Swedish fertility swings and public expenditure for children

Thomas Lindh, Institute for Futures Studies
Ying Hong, Swedish Institute for Futures Studies

In the 1930s Swedish birth rates had fallen to levels close to the death rates and in public discourse this was perceived as a major social and national crisis, spurring a range of social policy reforms. While total fertility rates in Sweden have varied over large spans the completed cohort fertility rates are almost constant around 2 children per woman for women born in the 20th century. Using unique data 1930-1997 on public expenditure for schools, child allowances and child care we estimate age-specific fertility for broad age groups as a function of these variables. The results indicate that the age group 25-29 is most sensitive to variations in this public expenditure thus providing an explanation of the swings in period fertility in terms of policy induced tempo variation. School expenditure is negatively correlated to fertility while child care and child allowance is positively correlated. This pattern is consistent with a quantity-quality trade-off by the parents. To check the predictive power of the model we use data from 1998-2007 and get an excellent prediction of the fertility turn-around after 1999 for all age groups except 35 and above where we tend to under-predict at the 10-year horizon. Still it raises the hope that the fertility effect of given policy mixes can be predicted with reasonable certainty.