No abstract

No abstract

Gale-Shapley's deferred acceptance (henceforth DA) mechanism has emerged as a prominent candidate for placing students to public schools. While DA has desirable fairness and incentive properties, it limits the applicants' abilities to communicate their preference intensities, which entails ex-ante inefficiency when ties at school preferences are broken randomly. We propose a variant of deferred acceptance mechanism that allows students to influence how they are treated in ties. It inherits much of the desirable properties of DA but performs better in ex ante efficiency. JEL: C78, D61, D78, I21, I28

No abstract

Despite its widespread use, the Boston mechanism has been criticized for its poor incentive and welfare performances compared to the Gale-Shapley deferred acceptance algorithm (DA). By contrast, when students have the same ordinal preferences and schools have no priorities, we find that the Boston mechanism Pareto dominates the DA in ex ante welfare, that it may not harm but rather benefit participants who may not strategize well, and that, in the presence of school priorities, the Boston mechanism also tends to facilitate greater access than the DA to good schools for those lacking priorities at those schools. (JEL D82, I21, I28)