of the costs and benefits of investing in children's education. Female Education, Income, and Contraceptive Use Although improvements in female education are associated with in- creased contraceptive use, the research that has led to this finding generally lacks controls for income, which may distort the relationship. Because increases in female education most likely result in higher incomes (and both are associated with lower fertility), it is important to distinguish which of these changes is primary in driving lower fertility. Income has not been included in most studies because of the difficulty of measuring it. The one study we found that does control for income suggests that female education is significantly associated with decreased fertility regardless of the effect of income on fertility (see Chapter 3~. Further work in this area would be most useful if it included such information. Community/Kinship/Household Extent of Nucleation of the Family and Child Fostering There is need for further research on the extent of joint decision making between spouses. The premise of the weak conjugal bond needs to be revisited in light of urban life-styles, exposure to western ideas via the mass media, and changing economic circumstances. Because of the heterogene- ity of sub-Saharan Africa, these factors may result in different responses. In some areas there may be greater pooling of resources and conjugal close- ness (see Chapter 4) and in other areas the family structure may not depart from a lineage orientation. Moreover, the effects of these factors on child fostering may be mixed. Such varied responses will have a profound effect on the future fertility desires of different African populations. Quantification of Kinship Factors Given the importance of kinship in influencing the demand for children, it would be useful to devise means for integrating kinship factors into quan
220 titative data analysis. FACTORS AFFECTING CONTRACEPTIVE USE Such efforts would clarify the relative effects of social organization and socioeconomic factors on the demand for births. Local Social Organization and the Diffusion of Family Planning Chapter 4 highlighted the potential role of local social networks in the provision of information and the legitimization of contraception. There is a need for further research on the possible contributions of these organiza- tions. Population Policies and Program Implementation- Service Availability at the Regional Level Not only in Africa but also in other parts of the developing world, attempts to evaluate and quantify the family planning supply environment have been limited to date. Although some information is available on the national level, there is a need for subnational data that indicate not only the quantitative but also the qualitative aspects of service delivery. Without these data, it is impossible to assess the effect of family planning programs on changes in contraceptive use. This list of research gaps is by no means exhaustive. However, it includes those items that would have been most beneficial to improving our analysis of the factors affecting contraceptive use in sub-Saharan Africa. Regardless of the limitations of our study, a central conclusion remains: We believe that although the social supports for high fertility have not disappeared, Zimbabwe, Botswana, and Kenya demonstrate that increases in contraceptive use can occur in sub-Saharan Africa. We believe that future fertility decline is likely in these three and other countries assuming provi- sion of family planning services, improvements in child mortality, and progress in female education.
