The Proximate Determinants of Fertility: An Estimation Approach

José Antonio Ortega , Universidad de Salamanca, Spain
David A. Sánchez Páez, Université Catholique de Louvain

The proximate determinants of fertility, originally formulated by Bongaarts(1978) with later contributions by Stover(1998) and Bongaarts(2015), remains the prevalent framework for studying fertility in developing countries. It highlights the role of contraception and abortion in the intentional reduction of fertility, and the impacts of sexual exposure and postpartum infecundability. However, the contribution of the different factors is not estimated from the data. It is imputed based on aggregate indicators for each determinant: the impact of contraception is based on overall contraceptive prevalence and an external imputation of effectiveness by method based on biomedical literature, mostly from the US in the 1970s. We propose an alternative estimation approach to the proximate determinants where the contribution of each factor is directly measured instead of imputed. For instance, the relative protection of contraception against the risk of pregnancy is measured by comparing pregnancy rates of fecund women using and not using contraception. The role of pregnancy termination, including both induced and spontaneous terminations, is estimated by the proportion of pregnancies that do not end up in a live birth. We estimate the role of the proximate determinants based on reproductive calendar data from Demographic and Health Surveys. We compare the decomposition obtained with the decomposition in Bongaarts(2015). The differences highlight in what contexts contraceptives are being used less effectively. While in this initial study we focus on general fertility, the approach can be applied to different population subgroups by age, union status or education in order to better target reproductive health programs.

See extended abstract

 Presented in Session P3. Poster Session Migration, Economics, Environment, Methods, History and Policy