Fertility evolution in Belgium and France during the 20th century. Filling the statistic gap with retrospective data

Thierry Eggerickx, Université Catholique de Louvain
Sandra Bree, Université Catholique de Louvain
Rafael Costa, Université Catholique de Louvain
Jean-Paul Sanderson, Université Catholique de Louvain

Whereas the onset of fertility decline since the 19th century has been a popular topic in European historical demography, much less attention has been given to fertility changes in the early 20th century. In fact, changes in fertility before 1940 and between 1950 and 1960 remain largely unexplored due to the lack of appropriate data. On the one hand, the use of individual data from public archives which are less than 100 years old (thus covering the 20th century) are prescribed by law. On the other hand, aggregate data are rather poor, as they are mostly available in the form of simple cross-sectional fertility indicators at the district level and only for the periods surrounding censuses (i.e., every 5 or 10 years). These data are not suitable for detailed studies of fertility evolution in the 20th century. The lack of chronological precision, for instance, makes it impossible to apprehend the impact of wars and crisis on reproductive behaviour. Likewise, cross-sectional indicators may hide adaptation strategies to these crises, such as postponement or birth spacing. In this context, data from recent censuses in Belgium and France offer new perspectives to the study of fertility in the 20th century. Using a retrospective approach, census data can be used to reconstruct the reproductive histories of female cohorts born since the end of the 19th century. The objective of this study is to examine the reliability of a retrospective approach of fertility using Belgian and French censuses. Several tests are run to evaluate the risks of biases in this approach, and assess its potential to fill the 20th century gap in fertility studies.