Educational differences in the timing of childbearing in Hungary

Ildikó Husz, Hungarian Academy of Sciences

This paper investigates the relationship between level of education and fertility. First, the way educational level is related to childbirth timing and fertility levels is examined. Also of interest was whether this relationship has changed since the early 1990s, i.e. during the major economic and social restructuring over the last fifteen years. It is well known that period fertility has greatly decreased on the national scale in this period, mainly owing to postponed births. However, we have hitherto had no information as to whether this change has affected every educational-level group equally, and an answer to this question could aid us in elucidating the reasons behind the postponement. Analysis of the live birth data yielded the following results. Around 1990, period fertility hardly varied by level of education. The highest level, as expected, was found in the least-qualified group. The observed fertility rate in other educational level groups was a few tenths lower, and showed very little spread. The differences were certainly not large enough to seen any empirical confirmation of the human capital theory. A decade later, the situation had changed fundamentally. The tempo effect in the national level of observed fertility, almost absent around 1990, has now become highly significant. The demographic behaviour of the least-qualified group now contrasts even more sharply from the others: they still have a relatively high fertility rate and a very young age at birth. Whereas mothers who had graduated from skilled-labour schools reacted to the economic recession by postponement of childbirth, the fertility decrease among those with secondary and tertiary qualifications displays a quantum-based effect in addition to the tempo effect. This is particularly striking for those with school leaving certificates, whose period fertility was particularly low around 2000 – only two thirds of the figure ten years before.

  See paper

Presented in Poster Session 1