Andrew Kloosterman


Kloosterman, Andrew and Schotter, Andrew. Complementary Institutions and Economic Development: An Experimental Study Games and Economic Behavior, 2016, 99, pp. 186-205.
Kloosterman, Andrew. Directed Search with Heterogeneous Firms: An Experimental Study Experimental Economics, 2016, 19(1), pp. 51-66.
Kloosterman, Andrew. Public Information in Markov Games Journal of Economic Theory, 2015, 157, pp. 28-48.

Working Papers


A Simple Experimental Test of the Coase Conjecture: Fairness in Dynamic Bargaining (joint with Jack Fanning)
In each round of an infinite horizon bargaining game, a proposer proposes a division of chips, until a responder accepts. The Coase conjecture predicts that incomplete information about responders' preferences for fairness leads to almost immediate agreement on an equal payoff split when discounting between rounds is small. We experimentally test this prediction when chips are equally valuable to both bargainers and when they are worth three times as much to proposers, and compare outcomes to an ultimatum game. Behavior offers strong support for the theory. In particular, when chips are more valuable to proposers, initial offers, initial minimum acceptable offers, and responder payoffs are significantly higher in the infinite horizon game than in the ultimatum game, while proposer payoffs are significantly lower.

Ultimatum Game Bargaining in a Partially Directed Search Market (joint with Stephen Paul)
We investigate a partially directed search market where buyers search for sellers and then final prices are determined by ultimatum game bargaining. In the search stage of this game, sellers post intervals of possible surplus splits to attract buyers and then buyers approach one seller from whom they attempt to purchase. Our main goal is to investigate the interaction between bargaining and competition in the preliminary search stage. The main results confirm that behavior in the ultimatum game is consistent with past findings (i.e., fairness matters), and the main effect on search is to drive up the posted lower bounds for buyer surplus above the competitive equilibrium (towards more equal splits). Our main treatment variable is the number of buyers in the market, and when the number of buyers is increased, lower bounds and ultimatum offers to buyers decrease. This is consistent with fairness perceptions being influenced by competition.

Stochastic/Repeated Games

Cooperation in Stochastic Games: A Prisoner's Dilemma Experiment
This experiment investigates cooperation in a stochastic version of the infinitely repeated prisoner's dilemma. The stochastic element introduces the importance of beliefs about the future for supporting cooperation as well as cooperation and defection on the equilibrium path. The results confirm that subjects cooperate as predicted after they gain sufficient experience. There is some evidence in favor of alternating cooperation and defection, but a maximum likelihood strategy estimation suggests that cooperation is conditioned mostly on past actions. For example, the popular repeated game strategies Grim Trigger and Tit-for-Tat are still popular here, although they are not equilibria in this environment.

An Experimental Study of Public Information in the Asymmetric Partnership Game
Markov games model a dynamic setting where there is uncertainty about the future. This paper analyzes a laboratory experiment on the novel asymmetric partnership game in the environment of Kloosterman 2015 where public information about the uncertain future is available to the players. In this game, two players jointly work on a project each period that only benefits one of them. Costly effort may be incentivized through intertemporal incentives; effort which is non-benefical today induces effort from their partner in future periods when it is beneficial. Behavior is shown to be best explained by an adaptation of the basin of attraction (as introduced in Dal Bò and Frèchette 2011) to allow for state-dependent beliefs. The public signals impact choices in the way that this theory predicts, and ultimately there is more effort when signals are more informative. Finally, a new way to investigate strategies in these types of games is considered, and the results show how the different states are used by subjects to employ partial effort strategies.

School Choice

Essentially Stable Matchings (joint with Peter Troyan and David Delacrétaz)
We propose a solution to the trade-off between Pareto efficiency and stability in matching markets. We definne a matching to be essentially stable if any claim initiates a chain of reassignments that ultimately results in the initial claimant losing the object she claimed to a third agent. Our definition is not only compatible with Pareto efficiency but also is practical, since explaining why claims are not valid is straightforward. We study the essentially stable set and show that it shares some, though not all, of the structure possessed by the stable set. We classify popular Pareto efficient mechanisms using our definition: those based on Shapley and Scarf's TTC mechanism are not essentially stable while Kesten's EADA mechanism is. Our analysis points to a trade-off among essential stability, Pareto efficiency, and strategyproofness: mechanisms exist that satisfy any two of these properties but no mechanism achieves all three.

School Choice with Asymmetric Information: Priority Design and the Curse of Acceptance (joint with Peter Troyan)
An implicit assumption in most of the matching literature is that all participants know their preferences. If there is variance in the effort agents spend researching options, some will know their preferences, while others may not. When this is true, (ex-post) stable outcomes need not exist and informed agents gain at the expense of less informed agents, outcomes we attribute to a curse of acceptance for the less informed students. However, when all agents have a "secure school" we recover positive results: equilibrium strategies are simple, the outcome is ex-post stable, and less informed students are made better off. Our results have potential policy implications for the current debate in school choice over how priority design affects outcomes.

Works in Progress

Imperfect Monitoring and Coordination Failure: Experimental Evidence

Optimal Information in a Principal Agent Problem