Does parental leave affect fertility and return-to-work? Evidence from a true natural experiment

We study the causal effects of changes in parental leave provisions on fertility and return-to-work behavior. We exploit a policy change that took place in 1990 in Austria which extended the maximum duration of parental leave from the child's first to the child's second birthday. As parental leave benefits can be automatically renewed when a new mother is still on leave from a previous child, this created a strong incentive to bunch the time off work in case of multiple planned children and/or to increase fertility. We study the quantitative effect of this incentive using an empirical strategy which resembles a true experimental set-up very closely. In particular, assignment to treatment is random and treated and controls face (almost) identical environmental conditions. We find that treated mothers have a 4.9 percentage points (or 15 percent) higher probability to get an additional child within the following three years; and a 3.9 percentage points higher probability in the following ten years. This suggests that not only the timing but also the number of children were affected by the policy change. We also find that parental leave rules have a strong effect on mothers' return-to-work behavior. Per additional months of maximum parental leave duration, mothers' time off work is reduced by 0.4 to 0.5 months. The effects of a subsequent policy change in 1996 when maximum parental leave duration was reduced from the child's second birthday to the date when the child became 18 months old brought about no change in fertility behavior, but a labor supply effect that is comparable in magnitude to the one generated by the 1990 policy change. This can be rationalized by the incentives created through automatic benefit renewal.