Risk-averse agents with peer pressure

Abstract

This article studies the effects of peer pressure on the incentives of risk-averse agents. It defines the peer pressure function and then assumes that each agent feels peer pressure not only when his effort level is below the standard level, but also when it is above that level. It also supposes that agents are heterogeneous in terms of their productivity and the degree to which they respond to peer pressure. It shows that a principal provides incentives that depend on the effects of peer pressure and risk-sharing.

Notes

¹ In this paper, one uses the term ‘risk components’ to refer to the product of the coefficient of constant absolute risk aversion and the uncertainty relating to output. One uses this term only to simplify the exposition.

² That is, agent 2 is superior to agent 1 with respect to productivity. Hereafter, one refers to agent 1 as ‘the bad agent’ and agent 2 as ‘the good agent’.

³ This view is supported by some studies. See, for example, Elster (1989).

⁴ As mentioned in footnote 1, the risk component (RC) is equal to rσ ².

⁵ Note that only when .

⁶ This occurs when the agents are risk-neutral (r = 0) and/or there is no uncertainty relating to output (σ² = 0).

⁷ From Equation 16, it is clear that β₂ > β₁ if m ₁ = m ₂ = 0.

⁸ However, his other effort level is still higher than that of the bad agent because of his productivity advantage.

⁹ Daido (2003) shows that the good agent's effort level is never less than that of the bad agent when the agents are risk-neutral, but protected by limited liability.

¹⁰ See Fehr and Schmidt (2003) for survey of this type of research.