The ``Boston'' school-choice mechanism: an axiomatic approach (Q2447152)

From MaRDI portal
Revision as of 10:13, 8 July 2024 by ReferenceBot (talk | contribs) (‎Changed an Item)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
scientific article
Language Label Description Also known as
English
The ``Boston'' school-choice mechanism: an axiomatic approach
scientific article

    Statements

    The ``Boston'' school-choice mechanism: an axiomatic approach (English)
    0 references
    0 references
    0 references
    24 April 2014
    0 references
    The ``Boston'' school-choice mechanism is a student-placement procedure in which students submit their preference lists to a central clearinghouse, and the clearinghouse runs the following algorithm to match as many students to their stated preferred schools as possible, subject to a school-dependent ranking of students: seats of each school are allocated to students who rank that school first, then to those who rank it second if there are any seats left, and so on. Schools can have quotas, students can be unqualified for some schools, and students can rank the ``empty placement'', meaning no admission to any school, as their preferred outcome. The authors give an axiomatic characterization of the ``Boston'' school-choice mechanism. The axioms are the following: {\parindent=6mm \begin{itemize}\item[--] Consistency: A placement mechanism is \textit{consistent} if whenever the school assignment of a student is fixed at the mechanism's outcome, by rerunning the mechanism for the remaining students and school seats, all remaining students are assigned the same schools. \item[--] Favoring higher ranks: A placement mechanism \textit{favors higher ranks} if whenever a student \(S\) is assigned a less preferred school than school \(C\), then the quota of \(C\) is filled with students who rank \(C\) at least as high as student \(S\). \item[--] Population monotonicity: A placement mechanism is \textit{population monotonic} if whenever a student is removed from the problem, the mechanism's outcome makes every remaining student (not necessarily strictly) better off than in the original outcome. \item[--] Resource monotonicity: A placement mechanism is \textit{resource monotonic} if whenever the supply of school seats increases, the mechanism's outcome makes each student (not necessarily strictly) better off than in the original outcome. \item[--] Individual rationality: A placement mechanism is \textit{individually rational} if for every preference profile and quota, every student favors its placement to the empty placement. \item[--] Rank-respecting invariance: A placement mechanism satisfies \textit{rank-respecting invariance} if the resulting placement is unchanged when students promote the rankings of their original assignments in such a way that this does not increase the competition for schools assigned to others. \end{itemize}} The authors obtain that a placement mechanism is the Boston mechanism induced by some priority profile if and only if it favors higher ranks and satisfies consistency, resource monotonicity, and rank-respecting invariance. An analogue of this result is also given where the quota of each unit is fixed to 1: in that case a placement mechanism is the Boston mechanism induced by some priority profile if and only if it favors higher ranks and satisfies individual rationality, population monotonicity, and rank-respecting invariance. In addition, the authors give examples of placement mechanisms showing that no axioms can be left out from the above characterizations. They also argue that their results can be modified to characterize a subclass of the Boston mechanisms in which every student is acceptable to every school.
    0 references
    Boston school-choice mechanism
    0 references
    matching
    0 references
    mechanism design
    0 references
    market design
    0 references
    Pareto efficiency
    0 references
    0 references
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references