We exploit local and temporal variation in the availability of public childcare for children under the age of three that induces exogenous variation in childcare attendance. We find a weak, positive average treatment effect (ATE) on maternal labor supply. The estimation of the average treatment effect is interesting – however, possibly masking important effect heterogeneity. Examining selection behavior and estimating marginal treatment effects along the distribution of observables and unobservables that drive individual treatment decisions reveal transmission channels and uncover substantial heterogeneity in marginal returns from public childcare reforms. By estimating marginal returns, we detect reverse selection on gains at the intensive margin, whereas a substantial share (40 percent) of mothers with median desire to public childcare react with increased probability to work full time. Thus, if the supply of public childcare is expanded from a modest to a more generous level of coverage, those with average resistance towards early public childcare do gain. At the extensive margin, positive selection on gains is found; however, only a small fraction of mothers with the lowest distaste for early public childcare shift from non-employment to part-time jobs.

我们利用三岁以下儿童在公共儿童保育方面的当地和时间差异，从而导致儿童保育出勤率的外源性差异。我们发现，对产妇劳动力供应的平均治疗效果（ATE）较弱。平均治疗效果的估计很有趣-但是，可能掩盖了重要效果的异质性。检查选择行为并沿着驱动个体治疗决策的可观察者和不可观察者的分布估算边缘治疗效果，揭示了传播渠道并揭示了公共育儿改革中边际回报的实质异质性。通过估算边际收益，我们可以发现密集边际收益的反向选择，而有相当一部分（40％）的母亲对公共托儿服务有中等期望，她们全职工作的可能性增加。因此，如果将公共托儿服务的覆盖范围从适度扩大到更为慷慨，那么那些对早期公共托儿服务平均抵制的人就会有所收获。在广泛的范围内，发现对收益的积极选择；然而，只有很少一部分母亲对早期公共育儿的厌恶程度最低，他们从非工作转向兼职。