We study the role of local institutions and regulations—school boundaries, school transportation provision, and zoning restrictions—in determining inequalities of educational opportunities for children. Motivated by our empirical findings on how the demand for both neighborhoods and schools responds to quasi-experimental variation in school quality and transportation, we build and estimate a spatial equilibrium model of residential sorting and school choice. We validate the model with our empirical quasi-experimental findings as well as with experimental estimates from an influential voucher program. We find that the evaluation of both people-based and place-based policies heavily hinges on spatial equilibrium effects. Abstracting from those would lead to either overestimating (voucher) or overturning (school choice expansion) the impact of these policies on the inequality of educational access.