Geoffroy de Clippel

Professor of Economics, Brown University

I am an economic theorist, with an interest in experiments too. My research covers topics in mechanism design, bounded rationality, bargaining, distributive justice, and cooperative games.

Working Papers:

Fairness Through the Lens of Cooperative Game Theory: an Experimental Approach (with K. Rozen) Version: May 2019, R&R AEJ Micro

Idea: This paper experimentally investigates cooperative game theory from a normative perspective. Axioms and solution concepts from cooperative game theory provide valuable insights into the data.

On Selecting the Right Agent (with K. Eliaz, D. Fershtman, and K. Rozen) Version: July 2019, R&R TE

Idea: Each period, a principal must assign one of two agents to some task. Profit is stochastically higher when the agent is qualified for the task, but the principal cannot observe qualification. Her only decision is which of the two agents to assign, if any, given the public history of selections and profits. She cannot commit to any rule. While she maximizes expected discounted profits, each agent maximizes his expected discounted selection probabilities. We fully characterize when the principal's first-best payoff is attainable in equilibrium, and identify a simple strategy profile achieving this first-best whenever feasible. We propose a new refinement for dynamic mechanisms (without transfers) where the designer is a player, under which we show the principal's next-best, when the first-best is unachievable, is the one-shot Nash. We show how our analysis extends to variations on the game accommodating more agents, caring about one's own performance, cheap talk and losses.

Bounded Rationality and Limited Datasets (with K. Rozen) Version: December 2018, R&R TE

Idea: Theories of bounded rationality are typically characterized over an exhaustive data set. We develop a methodology to help understand the empirical content of such theories when the data is limited. We apply our approach to an array of theories, illustrating its versatility. Our work adapts the classic, revealed-preference approach to new forms of revealed information. It identifies theories and datasets that are testable in the same elegant way as Rationality, as well as theories and datasets for which testing is more challenging. We also show that previous attempts to test consistency of limited data with bounded rationality theories are subject to a conceptual pitfall that can lead to false positives and empty out-of-sample predictions.

Testable Implications of Some Classic Assignment Methods (with K. Rozen) Version: July 2018

Idea: We study the testable implications of serial dictatorship, stable many-to-one matchings, and the core of housing markets. We show that serial dictatorship is easy to test, and explain how elements of the power ranking between agents can be identified. Stability is also easy to test for an interesting class of many-to-one matching problems, and is tightly related to serial dictatorship. We provide an insightful characterization of the core of Shapley and Scarf (1974)'s housing markets using revealed top-trading cycles. This characterization proves useful in many examples, but is also used to prove that testing the core is generally NP-hard.

Order-k Rationality (with S. Barberà, A. Neme, and K. Rozen) Version : October 2019

Idea: A decision maker (DM) may not perfectly maximize her preference over the feasible set. She may feel it is good enough to maximize her preference over a sufficiently large consideration set; or just require that her choice is sufficiently well-ranked (e.g., in the top quintile of options); or even endogenously determine a threshold for what is good enough, based on an initial sampling of the options. Heuristics such as these are all encompassed by a common theory of Order-k Rationality, which relaxes perfect optimization by only requiring choices from a set S to fall within the set’s top k(S) elements according to the DM’s preference ordering. Heuristics aside, this departure from rationality offers a natural way, in the classic ‘as if’ tradition, to gradually accommodate more choice patterns as k increases. We characterize the empirical content of Order-k Rationality (and related theories), and provide a tractable testing method which is comparable to the method of checking SARP.

Bargaining Over Contingent Contracts Under Incomplete Information (with J. Fanning and K. Rozen) Version : December 2019

Idea: We study bargaining over contingent contracts in settings where private information becomes verifiable when the time comes to implement the agreement. We suggest a simple, two-stage game incorporating important aspects of bargaining. We characterize equilibria where bargainers reach agreement, and study interim-efficient limits as bargaining frictions vanish. Under mild regularity conditions, all such limits belong to Myerson (1984)’s axiomatic solution.

Relaxed Optimization (with K. Rozen) Version: July 2019

Idea: We relax the first-order conditions in optimization to propose an approximation of rational consumer choice. We provide an axiomatically-founded measure of the extent to which the FOCs are violated, which is also interpretable in terms of a money-pump multiplier. The framework encompasses measurement errors, information unobserved to the modeler, and bounded rationality (e.g., misunderstanding prices, misperceiving utility tradeoffs). We develop the testable implications of the model for demand data--even when restricting to subclasses of regular utility functions--and study the properties of a new index of irrationality, the FOC-Departure Index (FDI), which can be applied in all contexts for which the first-order approach is meaningful.

Non-Bayesian Persuasion (with X. Zhang) Version: March 2020

Idea: Following Kamenica and Gentzkow (2011), the paper studies persuasion as an information design problem. We investigate how mistakes in probabilistic inference impact optimal persuasion. The ‘concavi?cation method’ is shown to extend naturally to a large class of belief updating rules, which we identify and characterize. This class comprises many non-Bayesian models discussed in the literature. We apply this new technique to gain insight in a variety of questions about, among other things, the revelation principle, the ranking of updating rules, and circumstances under which persuasion is bene?cial. Our technique extends to also shed light on the question of robust persuasion.

Communication, Perception and Strategic Obfuscation (with K. Rozen) Version: March 2020

Idea: We study the empirical content of simple Sender-Receiver games in which disclosures are mandatory but may be obfuscated. We focus on the fungibility between strategic inference and costly perception, developing a stylized theoretical framework that highlights this channel. Our framework yields crisp testable implications for equilibrium play, and naturally lends itself to an experimental design. Our laboratory results show that a large majority of Senders strategically obfuscate; and an aggregate analysis of Receiver’s stochastic-choice data suggests Receivers adjust their perception in response to strategic inference.

Incorporating Risk in Choice Theory: Some Observations Version: May 2020

Idea: This paper provides a warning: a property in the spirit of the sure-thing principle that may sound intuitive at first, and indeed standard in classic models, is systematically violated when considering choices that cannot be obtained through the maximization of a preference ordering. Beyond choice theory, this observation also has relevant implications for game theory, social choice, and mechanism design.

Bad Repetition (with K. Rozen) Version: July 2020

Idea: This paper experimentally explores the often-overlooked negative implications of repeated interactions, whereby promises of rewards and threats of punishments may lead people to take actions that are detrimental to themselves and/or others. For both finitely and infinitely repeated games, we demonstrate the prevalence of negative equilibria (which are worse than repetition of the one-shot Nash equilibrium) by departing from the stage games most commonly studied in the experimental literature. For the simple games we study, repetition can open the door to equilibria that capture stylized aspects of peer pressure and bystander complacency.

Coming Soon:

Level-k Mechanism Design with Small Modeling Mistakes (with R. Saran and R. Serrano)

Coming soon...

Selected Publications: (Click here for a complete list)

The Type-Agent Core for Exchange Economies under Asymmetric Information (JET 2007)

Idea: The type-agent core is a new solution concept for exchange economies with asymmetric information. It coincides with the set of equilibrium outcomes of a simple competitive screening game. Uninformed intermediaries help the agents to cooperate in an attempt to make some profit. The type-agent core is a subset of Wilson (1978)'s coarse core. It is never empty, even though it may be a strict subset of Wilson's fine core. In addition, it converges towards the set of constrained market equilibria as the economy is replicated.

Impartial Division of a Dollar (with H. Moulin and N. Tideman, JET 2008)

Idea: For impartial division, each participant reports only her opinion about the fair relative shares of the other participants, and this report has no effect on her own share. If a specific division is compatible with all reports, it is implemented. We propose a family of natural methods meeting these requirements, for a division among four or more participants. No such method exists for a division among three participants.

Marginal Contributions and Externalities in the Value (Ecma 2008)

Idea: The paper studies how to extend the theory of the Shapley value to problems involving externalities. Using the standard axiom systems behind the Shapley value leads to the identification of bounds on players’ payoffs around an “externality-free” value. The approach determines the direction and maximum size of Pigouvian-like transfers among players, transfers based on the specific nature of externalities that are compatible with basic normative principles.

No Profitable Decompositions in Quasi-Linear Allocation Problems (with C. Bejan, JET 2011)

Idea: We study the problem of allocating a bundle of perfectly divisible private goods from an axiomatic point of view, in situations where compensations can be made through monetary transfers. The key property we impose on the allocation rule requires that no agent should be able to gain by decomposing the problem into sequences of subproblems. Combined with additional standard properties, it leads to a characterization of the rule that shares the total surplus equally. Hence a traditional welfarist rule emerges as the unique consequence of our axioms phrased in a natural economic environment.

Reason-Based Choice (with K. Eliaz, TE 2012)

Idea: Two well-documented violations of rationality (the attraction and compromise effects) can be captured as the cooperative solution to an intrapersonal bargaining problem among two different criteria for choosing. We first axiomatically characterize our solution when the two criteria are known, and then discuss the testable implications of our model when the two criteria are unknown. Alternatively, our analysis may be reinterpreted as a study of (interpersonal) bilateral bargaining over a finite set of options.

On the Selection of Arbitrators (with K. Eliaz and B. Knight, AER 2014)

Idea: The problem of arbitrator selection is studied using implementation theory. Combining theory and experiments, we document some issues with the veto-rank procedure that is commonly used in practice, and develop a new sequential procedure, shortlisting, with better properties.

Competing for Consumer Inattention (with K. Eliaz and K. Rozen, JPE 2014)

Idea: How do markets respond when consumers are able to examine only a limited number of markets for the best price? A firm’s price can deflect or draw attention to its market, and consequently, limited attention introduces a new dimension of cross-market competition. We characterize the equilibrium of a stylized model capturing these features, and show that having consumers who are only partially attentive increases consumer welfare. With less attention, consumers are more likely to miss the best offers; but enhanced cross-market competition decreases average price paid, as leading firms try to stay under the consumers’ radar.

Behavioral Implementation (AER 2014)

Idea: Implementation theory assumes that participants’ choices are rational, in the sense of being consistent with the maximization of a context-independent preference. The paper investigates implementation under complete information when individuals’ choices need not be rational.

Strategic Disclosure of Feasible Options (with K. Eliaz, GEB 2015)

Idea: An important, yet overlooked, step in bargaining is figuring out the set of feasible options. Once options are on the table, different compromise rules can be used to decide on which final option to settle on. These compromise rules impact the willingness to disclose feasible options. Both in a static and in a dynamic setting, we show how the Nash compromise rule is second-best when it comes to disclosing options.

On the Redundancy of the Implicit Welfarist Axiom in Bargaining Theory (JET 2015)

Idea: There is a mismatch between the general motivation provided for Nash’s (1950) axioms and their actual mathematical content because they are phrased in the space of joint (Bernoulli) utilities. Alternatively, it is easy to rephrase these axioms in an economic environment so as to match their intuitive meaning, but Nash’s proof then applies only if one adds a cardinal welfarist axiom requiring that the solution of two problems that happen to have the same image in the space of joint utilities for some linear representation of von Neumann/Morgenstern preferences, must coincide in that space. Using a more elaborate argument, this paper shows, however, that this implicit cardinal welfarist axiom is redundant on a natural economic domain. This non-welfarist characterization of the Nash solution is shown to extend to a larger class of preferences that accommodate some forms of non-expected utility

Level-k Mechanism Design (with R. Saran and R. Serrano, ReStud 2019)

Idea: In my paper on behavioral implementation, bounded rationality was understood as a failure of preference maximization. The assumption of rational expectations was maintained, however. In this paper, agents do maximize a preference ordering, but the notion of Nash equilibrium is replaced by a level-k behavioral model.