The Pace of Fertility Decline

Fertility has fallen by an average of 1% per year in the 38 developing countries surveyed more than once since 1990. The pace of change varies widely among countries, however—from Yemen, where the TFR fell by 1.2 children per woman in five years, to Niger, where the TFR rose by 0.2 children per woman in six years. Although generally fertility has been falling, surveys conducted since 1990 suggest that the pace of decline has been slowing compared with fertility declines in the 1970s and 1980s. Considering all countries that have had more than one survey, including surveys before 1990 as well as after, instances of slowing in the rate of fertility decline outnumber accelerations in the rate of decline by two to one.

Many factors help to explain why the pace of fertility decline appears to be slowing. Fertility declines faster when fertility rates are high (21), as was the case in many developing countries in the 1970s and 1980s. Meanwhile, economic and social changes, new information and ideas, new reproductive attitudes, smaller family size ideals, and access to effective contraception were spreading rapidly.

At the same time, infant and child mortality was falling in response to better public health measures in most countries. Family sizes increased as more children survived their first few years of life. Thus demand for fewer children and use of contraception grew, resulting in rapid declines in fertility.

By the start of the 1990s, in contrast, family planning information and services had become widespread throughout much of the developing world. Most people already knew about contraception and where to obtain it. Moreover, the influence of new social norms favoring smaller families had already been reflected in lower fertility rates. Much of the latent demand for family planning had been met. Given these trends and smaller family sizes compared with the 1970s and 1980s, it should not be surprising that, while fertility continued to decline in the 1990s in most countries, the pace of decline has been slower.

Prospects for further decline. Despite fertility declines, in a number of developing countries women still are having more children than they say they would prefer, and many still prefer larger families than women in developed countries. Thus there remains substantial scope for further fertility declines. Fertility can be expected to fall further as more women become better able to prevent unwanted births. In addition, future fertility levels are likely to depend increasingly on changes in social and economic factors that lead people in developing countries to desire fewer children (21).

In the coming years fertility is likely to decline fastest where it is now highest—that is, in sub-Saharan Africa. So far, the pace of fertility decline in sub-Saharan Africa has been considerably slower than in other developing regions, since on average African couples prefer to have more children than do couples elsewhere (29).

In the 1980s fertility started to decline in a few sub-Saharan countries—the beginning of a trend to smaller family preferences and more contraceptive use in the region. Nevertheless, during the 1990s many sub-Saharan countries were only starting their fertility transitions, leaving room for more rapid declines than elsewhere, where the fertility transition was more advanced (29).

The “tempo effect.” Recently, fertility appears to have stopped falling altogether in some developing countries (52). One reason is a demographic “tempo effect.” This effect results from changes in the timing of childbearing rather than the overall demand for children. It probably explains some of the apparent stalling of fertility decline in such countries as Bangladesh, Colombia, Egypt, Indonesia, and Turkey, although the magnitude of the tempo effect is difficult to estimate (18).

The tempo effect operates as follows: When the average age at which women give birth rises, the TFR appears to decline temporarily, even if women continue eventually to reach the same completed family size as before. When completed family size is falling at the same time that the average age at first birth is rising, as has been the case in the past few decades until recently, the TFR declines even faster (18).

But when the average age at which women have their first birth stops rising, as was the case in some countries during the 1990s, the TFR can temporarily stop its decline or even rise somewhat. This effect is completely independent of any changes in family size preferences or levels of contraceptive use.

Reporting 30 Years of Survey Data

This report is the latest in a series of Population Reports issues covering survey findings on fertility, family planning, and health topics in developing countries. This report focused on data from 71 countries and 120 DHS or RHS surveys.

These surveys, which collect information from women (and sometimes from men) on fertility, contraceptive use, fertility desires, and other key reproductive health topics, provide the best available evidence on levels and trends in these indicators in the developing world and in some countries of the former Soviet bloc. Some 40 countries (38 in developing regions) were surveyed more than once since 1990, allowing examination of trends. Data from the YARHS are included here when appropriate.

In this issue of Population Reports, 60 developing countries are covered, representing nearly 50% of the population of the developing world. Excluding China, these surveys represent about 71% of the developing world. (Although Puerto Rico is a commonwealth freely associated with the United States, it is included in the report because it is part of the RHS survey program. It is reported in the Latin America and Caribbean region because of its geographic location.) The report treats separately the 11 surveyed countries in Eastern Europe and Central Asia that were part of the Soviet bloc.

About the Data

The data presented here come principally from special tabulations that the DHS and RHS programs produced for this report and from on-line DHS STATcompiler tabulations. Some data differ slightly from the data in published Final Country Reports. Differences are mainly due to additional cleaning of the data sets since publication of the final reports. In some cases final report data have been used. (STATcompiler data and many final country reports are available at http://www.measuredhs.com/).

Not all surveys include questions on all of the topics discussed in this report, nor are the respondents defined the same way in all surveys. As a result, the number of countries reported in different sections and tables sometimes varies.

Surveys are conducted in different years in different countries; the year each survey was conducted is shown in the tables. The comparisons in this report generally use the most recent survey data from each country. Unavoidably, for a few countries the most recent data are as much as 12 years old. The surveys included in this report are listed on p. 40.

DHS and RHS surveys do not cover all developing countries. Therefore, some tables and discussions at the beginning of this report include estimates of fertility levels and contraceptive prevalence for other countries calculated and published by the United Nations (http://www.un.org/esa/population/unpop.htm) and the United States Bureau of the Census (http://www.census.gov/ipc/www/idbnew.html).

A forthcoming issue of Population Reports will contain information about men’s family planning attitudes and behavior and about HIV/AIDS-related knowledge and behavior among both men and women.

Earlier survey findings can be found in previous issues of Population Reports. Survey findings from the 1980s appear in The Reproductive Revolution: New Survey Findings, Series M, No. 11, December 1992. Findings from the 1970s can be found in Fertility and Family Planning Surveys: An Update, Series M, No. 8, September–October 1985, Contraceptive Prevalence Surveys: A New Source of Family Planning Data, Series M, No. 5, May–June 1981, and The World Fertility Survey: Current Status and Findings, Series M, No. 3, July 1979. All five issues are available on CD-ROM (to order, see below).

Additional Survey Data Online

For readers who are interested in obtaining all of the data used for this issue of Population Reports, supplementary data tables are available for download and printing from the World Wide Web at http://www.infoforhealth.org/pr/m17/m17tables.shtml

A “surveys package” is also available upon request. This package includes a CD-ROM with Excel files of all tables used in preparing this report (including those below), scanned PDF files of all five of the Population Reports survey data issues (see above), and POPLINE abstracts of the most useful bibliographic items in these issues.

Requests for this CD-ROM package can be addressed to:

Orders Department (Surveys CD-ROM)
Information and Knowledge for Optimal Health Project
111 Market Place, Suite 310, Baltimore, MD 21202, USA
Phone: 410-659-6300 Fax: 410-659-6266
or by e-mail to: Orders@jhuccp.org
(Please include your full name and address and e-mail.)

All data tables published in this report (Tables 1–14) also appear as Web Tables on the Internet with additional data included from earlier DHS and RHS surveys for countries with more than one survey since 1990. Thus, for example, “Web Table 5” is comparable to “Table 5” in this report but with data from earlier surveys as well as the most recent. In addition, the following 11 supplementary data tables appear on the Internet but not in the printed report, as follows:

Supplementary Data Tables

Web Table A. Total Fertility Rates and Age-Specific Fertility Rates, All Women Ages 15–49, 1990–2001

Web Table B. Total Fertility Rates, All Women Ages 15–49, by Education and Residence, 1990–2001

Web Table C. Contraceptive Use, Unmarried, Sexually Active Women Ages 15–49, 1990–2001

Web Table D. Contraceptive Use by Women's Charac-teristics, Married Women Ages 15–49, 1990–2001

Web Table E. Intention to Use Contraception, Married Women Ages 15–49, 1990–2001

Web Table F. Reasons for Nonuse of Contraception, Married Women Ages 15–49, 1990–2001

Web Table G. Knowledge of Modern Contraceptives, Married Women 15–49, by Women's Characteristics, 1990–2001

Web Table H. Attitudes of Couples Toward Family Planning, 1990–2001

Web Table I. Age at First Marriage, Married Women Ages 15–49, 1990–2001

Web Table J. Breastfeeding and Postpartum Insusceptibility, Married Women Ages 15–49, 1990–2001

Web Table K. Abortion Rates, Married Women Ages 15–49, 1990–2001