Evaluating the effectiveness of in-work tax credits

One of the principle aims of the Working Families’ Tax Credit in the UK was to increase the participation of single mothers in work. The difference-in-differences methodology that is typically used to evaluate tax credit policies compares single mothers with single women without children. However, the characteristics of these groups are very different, and changes over time in relative covariates are likely to violate key identifying assumptions. We find that when we control for differential trends between women with and without children, the employment effect of the policy falls significantly. Moreover, closer inspection shows that while there was an effect on increasing the likelihood to work full-time (30 h or more), there was no effect on reducing the likelihood to be inactive. Looking closely at important covariates before and after the policy introduction, we can see sizeable changes in the relative returns to employment between the treatment and control groups, making it difficult to assess the policy effectiveness.