Dirk Engelmann

Inequality Aversion, Efficiency, and Maximin Preferences in Simple Distribution Experiments: Comment

We present simple one-shot distribution experiments comparing the relative importance of efficiency concerns, maximin preferences, and inequality aversion, as well as the relative performance of the fairness theories by Gary E Bolton and Axel Ockenfels and by Ernst Fehr and Klaus M. Schmidt. While the Fehr-Schmidt theory performs better in a direct comparison, this appears to be due to being in line with maximin preferences. More importantly, we find that a combination of efficiency concerns, maximin preferences, and selfishness can rationalize most of the data while the Bolton-Ockenfels and Fehr-Schmidt theories are unable to explain important patterns.

Indirect Reciprocity and Strategic Reputation Building in an Experimental Helping Game

We study indirect reciprocity and strategic reputation building in an experimental helping game. At any time only half of the subjects can build a reputation. This allows us to study both pure indirect reciprocity that is not contaminated by strategic reputation building and the impact of incentives for strategic reputation building on the helping rate. We find that while pure indirect reciprocity appears to be important, the helping choice seems to be influenced at least as much by strategic considerations. Strategic do better than non-strategic players and non-reciprocal do better than reciprocal players, casting doubt on previously proposed evolutionary explanations for indirect reciprocity.

An Experimental Study of the Repeated Trust Game with Incomplete Information

In the trust game first player 1 decides between non-cooperation or trust in reciprocity and then, in the latter case, player 2 between exploiting player 1 or rewarding him. In our experiment, player 2 can be a notorically rewarding player (this type is implemented as a robot strategy) or a human participant who may decide opportunistically. To allow for reputation formation, this game is played repeatedly. Learning can be analysed since participants play successively several repeated games with changed partners. In our computerized experiment, participants can explicitly rely on mixed strategies which allows testing the qualitative and quantitative aspects of reputation equilibria also at an individual level.

Indirect reciprocity and strategic reputation building in an experimental helping game

We study indirect reciprocity and strategic reputation building in an experimental helping game. At any time only half of the subjects can build a reputation. This allows us to study both pure indirect reciprocity that is not contaminated by strategic reputation building and the impact of incentives for strategic reputation building on the helping rate. We find that while pure indirect reciprocity appears to be important, the helping choice seems to be influenced at least as much by strategic considerations. Strategic do better than non-strategic players and non-reciprocal do better than reciprocal players, casting doubt on previously proposed evolutionary explanations for indirect reciprocity.

To Buy or Not to Buy? An Experimental Study of Consumer Boycotts in Retail Markets

We investigate experimentally how firms and consumers react to a sudden cost increase in a competitive retail market. We compare two conditions that exclusively differ with respect to how difficult it is to organize and enforce boycotts. We find that cost increases translate into sudden price increases, and that consumer boycotts are frequent in response. However, consumer boycotts are unsuccessful in holding down market prices even if collective action problems are completely eliminated. While consumer boycotts do not increase consumer rent, they reduce market efficiency. Consumer boycotts apparently serve to punish firms for seemingly unfair price increases.

Bidding Behaviour in Multi-Unit Auctions – An Experimental Investigation*

We present laboratory experiments of five different multi-unit auction mechanisms. Two units of a homogeneous object were auctioned off among two bidders with flat demand for two units. We test whether expected demand reduction occurs in open and sealed-bid uniform-price auctions. We also test revenue equivalence for these auctions as well as for the Ausubel, the Vickrey and the discriminatory sealed-bid auction. Furthermore, we compare the five mechanisms with respect to the efficient allocation of the units.

Mechanisms for Efficient Voting with Private Information About Preferences

In games of conflict, players typically have an incentive to exaggerate their interests. This concerns issues ranging from negotiations between political parties to conflict resolution within marriages. We experimentally study this problem using a simple voting game where information about preferences is private. With random matching, subjects overwhelmingly follow the dominant strategy to exaggerate their preferences. The exogenous linking mechanism by Jackson and Sonnenschein (2007) captures nearly all achievable efficiency gains. Repeated interaction in various settings, which could allow endogenous linking mechanisms to evolve, leads to significant gains in truthful representation and efficiency only when players can choose their partners.

Deconstruction and reconstruction of an anomaly

We present a striking example of the deconstruction and reconstruction of an anomaly. In line with previous experiments we show in a one-shot setting that the allegedly robust false consensus effect disappears if representative information is readily available. But the effect reappears if a small cognitive effort is required to retrieve the information. Most subjects apparently ignore valuable information if it is not handed to them on a silver platter. We conclude that the relevance of the false consensus effect depends on the difficulty of retrieving the information and that the underlying mechanism is an information processing deficiency rather than egocentricity. Moreover, we discuss the potential relevance of our findings for other well-known effects like the winnerʼs curse and overconfidence.

Tax Liability-Side Equivalence in Experimental Posted-Offer Markets

In theory, the incidence of a tax should be independent of the side of the market on which it is levied. This principle of liability-side equivalence underlies virtually all theories of tax incidence. Policy discussions, however, tend to place great emphasis on the legal division of tax payments. We use computerized experimental posted-offer markets to test liability-side equivalence. We find that market outcomes are essentially the same when the tax is levied on sellers as when it is levied on buyers. Thus, we cannot reject liability-side equivalence.

An experimental comparison of the fairness models by Bolton and Ockenfels and by Fehr and Schmidt

We present an experiment to compare the two fairness theories by Bolton and Ockenfels [ERC] and by Fehr and Schmidt [F&S]. If one wants to compare their predictive power, most of the experiments that are interpreted retrospectively are not helpful, since both theories make equal or very similar predictions. Both models rely on inequality aversion. The fundamental difference between them is that ERC assumes that subjects like the average payoff to be as close as possible to their own payoff while F&S assumes that subjects dislike a payoff difference to any other individual. To obtain explicitly opposite predictions by the two theories we chose a game that focuses on their fundamental difference. A person received a fixed payoff and chose between three different allocations of money between a person who received in all allocations more than her and a person who always received less. The allocations with an average payoff for the other two persons closer to hers, had both individual payoffs more distant from hers. ERC predicts that she chooses the allocation that is most unequal between the other two persons. The choice of the opposite allocation is predicted by F&S. Subjects knew that their decision could never influence their own payoff. To prevent interference of preferences for efficiency with our objective, we designed two treatments, one where following the ERC prediction leads to a maximization of total payoff, one where maximization of total payoff is in line with the F&S prediction. In the second treatment the results clearly confirm the F&S prediction. In the first treatment subjects chose in about equal proportions the two extreme allocations. Hence the performance of F&S is much better than that of ERC, although both theories ignore the importance that subjects assign to efficiency.