Schedule - Parallel Session 2 - Strategic Interactions

IDL Boardroom - 14:00 - 15:30

Compromise and Attraction Effects in Bargaining

Maria Montero; Fabio Galeotti; Anders Poulsen

Abstract

We report experimental data from bargaining situations where bargainers can make proposals as often and whenever they want, and can communicate via chat. We vary the set of feasible contracts in order to investigate the possible presence of compromise and attraction effects. Specifically, we compare situations with only two possible contracts and situations where a third contract is added. The third contract may have the effect of making one of the two original contracts appear as a compromise between two extreme options; alternatively, it may be Pareto dominated by one of the two original contracts but not by the other. If occupying an intermediate position in the contract set significantly increases the frequency of agreements on a contract, we observe a compromise effect; likewise, if adding a dominated contract significantly increases the frequency of agreements on the contract that dominates it, we observe an attraction effect. Both effects are instances of violations of the independence of irrelevant alternatives axiom. Our main findings are the following. First, if there is a contract that is both equal and maximizes total earnings, nearly all agreements are on this contract. Making the other (unequal) contract a compromise has no effect. Second, if one of the contracts is equal and the other maximizes total earnings, making the unequal contract a compromise increases its frequency. Third, there is a clear difference between contract sets that contain an equal contract and contract sets that contain only an approximately equal contract. The approximately equal contract is chosen significantly less often than the perfectly equal contract. Since it is not chosen as often to begin with, there is more room for the frequency of this contract to increase when it is a compromise, and this is observed in our data. Finally, we observe attraction effects but they are not as robust as the compromise effects. The stronger attraction effects we observe occur when there are no equal contracts available (including the dominated ones).

Maria Montero

Associate Professor, University of Nottingham/CeDEx

Freedom, Power, and Interference

Claudia Neri; Hendrik Rommeswinkel

Abstract

Freedom and power are pervasive components in any social, political, and economic interaction. Individuals interact by making decisions, affecting themselves to the extent that they have the freedom to do so, and affecting others to the extent that they have the power to do so. Thus, freedom and power are fundamentally related to the exercise of decision rights. Economics, which has traditionally considered decision rights solely for their instrumental value in achieving outcomes, has recently moved to consider decision rights also for their intrinsic value, i.e., the value beyond the expected utility associated with them. In this paper, we propose a behavioral theory of preference for decision rights, driven by preference for freedom, power, and non-interference, and we conduct a novel laboratory experiment in which the effect of each preference is distinguished. We employ the following terminology. An individual experiences freedom when his preferences over the possible outcomes influence the outcomes he achieves. An individual experiences power when his preferences influence another individual’s outcomes. An individual does not experience interference when his outcomes are not influenced by another individual’s preferences. Each concept captures the causal dependence of an individuals’ preferences on an individuals’ outcomes. In our behavioral theory, individuals have not only preferences over outcomes, which lead them to value decision rights instrumentally, but also preference for freedom, power, and non-interference, which lead them to value decision rights intrinsically. Evidence from our experiment confirms the existence of an intrinsic value of decision rights, as reported by the existing literature. Most importantly, our theoretical framework and experimental design allow to disentangle the factors generating the intrinsic value of decision rights. We highlight two main findings. First, we find no evidence of preference for power. This result suggests that preference for power, as casually observed in politics or other institutional settings, may simply be instrumental to other components of well-being, such as status recognition. Second, we find stronger evidence of preference for non-interference than of preference for freedom. This result suggests that individuals value decision rights neither because they enjoy the freedom of making a choice, nor because they enjoy having power over other individuals, but rather because they dislike letting other individuals interfere in their outcomes. Our framework and findings lead to a fundamental change in perspective on preference for decision rights. Individuals like to have decision rights in virtue of the absence of the decision rights of other individuals.

Claudia Neri

Assistant Professor, University of St.Gallen

Trust, Risk and the Repeated Trust Game

Andis Sofianos

Abstract

Trust is an important element of most social and economic transactions. For example, an employer assigns responsibilities to an employee based on trust, spouses will trust each other to be faithful and citizens will trust the elected officials to champion their interests. The problem is that quite often, we cannot be sure if our trust is being exploited or reciprocated. In this paper I run an experiment that first disentangles trust choices from risk preferences – the confounding of trust choices and risk preferences has not been accounted for in the literature. Secondly, my design helps to understand the link between declared trusting dispositions and trust choices in an infinitely repeated trust game with hidden action. Trusting attitudes are found to be an important motivator when playing trust in a trust game even after controlling for risk preferences, hence indicating that inherent trust is indeed identified when studying the trust game. Moreover, the likely mechanism through which intrinsic trust operates in guiding trusting individuals to play trust more often is investigated. I find that more trusting individuals are more likely to forgive or offer the benefit of the doubt to others after a disappointing outcome. This forgiving nature of more intrinsically trusting individuals appears to be the reason for the divergence in behaviour between more and less trusting individuals, since in the initial rounds there is no distinction between the two groups. Additionally, the effect of intrinsic trust appears to be operating independently from the formation of beliefs. This suggests that trusting is additionally rewarding for more intrinsically trusting individuals and is not due to different sets of beliefs on the likelihood of reciprocation from others.

Andis Sofianos

Student, University of Warwick

Effort Provision and Optimal Prize Structure in Contests with Loss-Averse Players

Ayse Gul Mermer

Abstract

This paper studies a multiple prize contest with expectation-based loss-averse contestants a la Koszegi and Rabin (2006). The predictions of the model is able to align the empirical evidence observed in recent laboratory experiments on effort provision of contestants. More specifically the model predicts that high-ability contestants overexert effort while low-ability contestant exert very little to no effort in comparison to predictions with standard preferences. Moreover, the optimal prize allocation in contests differs markedly in the presence of expectation-based loss aversion. In particular, multiple prizes can become optimal when the cost-of-effort function is linear or concave, where standard preferences predict the optimality of a single prize in these cases. Several unequal prizes might be optimal when the cost-of-effort function is convex.

Ayse Gul Mermer

Assistant Professor, University of Manchester