We examine how hospital treatment intensity is affected by an exogenous change in average reimbursement for an admission. Theory predicts that treatment intensity would be most affected for highly profitable services but unaffected for unprofitable services. We use Medicare inpatient data from 11 states for 16 disease categories that vary in the generosity of reimbursement to test this prediction. Using the coefficients from quantile regressions, we calculate a difference-in-difference estimate of the effect of the Balanced Budget Act (BBA) of 1998, comparing the pre- and post-BBA change in treatment intensity at high Medicare share hospitals to low Medicare share hospitals. We find that not-for-profit hospitals cut treatment intensity at the 50th, 75th, and 95th quantiles only for generously reimbursed services. Intensity at the 25th percentile was unaffected, regardless of generosity. We did not measure a statistically significant response at for-profit or public hospitals to the BBA.