ECIN Replication Package for "Designing School Choice Mechanisms: A Structural Model and Demand Estimation"

Designing the markets that allocate public school seats is a crucial policy consideration. This paper compares the design of school choice mechanisms in terms of economic efficiency, fairness, and strategic behavior. We estimate demand for schools using data from a large US public school system with novel indicators of students’ levels of strategic sophistication. We interpret our results in the context of what design choices policymakers should emphasize as they design their assignment process. The mechanism comparisons support the use of the Deferred Acceptance Algorithm. Even when implemented with constrained lists, DA has little justified envy according to the standard efficiency metrics from the literature. This supports the use of DA even in settings where there are large numbers of schools and districts are inclined to constrain the list length. Further, we find important benefits of reserving a set of seats to be assigned by a pure lottery. By identifying which design choices have large effects and which do not, we offer concrete policy recommendations for school districts to best achieve their objectives in school assignment.