The impact of income on continued childbearing across educational groups. A Longitudinal register-based study of Sweden

Ernesto Silva, Lund University
Martin Dribe, Lund University

In this paper we study the impact of income on continued childbearing using longitudinal data for Sweden from 1990 to 2005. We analyze how couple income, and the share of that income that is held by women, affects the risk of giving birth, considering also differences by parity and educational attainment. Our results indicate an overall positive impact of income on continued childbearing, which should not come as a surprise considering the Swedish welfare policies affecting the opportunity costs of children. The pattern tends to be strongest for the transition to the second birth with no big differences across educational categories. On the other hand, the relationship tends to be reversed for the fourth births, but selection effects cannot be dismissed. Another finding, which might be at odds with some of the theoretical literature on family economics, is a positive impact of women’s share of household income on the probability of having a birth. A suitable explanation for this could be given by the double veto hypothesis. Assuming that every member of the couple has a veto option when it comes to childbearing plans, then, if women are more likely to want another birth, greater bargaining power would allow them to override the veto power of their partners.