The Determinants of Fertility in Rural Peru: Program Effects in the Early Years of the National Family Planning Program

Author(s): Angeles G, Guilkey D K, Mroz T A

Year: 2002

Abstract:
After several attempts over a 20-year period, Peru enacted its National Policy on Population in July 1985. Using data from the 1991 Peru Demographic and Health Survey (PDHS91), a linked Peru Situation Analysis (PSA92) community and facility data set collected in 1992, and a unique region-level data set gathered specifically for this analysis, this paper examines the determinants of fertility in rural Peru before and after this important date. Particular attention is paid to assess the effect of family planning services on fertility. The empirical model that is used combines a model of the timing and spacing of births with a model of the timing of the placement of family planning (FP) services in communities. This modeling strategy allows us to control for the non-random placement of FP services that could potentially bias the measures of program impact. An illustration of the potential relationship between fertility and FP services can be seen in Figures 1 and 2. Figure 1 presents age-specific fertility rates (ASFR) for the period 1972-1991 from the fertility histories for women in the rural sample of the PDHS91. For all age groups except the youngest, fertility appears to be declining, and the rate of the decline seems to have accelerated in the 1980's. Figure 2 depicts the expansion of FP services within five kilometers of the rural PDHS communities for different type of providers. Public FP services were virtually non-existent in rural Peru during the 1970's and the expansion in services really started after the passage of the National Policy on Population in 1985. The timing and extent of the fertility decline appear to coincide with the growth of the government provision of FP services. Our data set allows us to estimate the determinants of the annual probability of a birth for every year between 1972 and 1991 and so we completely span this period of marked change. Clearly, any change in FP policy will not have an immediate impact on fertility. One of the goals of this paper will be to measure the lag in program impact if, in fact, there is an impact at all. The next section of this paper presents a brief review of Peru's family planning program. This context will be important in the interpretation of our empirical results. Section III discusses estimation difficulties that arise when programs are not randomly implemented and our estimation strategy that overcomes these difficulties. Section IV presents the data used to estimate the model, and Section V discusses the results. We conclude in Section VI.