Abstract
Large-scale infrastructure investments are often carried out in settings where their eventual usefulness or importance is difficult to predict. This paper studies optimal incentives for investment when the agent undertaking the investment has superior information on two dimensions: the cost of investment and the likelihood it is useful or beneficial to the principal. Usefulness eventually becomes public, but punishments are limited as the regulator aims at ensuring the agent earns non-negative profits in each period. We characterize the optimal incentive scheme and show it involves either: (i) investments by the agent even though he knows they are useless and rents to only cost-efficient types, or (ii) rents to all types. The possibility that rent is left to all types contrasts with the usual prediction in static (and also dynamic) mechanism design and arises even though the agent’s preferences are stable over time.