The optimality of (stochastic) veto delegation

TL;DR

This paper characterizes the optimal veto-based delegation mechanism in principal-agent problems with risk-averse principals and risk-neutral agents, showing that vetoes are optimal when the principal's risk aversion exceeds that of the agent.

Contribution

It provides a formal characterization of the optimal veto mechanism in a one-dimensional setting with state-independent preferences.

Findings

Veto mechanisms are optimal when the principal is more risk-averse than the agent.

The principal uses veto only when the state exceeds a critical threshold.

Optimal veto mechanisms do not involve randomization among non-status quo options.

Abstract

We analyze the optimal delegation problem between a principal and an agent, assuming that the latter has state-independent preferences. We demonstrate that if the principal is more risk-averse than the agent toward non-status quo options, an optimal mechanism is a {\em veto mechanism}. In a veto mechanism, the principal uses veto (i.e., maintaining the status quo) to balance the agent's incentives and does not randomize among non-status quo options. We characterize the optimal veto mechanism in a one-dimensional setting. In the solution, the principal uses veto only when the state surpasses a critical threshold.