Competing by Restricting Choice: The Case of Matching Platforms

Abstract

We show that a two-sided matching platform can successfully compete by limiting the number of choices it offers to its customers, while charging higher prices than platforms with unrestricted choice. We develop a stylized model of online dating where agents with different outside options match based on how much they like each other. Starting from these micro-foundations, we derive the strength and direction of indirect network effects, and show that increasing the number of potential matches has a positive effect due to larger choice, but also a negative effect due to competition between agents on the same side. Agents resolve the trade-off between these competing effects differently, depending on their outside options. For agents with high outside options, the choice effect is stronger than the competition effect, leading them to prefer an unrestricted-choice platform. The opposite is the case for agents with low outside options, who then have higher willingness to pay for a platform restricting choice, as it also restricts the choice set of their potential matches. Moreover, since only agents with low outside options self-select into the restricted choice platform, the competition effect is mitigated further. This allows multiple platforms offering different number of choices to coexist without the market tipping.

The Secret to Finding a Match: A Field Experiment on Choice Capacity Design in an Online Dating Platform

“…Our study makes several theoretical and practical contributions. First, our study enriches the literature on online matching market design (Pizzato et al 2010, Horton 2017, Halaburda et al 2018, Basu et al 2019, Shi and Zhang 2019, Li and Netessine 2020 by identifying the causal effects of choice capacity. Although designing a matching market to increase market performance is of interest to both academics and practitioners, identifying the causal effects of key design elements poses a methodological challenge because of the difficulty of controlling for network externalities and the resulting endogeneity in choice capacity (Aral and Walker 2011).…”

“…Alternatively, users might anticipate that an increased choice capacity increases competition and respond by becoming less selective (i.e., select less attractive candidates), which could increase the conversion rate. Following Halaburda et al (2018), we name these two distinct effects the choice effect and the competition effect, respectively. If users become less selective as their choice capacity increases, the number of matches (matching outcomes) further increases along with the increase in choices.…”

Random Versus Directed Search for Scarce Resources

“…While not directly related to this paper, the literature on market and search platform design has pointed out that limiting choices (Halaburda et al, 2017), limiting the number of potential matches (Arnosti et al, 2015), limiting preference signaling (Coles et al, 2013), or preventing one side of the market from initiating contact with potential matches (Kanoria and Saban, 2017) could all be welfare improving.…”

Platform competition: Betfair and the UK market for sports betting

“…Hałaburda, Piskorski, and Yildirim () explain how a two‐sided matching platform can successfully compete by limiting the number of choices it offers to its customers, even while charging higher prices than platforms with an unrestricted choice. Using a stylized model of online dating, the authors show that increasing the number of potential matches has a positive effect (the choice effect), since agents are more likely to find desirable candidates, but also a negative effect (the competition effect), since their probability of being accepted by their preferred candidate is lowered.…”