Pick-an-object Mechanisms

TL;DR

This paper introduces pick-an-object (PAO) mechanisms for one-sided matching markets, characterizes their implementation and equilibrium properties, and demonstrates through experiments that PAO encourages more truthful behavior than direct mechanisms.

Contribution

It proposes PAO mechanisms, characterizes their implementability and equilibrium properties, and compares their practical performance to other mechanisms through laboratory experiments.

Findings

PAO mechanisms can implement a wide range of allocation rules.

Agents behave more truthfully under PAO and OSP mechanisms in experiments.

PAO mechanisms are more effective in promoting truthful behavior than direct mechanisms.

Abstract

We introduce a new family of mechanisms for one-sided matching markets, denoted pick-an-object (PAO) mechanisms. When implementing an allocation rule via PAO, agents are asked to pick an object from individualized menus. These choices may be rejected later on, and these agents are presented with new menus. When the procedure ends, agents are assigned the last object they picked. We characterize the allocation rules that can be sequentialized by PAO mechanisms, as well as the ones that can be implemented in a robust truthful equilibrium. We justify the use of PAO as opposed to direct mechanisms by showing that its equilibrium behavior is closely related to the one in obviously strategy-proof (OSP) mechanisms, but implements commonly used rules, such as Gale-Shapley DA and top trading cycles, which are not OSP-implementable. We run laboratory experiments comparing truthful behavior when using PAO, OSP, and direct mechanisms to implement different rules. These indicate that agents are more likely to behave in line with the theoretical prediction under PAO and OSP implementations than their direct counterparts.