More than 50 studies have examined the programmatic incumbent support hypothesis, which posits that once enacted, programmatic policies increase electoral support for the incumbent. Despite the careful attention to causal inference in this work, empirical findings have been strikingly inconsistent. We make the case that these inconsistent results are likely explained by incumbents' strategic responses to the enactment of a programmatic policy. Specifically, incumbents have good reasons to distribute different amounts of non-programmatic goods to voters who do and do not receive a programmatic policy. To examine this conjecture, we turn to the case of Japan, where municipalities receive allocations of non-programmatic goods and vary in their eligibility for a programmatic policy (a snow subsidy) according to plausibly exogenous factors. Using a geographic regression discontinuity design, we find that municipalities receiving the programmatic policy receive systematically more non-programmatic goods than municipalities that do not.
