The implementation of family planning programs has been the principal population policy instrument in the developing world over the past few decades. This paper reviews that controversy over the role these programs have played in reducing fertility. Opposing views on a number of contentious issues (for example, the significance of unmet need and unwanted fertility) are summarized and a consensus position is presented. Surprisingly, recent estimates of the fertility reductions achievable through the implementation of strong family planning programs differ only modestly--from 1 to 1.4 births per woman. Since only a small proportion of countries have implemented strong programs, the (unweighted) average impact of programs in 1985-90 is estimated at only 0.33 births per woman. However, the (weighted) average, which gives the program impact for the developing world as a whole, is estimated at 0.96 births per woman in the late 1980s. This suggests that programs have been responsible for about 43% of the fertility decline in the developing world between 1960-65 and 1985-90.

This report summarizes presentations and discussions at the Workshop on the Social Processes Underlying Fertility Change in Developing Countries, organized by the Committee on Population of the National Research Council (NRC) in Washington, D.C., January 29-30, 1998. Fourteen papers were presented at the workshop; they represented both theoretical and empirical perspectives and shed new light on the role that diffusion processes may play in fertility transition. These papers served as the basis for the discussion that is summarized in this report.

Family planning programs have been highly successful over the past 30 years in providing women in developing countries with desired access to contraceptive services and helping to reduce fertility rates. Notwithstanding this success, there is still an urgent need for these programs. The world's population is increasing, with annual population growth still approximately 80 million people. Nearly all of this growth is occurring in developing nations, where fertility rates remain relatively high. This high fertility runs counter to the preferences expressed by millions of women, who actually want to have smaller families. Family planning programs are also desirable because they are associated with a range of other benefits, most notably improvements in women's and children's health. Host countries provide about 60 to 75 percent of funding for family planning. However, funding and technical assistance from donor nations, especially the United States, have been crucial to the past success of family planning programs and are equally important for strengthening and expanding program efforts to meet future challenges.

This paper critically reviews the methods for measuring the impact of family planning programs on fertility used primarily in the last decade, based on the literature as well as site visits and interviews. The strengths and weakness of several acceptor-based and population-based methods are evaluated for use in particular settings. The acceptor- based methods include: reproductive process analysis, standard couple- years of protection (SCYP), CYP, individual matching and component projection (CONVERSE). The population-based methods are: decomposition (standardization), trend analysis, prevalence, areal regression, multilevel regression and experimental design. Current evaluations are focusing on identification of the aspects of programs that have an impact, their interaction, and their relative importance in different settings. Except for the use of CYP in several Asian programs to monitor performance of the program, population-based methods are in vogue recently. Reasons for their popularity are the availability of national survey data, microcomputers and software such as TARGET, and the need for target setting in country policy. Areal and multilevel regression and experimental design are the methods receiving the most attention. The success of the Matlab project has stimulated interest in experimental design.