Many students enroll in less selective colleges than they are qualified to attend, despite low graduation rates at these institutions. Some scholars have argued that qualified students should enroll in the most selective colleges because they have greater resources to support student success. However, selective college attendance is endogenous, so student outcomes could be due to individual ability, not institutional characteristics. Previous work on college selectivity has focused on the earnings effects of attending elite private universities, overlooking both college graduation impacts and the public institutions that educate most students. I estimate the effect of selective colleges on the probability of bachelor's degree completion using a restricted-access national dataset and an instrumental variables approach to address the endogeneity of college choice. I find that a 100-point increase in the average SAT score for admitted students is associated with an increase in the probability of graduation by 13 percentage points. In addition, I find suggestive evidence that enrolling in a selective public college has a positive effect on degree completion. The results are robust to a series of sensitivity tests and alternate specifications. The findings suggest strong benefits to enrolling in the most selective colleges that students are qualified to attend and have important implications for decisions to pursue postsecondary education in the face of high student loan debt.