🤖 AI Summary
This paper studies stochastic object allocation with capacity constraints, focusing on the equivalence of ex-post efficiency notions under lottery-based rules. Methodologically, it integrates mechanism design, Maskin monotonicity analysis, and stochastic dominance comparisons. The main contribution is the first rigorous proof that, under the axioms of ex-post non-wastefulness and probabilistic monotonicity, ex-post pairwise efficiency is equivalent to ex-post Pareto efficiency. This result unifies and strengthens the axiomatic characterizations of classical mechanisms—including Random Serial Dictatorship, Trading Cycles, and Hierarchical Exchange—by establishing a common foundational efficiency benchmark. Moreover, it provides a more concise and expressive theoretical foundation for analyzing the fairness–efficiency trade-off in stochastic allocation, enhancing both conceptual clarity and analytical tractability in the design and evaluation of randomized matching mechanisms.
📝 Abstract
We consider object allocation problems with capacities (see, e.g., Abdulkadiroglu and Sonmez, 1998; Basteck, 2025) where objects have to be assigned to agents. We show that if a lottery rule satisfies ex-post non-wastefulness and probabilistic (Maskin) monotonicity, then ex-post pairwise efficiency is equivalent to ex-post Pareto efficiency. This result allows for a strengthening of various existing characterization results, both for lottery rules and deterministic rules, by replacing (ex-post) Pareto efficiency with (ex-post) pairwise efficiency, e.g., for characterizations of the Random Serial Dictatorship rule (Basteck, 2025), Trading Cycles rules (Pycia and Unver, 2017), and Hierarchical Exchange rules (Papai, 2000).