This paper studies the effects of a large-scale urban school construction program on student test scores, student mobility, and residential housing markets. The paper takes advantage of a natural experiment in which a 15-year, $1.4 billion school construction program – one of the largest per-capita construction programs in the U.S. over the period – to isolate the causal effect of capital investments on student performance. Our empirical strategy uses the fact that occupancy dates varied widely across schools, with the first school completed in 1998 and the last slated for completion in 2014.  Specifically, we use a difference-in-differences comparison of test scores, housing prices, and public school attendance in neighborhoods on different construction schedules. For our test score analysis, we combine this approach with panel data methods (e.g., “fixed-effects”) to address concerns about students (parents) migrating into newly built schools. Our findings suggest that six years after occupancy, school construction increases reading scores by 0.15 standard deviations relative to reading scores prior to building occupancy. Further, we find evidence that school construction raises housing prices in affected neighborhoods by about 10 percent and that new construction leads to increased public school enrollment in affiliated catchment areas. Our findings suggest that infrastructure improvements can be thought of as part of the broader school reform toolkit in urban districts and that increased property tax revenues might defray a portion of the cost to districts.