Bernd Irlenbusch

An experimental study on tournament design

Abstract Since recently, rank order tournaments have become quite popular for providing incentives in employment relationships. However, the consequences of different tournament designs are widely unexplored. This paper experimentally investigates different tournament design alternatives along two dimensions: tournament size and prize structure. We find that average effort tends to increase with a higher proportion of winner prizes. Additionally, variability of effort is lower if the number of winner prizes is high. Especially two-person tournaments are prone to collusion. Furthermore, we observe a restart effect.

Sabotage in Tournaments: Evidence from a Laboratory Experiment

Although relative performance schemes are pervasive in organizations, reliable empirical data on induced sabotage behavior are almost nonexistent. We study sabotage in repeated tournaments in a controlled laboratory experiment and observe that effort and sabotage are higher for higher wage spreads. Additionally, we find that also in the presence of tournament incentives, agents react reciprocally to higher wages by exerting higher effort. Destructive activities are reduced by explicitly calling them by their name “sabotage.” Communication among principal and agents can curb sabotage when they agree on flat prize structures and increased output. If sabotage is not possible, the principals choose tournament incentives more often. This paper was accepted by Peter Wakker, decision analysis.

Sabotage in Corporate Contests – An Experimental Analysis

Abstract In corporate contests, employees compete for a prize. Ideally, contests induce employees to exert productive effort which increases their probability of winning. In many environments, however, employees can also improve their own ranking position by harming their colleagues. Such negative incentive effects of corporate contests are largely unexplored, which can partly be attributed to the fact that sabotaging behavior is almost unobservable in the field. In this study we analyze behavior in experimental contests with heterogeneous players who are able to mutually sabotage each other. We find that sabotaging behavior systematically varies with the composition of different types of contestants. Moreover, if the saboteur’s identity is revealed sabotage decreases while retaliation motives prevail. Our results promise to be valuable when designing corporate contests.

An Intergenerational Common Pool Resource Experiment

Many renewable resources are in intergenerational common pools, exploited by one generation after another. In our experiment, the stock available to each generation depends on the extent of exploitation by previous generations and on resources growth rate, which is either slow or fast. Subjects show altruistic restraint in exploitation, but not enough to achieve the social optimum. The presence of an intergenerational link induces subjects - both in slow and in fast - to expect less resource exploitation from each other than subjects expect in a single generation control. On average, expectations are too optimistic, especially in slow, where intended free-riding behavior is predominant.

An Experimental Test of Design Alternatives for the British 3G / UMTS Auction

In spring 2000, the British government auctioned off licences for Third Generation mobile telecommunications services. In the preparation of the auction, two designs involving each a hybrid of an English and a sealed-bid auction were suggested by the government: a discriminatory and a uniform price variant. We report an experiment on these two designs, and also compare the results to those with a pure English auction. Both hybrids are similar in efficiency, revenue differences disappear as bidders get experienced. Compared to the discriminatory format, the pure English auction gives new entrants better chances.

Leading With(Out) Sacrifice? A Public-Goods Experiment with a Privileged Player

We analyze two team settings in which one member in a team has stronger incentives to contribute than the others. If contributions constitute a sacrifice for the strong player, the other team members are more inclined to cooperate than if contributions are strictly dominant for the strong player.

Leading With(Out) Sacrifice? A Public-Goods Experiment with a Super-Additive Player

We analyse two team settings in which one member in a team has stronger incentives to contribute than the others. If contributions constitute a sacrifice for the strong player, the other team members are more inclined to cooperate than if contributions are strictly dominant for the strong player.

Experimental perspectives on incentives in organisations

Controllability is the decisive advantage of experimental research. Therefore this method suggests itself for investigating clear-cut hypothesis regarding the impact of managerial design issues on organisational behaviour. The current paper reviews selective experimental findings on behaviour under various incentives schemes. It is found that the effectiveness of fixed wages is to a large extent determined by social norms like reciprocity. Fixed wages are analysed if effort is intransparent and in the presence of career concerns. Furthermore they are contrasted with piece rate schemes. Different design options of relative compensation, are discussed with a special emphasis on taking sabotage activities into account.

Fairness Crowded Out by Law: An Experimental Study on Withdrawal Rights

Withdrawal rights protect buyers in distance selling, for example when ordering via the Internet. After introducing such a law in Germany the proportion of returned goods drastically increased although most sellers had already offered a return option before. We experimentally investigate scenarios in which sellers can voluntarily offer a withdrawal right. In a second treatment it is provided by law. We find indications that a voluntary provision is perceived as friendly, so that buyers reciprocate by not exploiting sellers too heavily. A third treatment reveals that small return costs for buyers have only a marginal influence on withdrawal behaviour.

Turning the Lab into Jeremy Bentham’s Panopticon - The Effect of Punishment on Offenders and Non-Offenders

The most famous element in Bentham’s theory of punishment, the Panopticon Prison, expresses his view of the two purposes of punishment, deterrence and special prevention. We investigate Bentham’s intuition in a public goods lab experiment by manipulating how much information on punishment experienced by others is available to would-be offenders. Compared with the tone that Jeremy Bentham set, our results are non-expected: If would-be offenders learn about contributions and punishment of others at the individual level, they contribute much less to the public project. Our results confirm the special prevention effect but show that the deterrence effect is smaller the more information on individual punishment is available.The most famous element in Bentham’s theory of punishment, the Panopticon Prison, expresses his view of the two purposes of punishment, deterrence and special prevention. We investigate Bentham’s intuition in a public goods lab experiment by manipulating how much information on punishment experienced by others is available to would-be offenders. Compared with the tone that Jeremy Bentham set, our results are non-expected: If would-be offenders learn about contributions and punishment of others at the individual level, they contribute much less to the public project. Our results confirm the special prevention effect but show that the deterrence effect is smaller the more information on individual punishment is available.