Exogenous targeting instruments for eliciting efficiency in groups

Abstract

<p>When an individual agent's actions are unobservable, there is often an incentive for that agent to act in a way which is not socially optimal. This is the crux of the moral hazard problem and is compounded when there are a group of agents. Moral hazard in groups is a common social dilemma which encompasses situations such as the worker effort problem, contributions to public goods, and the abatement of non-point source pollution. Holmstrom (1982) suggests what is referred to as a forcing contract to induce a group to choose the optimal or target outcome. However, the optimal outcome is only one of many possible Nash equilibria under these forcing contracts. Segerson (1988) suggests the use of exogenous targeting instruments which do implement the optimal outcome as a unique Nash equilibrium. The purpose of this thesis is to test these different instruments in a controlled laboratory environment to determine whether they are able to induce a group of agents to select the socially optimal outcome. There are many studies which conclude that the target outcome cannot be attained under the Holmstrom (1982) forcing contract. Most notable among these studies is the 1997 paper by Nalbantian and Schotter. They conclude that costly monitoring or competitive teams are required to mitigate the problem of moral hazard in groups in an experiment based on the worker effort problem. The results from the experiment discussed in this dissertation show that some exogenous targeting instruments can induce groups of both homogenous and heterogeneous individuals to choose a target outcome. However, significant differences from Nash behaviour are observed at the individual level. It is shown that these differences can be explained by preferences for equity and decision error.</p>