A growing literature on inference in difference-in-differences (DiD) designs with grouped errors has been pessimistic about obtaining hypothesis tests of the correct size, particularly with few groups. We provide Monte Carlo evidence for three points: (i) it is possible to obtain tests of the...

Time-limited in-work credits are cheaper, and more targeted, than conventional in-work credits, but are thought to have small to zero long-term impacts. We study two time-limited in-work credits introduced in the mid-2000s in the UK and find they reduced welfare participation and increased...