Inferring repeated-game strategies from actions: evidence from trust game experiments

Summary. Combining a strategy model, an inference procedure and a new experimental design, we map sequences of observed actions in repeated games to unobserved strategies that reflect decision-makers’ plans. We demonstrate the method by studying two institutional settings with distinct theoretical predictions. We find that almost all strategies inferred are best responses to one of the inferred strategies of other players, and in one of the settings almost all of the inferred strategies, which include triggers to punish non-cooperators, are consistent with equilibrium strategies. By developing a method to infer unobserved repeated-game strategies from actions, we take a step toward making game theory a more applied tool, bridging a gap between theory and observed behavior.Received: 23 December 2002, Revised: 19 April 2005, JEL Classification Numbers: C72, C80, C90.The authors are indebted for discussions with Ray Battalio, David Cooper, Robin Dubin, John Duffy, Ellen Garbarino, Susan Helper, Margaret Meyer, John Miller, Jim Rebitzer, Mari Rege, Al Roth, and John Van Huyck. The authors also benefited from discussants at economic department seminars at Case Western Reserve, McMaster and McGill University, University of Pittsburgh, SUNY-Stony Brook, and Texas A&M, and participants at the 2002 European Winter Meeting of the Econometric Society. We are grateful for the financial support provided by the Department of Economics at the University of Pittsburgh and Case Western Reserve University.     

Australian Audit Committees — Do They Meet Best Practice Guidelines?

The paper reports the results of a study of Australia's largest 250 companies' audit committees in two years (1998 and 2001). The audit committees were scrutinised on the bases of audit committee independence and the frequency of meetings. Australia's audit committees did not score highly on either criterion and most companies fell short of best practice guidelines. Further, it was puzzling and of concern that Australian audit committees appear to have deteriorated in terms of independence and frequency of meetings between 1998 and 2001.     

Computing equilibria of GEI by relocalization on a Grassmann manifold

An algorithm is described to compute equilibria of the general economic model with incomplete asset markets, that is, of GEI. The algorithm is based on the existence of a route of zeros of a homotopy whose domain includes the price simplex and a Grassmann Manifold. This route is followed, in effect, by localizing and following diffeomorphic pieces in Euclidean space, and by relocalizing as is necessary.     

Economic Comparison of Alternative Tillage Systems under Risk.

Stochastic dominance efficiency criteria are used to rank the net farm return distributions for four different tillage systems under six different farm scenarios. Pairwise comparisons of tillage systems are carried over incremental upper and lower risk-aversion coefficient (RAC) bounds identified for each farm scenario, based on the size and spread of the outcome distributions to identify regions where dominance may switch between tillage systems. Ridge-till systems are generally the dominant tillage system for all farm scenarios considered. No-till systems on clay loam soils are more dominant in the risk-preferring range and less dominant in the risk-averse range, which relates to the relatively larger range of net return values and smaller mean net return values generated by this tillage system. Conversely, the moldboard plow and chisel plow systems are more dominant in risk-aversion intervals and less dominant in the risk-preferring intervals for clay loam soils. In sandy soil scenarios, the ridge-till and no-till systems are dominant over the two fall tillage systems for all farm sizes, indicating that these tillage systems are more competitive with conventional tillage systems in lighter soil type situations. A sensitivity analysis between moldboard plow and no-till systems indicates that no-till would dominate in risk-preferring intervals, and an increase in no-till net farm returns of 16 per acre would change dominance in favor of no-till in risk-averse interval space. On a utilisé les critères d'efficacitéà dominance stochastique pour classer les divers niveaux de recettes agricoles nettes dans quatre systèmes différents de travail du sol, selon six scénarios d'exploitation. Des comparaisons en paires des systèmes de travail ont été faites en regard de tranches croissantes des limites supérieures et inférieures du coefficient d'aversion du risque (CER), établies pour chaque scénario d'après l'importance et l'étendue des distributions du rendement, dans le but d'identifier les régions où la dominance serait susceptible de passer d'un système de travail à l'autre. Le système de travail sur billons était en général le système dominant dans tous les scénarios d'exploitation examinés. La culture sans travail du sol sur loam argileux était plus dominante dans les intervalles d'acceptation du resque et moins dominante dans les intervalles d'aversion du resque, cequi s'explique au fait que ce système de travail donne lieu à une fourchette relativement plus large, mais aussi à une moyenne plus faible du revenu net. À l'inverse, le labour à la charrue classique et au chisel dominait davantage dans les intervalles d'aversion du risque. Dans les sols sableux, le travail sur billons et la culture sans labour l'emportaient sur les deux autres régimes dans toutes les tailles d'exploitation, ce qui porte à croire que, dans les sols plus légers, ces régimes sont plus concurrentiels avec les systèmes classiques. L'analyse de sensibilité entre le labour à la charrue et les systèmes sans travail du sol montre que ces derniers dominent dans les intervalles d'acceptation du risque et qu'un accroissement du revenu agricole net de 16 /acre ferait glisser la dominance vers les intervalles d'aversion du risque.     

Research notes and communications management and organizational change: A note on the railroad industry

Hypotheses which relate top-level managers' age, years of company and industry service, and education to strategic change are studied with a sample of 855 managers from 27 railroads. Results generally support hypotheses that younger managers and those with less experience are more likely to alter their strategies with changing environmental conditions.     

Evolutionary Stability in the Investment-Opportunism-Retaliation Game

Can economically efficient outcomes be obtained and sustained in the absence of externally enforced property rights? We study the evolutionary properties of a game that exhibits two well-defined Nash equilibria: one generates an inefficient outcome while the other set generates an efficient outcome supported by the potential for retaliation. Although standard forward-looking refinements eliminate the efficient equilibrium, neither equilibrium type satisfies strict evolutionary stability criteria. However, both types of equilibrium define strategies that are neutrally stable, which makes them vulnerable to drift in dynamic environments. We conduct computer simulation experiments in which players learn adaptively via a tournament selection mechanism called sophisticated experimentation. Our simulations demonstrate that while the system spends a disproportionately high proportion of time in the inefficient equilibrium set, the efficient equilibrium is pervasive as the system drifts back and forth between the equilibrium sets, never settling on one or the other.     

Labor force migration, non-labor force migration, and non-employment reasons for migration

This paper documents the extent and nature of non-employment factors in migration. The labor force status of over 18 million recent interstate migrants in the United States and stated reasons for moving reported in several surveys in the U.S. are examined. Labor force migrants are heterogeneous in terms of the relationships between acquisition of employment and the migration decision, and in terms of the influence of decision makers outside of the migrant household including firms and the government. Numerically important categories of migrants not traditionally captured in migration models exist, including the elderly, the military, and movers from abroad. Whereas non-employment reasons are secondary to employment reasons for a majority of labor force migrants, non-employment factors are singularly important or operate in combination with employment factors for the majority of all migrants. The heterogeneity of migrant types and migration reasons needs to be better captured in migration models.     

THE RELATIONSHIP BETWEEN THE BLACK MARKET AND OFFICIAL EXCHANGE RATES: AN EXAMINATION OF LONG-RUN DYNAMICS IN INDIA

This study is the first to use Johansen's cointegration approach for India in the analysis of the long‐term dynamics between the black and official exchange rates for the period 1953–1993. The study also estimates the long‐run elasticity of the official rate with respect to the black market rate. As monthly data over 40 years are used, and a more robust methodology is employed, the results are likely to be more reliable as compared with the earlier work on India. The results of our study suggest that while there is a long‐term relationship between the two rates, the direction of causality is from the black rate to the official exchange rate. This is plausible in the Indian context where policy has generally lagged behind events in the black market. The hypothesis of a constant black market premium is rejected, implying that there is a mismatch between the percentage change in the official exchange rate and the percentage change in the black market rate.     

Sample attrition and labor force dynamics: Evidence from the Spanish labor force survey

In this paper we examine a number of issues that arise in investigating labor force dynamics using the Spanish Labor Force Survey (EPA). These issues are by no means specific to the Spanish case and apply to most European-style labor force surveys. Our main conclusions may be summarized as follows. First, survey nonresponse cannot be neglected. Second, the EPA tends to underestimate employment and participation of high-educated young people, and to overestimate those of the low-educated elderly. Finally, we find little evidence that attrition causes important selection biases in estimating quarterly transition probabilities.