An Econometric Analysis of the Number and Timing of Births: The Case of Vietnam
This thesis represents an intensive econometric study of two important aspects of reproductive behavior -- namely the level of fertility and the intervals between births. The analysis is economics-based, but the standard microeconomic framework is extended to take into account the effects of social factors on reproduction. The estimation is based on data from the 1988 Vietnam Demographic and Health Survey.
The initial analysis of the determinants of fertility (defined as the number of children ever born per woman) is conducted within a static framework, and three alternative statistical models (OLS, Poisson, and ordered-Iogit) are estimated and the results compared. Fertility decision-making is then modelled as a dynamic sequential-response process, with particular attention to the effects of sex preference and child mortality on subsequent fertility. The analysis of birth intervals is based on the estimation of both non-parametric life table methods and the (semi-parametric) proportional hazard model.
Among the main results of the study is the finding that individual and social characteristics generally have little effect on reproductive behavior in the first ten years or so of reproductive life, apparently reflecting the strong desire of typical couples to have at least two children. Such characteristics of the wife as education, knowledge of contraceptives, the industrial sector in which she works, and whether she lives in an urban or rural area are found to have statistically significant effects on both the level and timing of completed reproductive behavior, although the magnitudes of the effects are generally moderate. Certain characteristics of the husband (notably, his educational attainment and the industrial sector in which he works) also affect the level and timing of births. However, no evidence is found to support an opportunity-cost effect of either the wife's or the husband's education on fertility. That finding is readily interpreted in the Vietnamese context and may explain why the estimated effects of education are considerably smaller in Vietnam than in other, more market-oriented countries. Finally, in Vietnam the preference for sons is strong and largely independent of socioeconomic variables: furthermore, the evidence indicates that parents, whatever their levels of education or other characteristics, wish to have one living son, not necessarily more.