Peng Shi

Online platforms that match customers with suitable service providers utilize a wide variety of matchmaking strategies: some create a searchable directory of one side of the market (i.e., Airbnb, Google Local Services); some allow both sides of the market to search and initiate contact (i.e., Care.com, Upwork); others implement centralized matching (i.e., Amazon Home Services, TaskRabbit). This paper compares these strategies in terms of their efficiency of matchmaking, as proxied by the amount of communication needed to facilitate a good market outcome. We find that the relative performance of these strategies is driven by whether the preferences of agents on each side of the market is easy to describe or satisfy. ``Easy to describe'' means that the preferences can be readily captured in a short questionnaire, and ``easy to satisfy'' means that an agent has high preferences for many potential partners. For markets with suitable characteristics, each of the above matchmaking strategies can provide near-optimal performance guarantees according to an analysis based on information theory. The analysis provides prescriptive insights for online platforms.

This paper investigates the prediction accuracy of discrete choice models of school demand, using a policy reform in Boston that altered where applicants can apply under school choice. We find that the discrete choice models do not consistently outperform a much simpler heuristic, but their inconsistent performance largely arises from prediction errors in applicant characteristics, which are auxiliary inputs. Once we condition on the correct inputs, the discrete choice models consistently outperform, and their accuracy does not significantly improve upon refitting using post-reform data, suggesting that the choice models capture stable components of the preference distribution across policy regimes.

This paper develops a tractable methodology for designing an optimal priority system for assigning agents to heterogeneous items while accounting for agents’ choice behavior. The space of mechanisms being optimized includes deferred acceptance and top trading cycles as special cases. In contrast to previous literature, I treat the inputs to these mechanisms, namely the priority distribution of agents and quotas of items, as parameters to be optimized. The methodology is based on analyzing large market models of one-sided matching using techniques from revenue management and solving a certain assortment planning problem whose objective is social welfare. I apply the methodology to school choice and show that restricting choices may be beneficial to student welfare. Moreover, I compute optimized choice sets and priorities for elementary school choice in Boston.

We study a setting in which dynamically arriving items are assigned to waiting agents, who have heterogeneous values for distinct items and heterogeneous outside options. An ideal match would both target items to agents with the worst outside options and match them to items for which they have high value. Our first finding is that two common approaches—using independent lotteries for each item and using a waitlist in which agents lose priority when they reject an offer—lead to identical outcomes in equilibrium. Both approaches encourage agents to accept items that are marginal fits. We show that the quality of the match can be improved by using a common lottery for all items. If participation costs are negligible, a common lottery is equivalent to several other mechanisms, such as limiting participants to a single lottery, using a waitlist in which offers can be rejected without punishment, or using artificial currency. However, when there are many agents with low need, there is an unavoidable trade-off between matching and targeting. In this case, utilitarian welfare may be maximized by focusing on good matching (if the outside option distribution is light tailed) or good targeting (if it is heavy tailed). Using a common lottery achieves near-optimal matching, whereas introducing participation costs achieves near-optimal targeting.