(October 2000) Population projections give us an idea about where a country is headed demographically — if we can accurately predict a country's trends in fertility. Mortality (AIDS in Africa) and migration (in the United States) play a significant role as well, but fertility is the driving force behind population growth in the majority of less developed countries.

Projections aside, most, if not all, observers expect that high fertility will decline in all less developed countries, at some point ending the well-known population "explosion." In fact, most projections assume that fertility rates in less developed countries will decline to about two children per woman. That assumption has a tidy result: population growth someday ends since couples just "replace" themselves with two children.

But how realistic is that assumption? Is it reasonable to assume that the preferred family size in less developed countries will track with, say, the United States, where the two-child family has become the norm? Or to assume that fertility will drop well below two children, as in Europe? Recent fertility surveys from two key countries illustrate how much in doubt the outcome really is.

The 1999-2000 Bangladesh Demographic and Health Survey (BDHS) and the 2000 Egypt DHS (both conducted by ORC Macro) offer some insights. Fertility in Bangladesh was over six children per woman as recently as the mid-1970s and declined slowly, despite government efforts to encourage family planning. A big break occurred in the early 1990s, when the 1993-1994 DHS showed a rather sudden drop to just 3.4 children per woman, a report that caused lively debate. The latest figures represent an equally unexpected development: The total fertility rate, which dropped almost imperceptibly to 3.3 between 1994 and 1996, has remained constant for the last three years.

More surprising still, along with the Bangladesh survey came the release of the Egypt survey, which showed exactly the same thing. The total fertility rate from this survey (for 1997 to 2000) was 3.5, not statistically different from the 3.6 measured for the period 1993 to 1995 in a previous DHS. (Successive surveys may measure fertility a bit differently due to variations in samples and other factors, but the examples cited here are believed to be sufficiently comparable for analysis.)

Other examples of this kind of slowdown in fertility decline — in Argentina, Paraguay, and Jamaica — show that the trend is far from isolated. Argentina is particularly noteworthy in that its total fertility rate has been about three children per woman or a little less for almost 50 years.

To be sure, examples of all types of fertility decline can be shown. Thailand experienced a relatively uninterrupted decline to below-replacement fertility, while Indonesia experienced a slowdown only to resume a gradual decrease. Still, both the Bangladesh and the Egypt surveys indicate that long-term assumptions about fertility decline should be treated with considerable caution and monitored continuously. It is possible that the earlier stages of fertility decline can be accounted for by couples who were the most receptive to the idea of limiting family size and to the use of modern contraceptive methods to achieve that goal. It may be that a fertility decline from six children to, say, 3.5, will prove much easier than from 3.5 to 2.0. And it may well be that two children will never be the norm in some countries.

Carl Haub holds the Conrad Taeuber Chair of Population Information at PRB.