Difference-in-Differences

Difference-in-differences (DiD) estimates a causal effect by comparing the before-and-after change in a treated group against the before-and-after change in an untreated comparison group. Subtracting one change from the other cancels out fixed differences between the groups and shared trends over time.

Its credibility hinges on the parallel-trends assumption: that the two groups would have moved in step had the treatment never happened. You can never fully verify this, but checking that the groups tracked each other before the treatment makes it more believable. DiD is a workhorse for evaluating policies that hit one region or group but not another.