Lester Chan

I am an Assistant Professor in the School of Economics & Gregory and Paula Chow Center for Economic Research at Xiamen University. I received my Ph.D. in Economics from Boston University in 2021.

My research fields are microeconomic theory and industrial organization, focusing on contract theory, business economics, platforms, and potential games.

You can find my CV here and my research statement here.

You can contact me at lesterchantl[at]gmail.com.


Quality Strategies in Network Markets, forthcoming at Management Science [WP]

This paper studies network market problems in which firm(s)/platform(s) sets quality in addition to price. A well-established result in the network economics literature is that a profit-maximizing firm concerns only how quality is valued by the marginal consumer but not by inframarginal consumers, aka the Spence effect/distortion. For markets with strong network effects under which multiple market-tipping equilibria exist, I show that the validity of the above result depends on the choice of the equilibrium selection criterion. Precisely, I show that all criteria commonly used in this literature give rise to the Spence effect, whereas the well-justified risk dominance criterion in game theory and its generalizations do not. Novel quality strategies are derived based on the latter criteria.

Divide and Conquer in Two-Sided Markets: A Potential-Game Approach, The RAND Journal of Economics, 52(4): 839-858 (2021). [WP]

Network effects typically generate multiple equilibria in two-sided markets. To overcome the methodological challenge of selecting an appropriate equilibrium, this article shows that many two-sided market models are weighted potential games, and therefore a refinement of Nash equilibrium justified by many theoretical and experimental studies, potential maximization, pins down an equilibrium. Under potential maximization, platforms often subsidize one side and charge the other, i.e., divide and conquer. The primary determinant of which side to subsidize/monetize is cross-side network effects. This divide-and-conquer strategy implies that platforms are often designed to favor the money side much more than the subsidy side.

Working Papers

Weight-Ranked Divide-and-Conquer Contracts, revise and resubmit at Theoretical Economics

This paper studies bilateral contracting between one principal and multiple agents. Multiple equilibria arise due to agents' strategic interactions. In general, the principal's optimal contracting scheme varies with the choice of equilibrium selection criterion or implementation requirement. Nevertheless, for a large class of models where agents' payoffs constitute a weighted potential game, I show that one contracting scheme is optimal for a large class of equilibrium selection criteria and implementation requirements. This scheme ranks agents in ascending order of their weights in the weighted potential game and induces them to accept their offers in a dominance-solvable way, starting from the first agent. With the general results, I derive robust predictions and policy guidance for a wide variety of applications, including networks and pure/impure public goods/bads.

Algorithm-Based Platform Envelopment, (with Liang Chen and Zhou Zhou) revise and resubmit at Academy of Management Review [UTD24]

Platform research has much focused on network effects as the key mechanism for value creation, and on that basis, unveiled such strategies as envelopment in explaining how a large entrant extends market power to an adjacent platform market. By contrast, we follow on from a previous debate on “data network effects” and tease apart networks effects and data-driven learning. Data-driven learning refers to a platform deriving from its data matching algorithms which are complementary to installed base in creating platform value. Our model demonstrates how an entrant with a superior algorithm may outcompete an incumbent possessing an installed base advantage, a strategy we term “algorithm envelopment.”

This paper introduces signed weighted potential games and potential minimaximization. The former generalize weighted potential games by allowing negative weights for players, and they are strategically equivalent to two-team zero-sum games. The latter refines correlated equilibrium by selecting the global saddle point of the potential function, and it reduces to potential maximization if all players have positive weights. By exploiting these two generalizations, I generalize the main result of Chan (2022) on multi-agent contracting. Precisely, in a more general contracting environment, weight-ranked divide-and-conquer contracts, in a generalized form, remain optimal for the principal for a large class of equilibrium selection criteria.

A principal incentivizes a team of agents to exert efforts on a project by offering them bonuses upon project success. In addition to bonuses she also designs the project's technology, which maps agents' effort profiles to project success probabilities. Multiple equilibria typically arise due to agents' strategic interactions. Unlike the case where the principal only sets bonuses, I show that it is possible to derive a joint bonus-technology design that is optimal for her for a large class of equilibrium selection criteria/implementation requirements. An implication is that this joint design achieves a good balance between optimality and robustness for her. The corresponding technology exhibits strategic independence among agents' efforts, eliminating their coordination concerns. In addition, I show that any technology exhibiting strategic substitutability among efforts is suboptimal for the principal for all standard implementation requirements.