The End of the Fertility Transition in the Developing World
Future declines in fertility in some developing countries may not be as rapid as in the past, and in a number of countries fertility may stall above the replacement level for a prolonged period.
Since the 1960s many developing countries have experienced rapid fertility declines. It is widely expected that countries that are still in transition will continue their declines until fertility drops to or even below replacement. This assumption has been incorporated into population projections made by the United Nations and the World Bank.
An examination of fertility trends since the late 1980s in countries with multiple Demographic and Health Surveys (DHS) reveals, however, that fertility as measured by the total fertility rate stalled in mid-transition in seven countries: Bangladesh, Colombia, the Dominican Republic, Ghana, Kenya, Peru, and Turkey. In each of these countries fertility was high (>6 births per woman) in the 1950s and then declined to fewer than 5 births per woman in the early or mid-1990s, before entering a stall during the 1990s. The level of stalling varied from 4.7 births per woman in Kenya to 2.5 births per woman in Turkey. This surprising development has implications for future population growth, because this growth is sensitive to minor variations in fertility trends.
An analysis of trends in the determinants of fertility in these seven countries leads to three conclusions.
First, fertility stalls are accompanied by a leveling off or sharp deceleration in the trends in contraceptive use and the demand for contraception, and by a leveling off in fertility preferences as measured by the wanted total fertility rate.
Second, there is no common trend in socioeconomic determinants during fertility stalls. In some stalling countries (Ghana and Kenya) development indicators changed little, while in others socioeconomic development continued at a fairly rapid pace. However, the level of fertility relative to the level of development seems to play a role as a cause of stalls. At the onset of six out of the seven stalls, fertility was lower than expected for the level of development. It is not clear why countries have reached this low fertility relative to their socioeconomic predictors, but once this is the case fertility can subsequently be expected to move closer to the predicted level, thus making a stall more likely.
Third, little support is found for the hypothesis that declining access to contraception is a main cause of stalling fertility. Program effort scores rose during all stalls except in the Dominican Republic. In addition, measures of unmet need and unwanted fertility showed no significant recent upward trend in the stalling countries, although Kenya experienced slight increases.
Any policy response to address stalling fertility should be tailored to the circumstances of the individual country and, in particular, its levels of wanted and unwanted childbearing. A country in which unwanted fertility is higher than a few tenths of a birth can especially benefit from further investments in family planning programs. Such investments are most needed in countries such as Bangladesh, Ghana, Kenya, and Peru with the highest levels of unmet need and unwanted fertility. Family planning programs can reduce unwanted fertility but their effect on desired family size is apparently weak.
Countries in which wanted fertility has stalled well above the replacement level will therefore need declines in preferences to complete their fertility transition. Such declines are usually achieved by improvements in socioeconomic conditions. Among the seven stalled countries, Ghana and Kenya have relatively high wanted fertility levels (3.7 and 3.6 births per woman, respectively), and their levels of development as measured by real GDP per capita, child survival, and proportion schooled are low and have leveled off. In these two countries improvements in development will almost certainly be needed if desired family size and actual fertility are to fall substantially below current levels.
In contrast, in Bangladesh, Colombia, the Dominican Republic, Peru, and Turkey wanted fertility has already dropped to about two births per woman and any further declines in overall fertility are likely to come from reductions in unwanted fertility.
The causes of stalling fertility transitions (PDF)
Policy Research Division Working Paper (no. 204)
Publication date: 2005
Completing the fertility transition in the developing world: The role of educational differences and fertility preferences (PDF)
Policy Research Division Working Paper (no. 177)
Publication date: 2003
Completing the fertility transition in the developing world: The role of educational differences and fertility preferences (abstract) (HTML)
Population Studies 57(3): 321-336
Publication date: 2003
Future trends in contraceptive prevalence and method mix in the developing world (abstract) (PDF)
Bongaarts,John; Johansson,Elof D.B.
Studies in Family Planning 33(1): 24-36
Publication date: 2002
The potential role of contraception in reducing abortion (abstract) (PDF)
Bongaarts,John; Westoff,Charles F.
Studies in Family Planning 31(3): 193-202
Publication date: 2000
The fertility impact of changes in the timing of childbearing in the developing world (abstract) (HTML)
Population Studies 53(3): 277-289
Publication date: 1999
Location: Developing Countries
Poverty, Gender, and Youth
Duration: 1/1997 - ongoing
John B. Casterline (Pennslyvania State University)