While Medicaid is currently financed by open-ended matching, in which the federal government pays an uncapped percentage of program expenditures, there has been interest in trans-forming the financing structure to block grants, which limits federal cost-sharing to a fixed amount. To understand the implications of this reform, I measure the effect of match rates on Medicaid spending by using the variation induced by a kink in the match rate formula. I find that a percentage point increase in the federal match raises per-beneficiary spending by 3 percent. Using this estimate, I discuss the welfare impact of a block grant reform.