Cash Transfers and Fertility: How the Introduction and Cancellation of a Child Benefit Affected Births and Abortions

Recognition Program

Open Access       

Authors: Libertad González and Sofia Trommlerová

Journal of Human Resources, February, 2021

We study the impact of a universal child benefit on fertility, identifying separately the effects driven by conceptions and those by abortions, and analyzing the potentially asymmetric impact of the benefit’s introduction and its later cancellation. We focus on a generous lump-sum maternity allowance that was introduced in Spain in 2007 and then eliminated in 2010. Using administrative, population-level data, we create a panel data set of the 50 Spanish provinces, with monthly data on birth rates and weekly data on abortion rates between 2000 and 2017. Our identification is based on the timing of the introduction and cancellation of the policy (both announcement and implementation), from which we infer when the changes in births and abortions can be expected. We find that the introduction of the policy led to a 3% increase in birth rates, due to both a decrease in abortions and an increase in conceptions. The announcement of the cancellation led to a transitory increase in birth rates just before the benefit termination was implemented, driven by a short-term decrease in abortions. The actual cancellation then led to a 6% decline in birth rates. A heterogeneity analysis suggests that the positive fertility effect of the benefit’s introduction was driven by high-skilled parents, while the negative impact of the cancellation was larger among low-skilled and out-of-labor-force parents, and in poorer regions and areas that were more affected by the 2008 recession. We also find suggestive evidence that the child benefit had both a timing (“tempo”) effect, such that some women had children earlier, and a level (“quantum”) effect, where some women had more children than they would have had otherwise.

This paper originally appeared as Barcelona School of Economics Working Paper 1153
This paper is acknowledged by the Barcelona GSE Research Recognition Program