Increasingly heterogeneous ages at first birth by education in Southern European and Anglo-American family-policy regimes: A seven-country comparison by birth cohort.

According to the 'reproductive polarization' hypothesis, family-policy regimes unfavourable to the combination of employment with motherhood generate greater socio-economic differentials in fertility than other regimes. This hypothesis has been tested mainly for 'liberal' Anglo-American regimes. To investigate the effects elsewhere, we compared education differentials in age at first birth among native-born women of 1950s and 1960s birth cohorts in seven countries representing three regime types. Women with low educational attainment have continued to have first births early, not only in Britain and the USA but also in Greece, Italy, and Spain. Women at all other levels of education have experienced a shift towards later first births, a shift that has been largest in Southern Europe. Unlike the educationally heterogeneous changes in age pattern at first birth seen under the Southern European and Anglo-American family-policy regimes, the changes across birth cohorts in the study's two 'universalistic' countries, Norway and France, have been educationally homogeneous.

First births by age and education in Britain, France and Norway.

Progressively later starting of childbearing has been a feature of cohort change in fertility across Europe and elsewhere over recent decades. Growing differences in the age patterns of childbearing between the Anglo-American and continental European countries, however, have also been found. The present study uses large linked-record databases in Britain, France and Norway to analyse these differences in more detail, focussing on age at entry to motherhood (first childbearing) by level of educational attainment among women born in the 1950s and in the 1960s. The shift between these two cohorts towards a later pattern of first childbearing in Britain was confined to women with secondary school qualifications and above. For women born in the 1960s, the peak age for risk of first childbearing among those with secondary school qualifications grew to be between seven and eleven years later than among women without secondary school qualifications. In France and Norway, the peak ages for risk of first childbearing shifted more uniformly across education levels between the two cohorts. For these 1950s and 1960s cohorts, improvements in women's educational levels also occurred more uniformly in France and Norway, moving more women into education categories characterised by later patterns of first childbearing.