The causal role of reduced child mortality on contemporary fertility transitions

Abstract

To what extent does reduced child mortality explain the fertility transition in developing countries? Is decreasing child mortality a prerequisite-a necessary condition-for decreasing fertility? May decreasing child mortality trigger by itself-as a sufficient cause-the fertility transition? Answering these questions is important to understand the dramatic demographic changes currently underway in developing countries and to guide population and health policies. If improving child survival is a precondition for birth control, family planning programs in the least developed regions are unlikely to succeed, especially if these programs have a vertical organisation independent of child health interventions. In turn, if reducing child mortality is a sufficient condition, family planning programs may be somewhat superfluous and there would be grounds for the statement that "development is the best contraceptive."

A book published about twenty years ago (Preston, 1978) is a landmark for the study of the effect of child mortality on fertility. Probably because of the lack of appropriate data at that time, this book, however, did not study the relationship at the aggregate level in developing countries. This article aims at filling the gap by examining country-level rates at two points in time and by analysing, in a multivariate fashion, the determinants of the fertility transition onset and pace in small geographic units in Costa Rica-a less developed country with a relatively long history of both child mortality and fertility decline.

A strong association between child mortality and fertility is a well documented fact in the literature. Countries with low infant mortality almost always have low birth rates (Heer 1966, Berelson et al. 1978). Couples that have experienced child losses are, in turn, less likely to use contraception, tend to have more children, and have shorter birth intervals (Taylor et al. 1976). However, this well documented association is neither proof of causation nor says anything about the causation direction. The observed association may have three closely linked origins: (1) The sharing by both child mortality and fertility of a common set of determinants, such as mother's education, access to health services, breastfeeding practices, and less observable traits such as a preference for high-quality children or a less fatalistic outlook at life (Hanson et al. 1994). (2) The fact that lower fertility may reduce child deaths by, among other reasons, lessening the maternal depletion associated to pregnancies and lactation (Trussell & Pebley 1984), diminishing sibling competition for scarce family resources and maternal care including breastfeeding (Pebley & Millman 1986), and decreasing the rates of transmission of infections in child-crowded environments (Blacker 1987, Haaga 1989). (3) The third possibilitythe one addressed in this articleis that the direction of causation runs from child survival to fertility. Although disentangling these three type of causal links is an impossible task with the data available, this article hopes that by statistically controlling the effect of third variables and by paying attention to the sequence of events over time, it will be possible to reach some conclusions about the third causal link, i.e. about the role of child mortality on contemporary fertility transitions.