When is the Boston Mechanism Dominance-Solvable?∗†

It is well known that the widely-used Boston mechanism for allocating students to schools is not strategy-proof. An observation shows that relaxing the solution concept to dominance solvability could allow the Boston mechanism to achieve a desirable outcome. I define the acyclic priority structure of schools and find that the Boston mechanism is dominancesolvable if and only if the priority structure is acyclic. Moreover, when the priority structure is acyclic, the dominance-solvable outcome is stable and Pareto efficient. However, acyclic priority structure is restrictive and thus is difficult to achieve in reality. JEL: C72, C78, D78, I20