On the optimal design of peer monitoring systems in capital budgeting

Abstract This paper considers peer monitoring in a capital budgeting setting with information asymmetry between principal and agents. The paper demonstrates in a two‐agent, two‐project setting that the principal can have each agent reveal their private information on the other agent's project, at an arbitrarily low cost. Next, an examination of monitoring resource allocations is made. Specifically, a comparison is made of broad monitoring, which spreads monitoring resources equally across agents' projects, and focused monitoring, which applies all monitoring resources to a single agent's project. Focused monitoring is preferred within an intermediate range of profitability, above which broad monitoring is preferred and below which neither is preferred. If project profitability is heterogeneous and focused monitoring is preferred, the more profitable project receives the focus. Both of these results occur because there is a threshold of monitoring resources that must be deployed for monitoring to provide any benefit, and the threshold is higher (more resources must be deployed) for lower profitability projects. That is, even if all monitoring resources are devoted to a low profitability project, they may provide no benefit to the principal. A discussion of how peer monitoring can be engineered by the firm is included.