We examine the force of three types of behavioral dynamics in quantity-setting triopoly experiments: (1) mimicking the successful firm, (2) rules based on following the exemplary firm, and (3) rules based on belief learning. Theoretically, these three types of rules lead to the competitive, the collusive, and the CournotNash outcome, respectively. In the experiment we employ three information treatments, each of which is hypothesized to be conducive to the force of one of the three dynamic rules. To a large extent, the results are consistent with the hypothesized relationships between treatments, behavioral rules, and outcomes.

Abstract not found

An experiment designed to assess the effects of advance production decisions on posted-offer market performance is reported. Six of the twelve triopolies were conducted under standard posted-offer rules. In the remaining markets sellers made binding production commitments prior to posting prices, an alteration that shifts the unique stage-game Nash equilibrium from the competitive to the Cournot outcome. As predicted, the advance production decisions raised prices and lowered output. Nevertheless, stable Cournot outcomes were never observed.

Cournot and Bertrand oligopolies constitute the two most prevalent models of firm competition. The analysis of Nash equilibria in each model reveals a unique prediction about the stable state of the system. Quite alarmingly, despite the similarities of the two models, their projections expose a stark dichotomy. Under the Cournot model, where firms compete by strategically managing their output quantity, firms enjoy positive profits as the resulting market prices exceed that of the marginal costs. On the contrary, the Bertrand model, in which firms compete on price, predicts that a duopoly is enough to push prices down to the marginal cost level. This suggestion that duopoly will result in perfect competition, is commonly referred to in the economics literature as the “Bertrand paradox”. In this paper, we move away from the safe haven of Nash equilibria as we analyze these models in disequilibrium under minimal behavioral hypotheses. Specifically, we assume that firms adapt their strategies over time, so that in hindsight their average payoffs are not exceeded by any single deviating strategy. Given this no-regret guarantee, we show that in the case of Cournot oligopolies, the unique Nash equilibrium fully captures the emergent behavior. Notably, we prove that under natural assumptions the daily market characteristics converge to the unique Nash. In contrast, in the case of Bertrand oligopolies, a wide range of positive average payoff profiles can be sustained. Hence, under the assumption that firms have no-regret the Bertrand paradox is resolved and both models arrive to the same conclusion that increased competition is necessary in order to achieve perfect pricing. 1.

We report results of experiments designed to test the predictions of the best reply process. In a Cournot oligopoly with four Þrms, the best reply process should theoretically explode if demand and cost functions are linear. We Þnd, however, no experimental evidence of such instability. Moreover, we Þnd no differences between a market which theoretically should not converge to Nash equilibrium and one which should converge because of inertia. We investigate the power of several learning dynamics to explain this unpredicted stability.

This paper reports an experiment designed to assess the effects of a rotation in the marginal cost curve on convergence in a repeated Cournot triopoly. Increasing the cost curve’s slope both reduces the serially-undominated set to the Nash prediction, and increases the peakedness of earnings. We observe higher rates of Nash equilibrium play in the design with the steeper marginal cost schedule, but only when participants are also rematched after each decision. Examination of response patterns suggests the treatment with a steeper marginal cost curve and with a re-matching of participants across periods induces the selection of Nash Consistent responses.

Convergence dans l’oligopole de Cournot fini avec apprentissage social et individuel Résumé La convergence vers l’équilibre de Nash dans l’oligopole de Cournot est un problème qui apparaît de manière récurrente dans les études économiques. Le développement des jeux évolutionnaires a permis l’utilisation d’un concept d’équilibre en adéquation avec les dynamiques d’ajustement et la stabilité évolutionnaire de l’équilibre de Cournot a été étudiée par plusieurs articles. Ces articles montrent que l’équilibre Walrasien est la seule solution évolutionnairement stable du jeu de Cournot. Vriend(2000) propose l’utilisation des algorithmes génétiques pour l’étude des dynamiques d’apprentissage et il obtient la convergence vers la solution de Cournot avec l’apprentissage individuel. Nous montrons dans cet article pourquoi l’apprentissage social conduit à la solution de Walras et comment l’apprentissage individuel peut effectivement permettre la convergence vers la solution de Cournot. De plus, ces résultats sont obtenus dans un cadre plus général avec une application à l’aide d’expériences informatiques.

We study a general sub-class of concave games, which we call socially concave games. We show that if each player follows any no-external regret minimization procedure then the dynamics converges in the sense that both the average action vector converges to a Nash equilibrium and that the utility of each player converges to her utility in that Nash equilibrium. We show that many natural games are socially concave games. Specifically, we show that linear Cournot competition and linear resource allocation games are socially-concave games, and therefore our convergence result applies to them. In addition, we show that a simple best response dynamic might diverge for linear resource allocation games, and is known to diverge for a linear Cournot competition. For the TCP congestion games we show that “near” the equilibrium these games are socially-concave, and using our general methodology we show convergence of specific regret minimization dynamics.

This paper summarizes some recent research pertaining to laboratory markets, and then discusses some of the implications of this research for applied economics and policy. Three results are discussed: (a) In one-sided markets where sellers post prices pricetracking response to demand shocks is very poor; (b) Despite the prominence of the Cournot model as a centerpiece for applied research, it is very difficult to construct an environment where Cournot outcomes are observed; (c) Market power, or the capacity of sellers to unilaterally deviate from the competitive prediction, is a prominent In the forty years since Chamberlain (1948) first reported results of a market experiment, the use of laboratory methods by economists has increased exponentially. Laboratory methods have become a standard research tool in areas ranging from individual decision theory to macroeconomics, and papers reporting experimental

I report results of an experiment designed to study the relation between the process of information search and learning in a Cournot oligopoly, with limited a priori information. Different theories of learning have been applied to this setting, each yielding a specific market outcome in the long run, and postulating specific informational requirements. By allowing players to choose the information they wish to acquire, and controlling for these choices, I study the features of the learning model actually followed by the subjects, and the relation between the information they gather and the market behavior they adopt. According to my results, learning appears to be a composite process, in which different components coexist. Belief learning seems to be the leading element, as subjects try to form expectations about their opponents ’ future actions and to best reply to them. When subjects also look at the strategies individually adopted by their competitors, though, they tend to imitate the most successful behavior, which makes markets more competitive. Finally, reinforcement learning also plays a nonnegligible role, as subjects tend to favor strategies that have yielded higher profits in the past. I show that these different elements may be usefully incorporated into a more sophisticated learning model, shaped after self tuning EWA learning model.