Are low Danish fertility rates explained by changes in timing of births?

The most commonly used indicator of fertility, the period total fertility rate (TFR(p)), tends to underestimate actual fertility when women delay childbearing. The objective of this study was to examine to which extent fluctuations in Danish fertility rates result from changes in timing of births and, thus, whether the conventional TFR(p) is a distorted indicator of fertility quantum. In addition, we investigated whether such changes in timing explained the observed regional differences in the TFR(p) in Denmark.

The study applied age-, period-, county-, and parity-specific data from the Danish Fertility of Women and Couples Dataset, 1980-2001. We evaluated fluctuations in period fertility rates by the tempo-adjusted TFR(') - a proposed variant of the conventional TFR(p) taking period changes in timing of births into account. Tempo-effects were given by the difference between TFR(p) and TFR(') , and these period measures were compared to actual cohort fertility.

Mean age at childbearing increased with more than 3 years over the period 1980-2001 leading to considerable differences between TFR(p) and TFR(' ) . A tempo-effect of up to 0.347 children per woman was observed. Comparisons with actual cohort fertility showed consistency with the TFR('). However, tempo-adjustment did not attenuate observed regional differences.

This study indicates that the conventional TFR(p) consistently underestimates the fertility quantum in periods characterized by changes in timing of births, and that the TFR(' ) generally provides good indication of actual cohort quantum for the period in question. Considerable tempo-effects were observed within counties; however, differences between regions were not explained by changes in timing of births.