How much do mandatory minimums matter?

Abstract Mandatory minimum (MM) sentencing is frequently identified as a driver of long sentences among federal drug offenders, but estimating its causal impact is complicated by prosecutorial discretion in charging. First, I show evidence of selective manipulation of MM eligibility, with prosecutors disproportionately choosing minority defendants for eligibility relative to comparable white defendants. This selection behavior creates bias in naive estimates of MM effects on sentence length. To recover the causal impact of MM eligibility, I use a regression discontinuity design with extrapolation that disentangles statutory effects from prosecutorial selection. I find that eligibility increases sentence length by approximately 10 months (14 percent) uniformly across case types, with no discernible heterogeneity across race. Together, these results indicate that racial disparities in MM sentencing arise primarily from prosecutorial charging decisions rather than from features of the statute itself.