We examine the effects of the Kalamazoo Promise, one of the earliest and most generous place-based college scholarships, on college enrollment, persistence, and completion. We employ two approaches for identification. First, because the value of the scholarship varies based on past enrollment decisions, which are plausibly exogenous because of the surprise announcement of the program, we use an inverse-probability-weighted differences-in-differences approach that compares (similar) eligible and non-eligible students before and after the program. Second, we use propensity scores to match students from the Kalamazoo school district with similar students from neighboring school districts for whom we create pseudo-eligibility under the Promise rules. While we explore standard heterogeneous treatments by gender, ethnicity, and socioeconomic status, our administrative data allow us to further look at outcomes by past attendance, high school grades, and standardized test scores. In so doing, we offer a complete look into how the Kalamazoo Promise affected both the distribution of high school students going to (and persisting in) college as well as the conditional and unconditional distributions of colleges attended. We discuss the implications of our results on the possibility for place-based programs to promote educational opportunity and alleviate social inequality.