A pooled time series analysis on the relation between fertility and key fertility-related demographic behaviour across space and time

Henriette Engelhardt, Austrian Academy of Sciences
Alexia Fuernkranz-Prskawetz, Vienna Institute of Demography
Marija Mamolo, Vienna Institute of Demography and Università di Roma "La Sapienza"

Cross-country differentials in period fertility rates are commonly explained by cross-country differentials in socio-economic, demographic and institutional settings. Recent studies have indicated "... a reversal of many well-known relationships that have been used to explain cross-country differences in fertility patterns" (Billari and Kohler 2004). In this paper we focus on fertility-related demographic behaviour summarised through period indicators, such as the Mean Age at First Birth (MAFB), the Total First Marriage Rate (TFMR), the Mean Age at First Marriage (MAFM), the Total Divorce Rate (TDR) and the percentage of extramarital births, and investigate whether unmeasured country and time heterogeneity in those indicators can explain the change in the cross-country correlation coefficients. In particular, pooled time series analysis is applied to explain the reversal of the cross-country correlation between period fertility and these key fertility-related demographic indicators. We use data from the Council of Europe and focus on the period 1960-2002. Our empirical findings reveal substantial differences across time periods and across countries in the effects of these demographic indicators on fertility. The main effect of the indicators considered is negative in all the regressions. Nevertheless, when controlling for time heterogeneity, we observe a progressively less negative effect of these indicators on fertility. These findings support the conclusion that at the macro-level the postponement of key fertility related demographic events might have a declining negative impact on fertility; in particular in those countries that have adjusted their social, family and labour market policies accordingly.