Wealth-Robust Intertemporal Incentive Contracts

We study optimal incentive contracts in a continuous time principal-agent setting with hidden actions. The agent, whose effort controls the output, has a concave utility function which is non-separable in wealth and monetary cost of effort. The principal is risk neutral and optimally selects the effort to be induced and the contract design. Output follows a mean-reverting process with random coefficients. We characterize the class of W-robust compensation schemes that elicit a desired effort which is immune to the principal's mispecifications of the future wealth of the manager. We demonstrate the existence of a solution to the principal's problem, characterize the optimal effort policy, derive the optimal W-robust contract and show that our contract dominates randomized contracts