This study examines fertility trends for 30 sub-Saharan African countries with Demographic and Health Surveys (DHS) between 1986 and 2006. The study uses women’s birth histories to construct panel data that enable calculation of fertility by mother’s birth cohort. The panel comprises women born between 1937 and 1990 and covers their fertility behavior between 1952 and 2005. The study’s focus on cohort trends in fertility distinguishes it from many other studies of fertility transition and enables examination of women’s fertility over their own lifetimes, rather than what fertility rates at a single point in time.
Over the past five decades, most countries studied show signs of fertility decline in the cohort measures examined. Comparing women born between 1950 and 1959 with those born between 1960 and 1969 reveals that 17 of the 30 countries studied show rates of fertility decline indicative of fertility transition. However, there is significant variation among countries in fertility levels and trends.
This study also finds some “stalling” in fertility declines in a few countries, as decreases in fertility become less pronounced than in the past. Fertility decline stalling for most age groups can be witnessed in the Congo (Brazzaville) and Madagascar. In several other countries, stalling is only witnessed in certain age groups, while in some countries fertility is increasing among certain age groups. Thus these data contain suggestions of stalling fertility transitions. For nearly all countries, however, cohort fertility rates for women born in the most recent past show decreases suggesting little evidence of an unequivocal reversal in Africa’s fertility transition.
The largest percentage fertility declines by age group appear for women aged 15-19. This pattern suggests that women are delaying the start of reproduction. Beyond this youngest reproductive age group, fertility declines are similar for younger and older age groups for most countries in the study. There is some evidence that earlier cohorts have larger fertility declines among older women, followed by declines among younger women in subsequent cohorts.
The analysis offered here is intended to provide a starting point for further analysis of the retrospective cohort fertility data detailed in this report. Retrospective data may suffer from bias related to recall errors or sample selection. The extent of bias related to retrospective data is unknown and could be a factor in the trends represented here. The large number of data tables and figures included in this report (see Appendix B for data tables and figures for each country studied) present new opportunities for exploring fertility trends in sub-Saharan Africa and are included for this purpose.