Jaeyoung Sung

Linearity with Project Selection and Controllable Diffusion Rate in Continuous-Time Principal-Agent Problems

I extend Holmstrom and Milgroms model by allowing the agent to control privately or publicly the diffusion-rate process, and show that the optimal contract is still linear. I discuss the conditions under which conflicts over the choice of diffusion rate do and do not arise. As an application, I examine project- selection problems of a firm. Conflicts between the manager and investors arise because giving the manager too much incentive to work to increase the profit from ongoing operations can induce him to be too conservative in project selection, possibly resulting in optimal contracts with low sensitivities.

Executive Pay, Talent and Firm Size: Why has CEO Pay Grown so Much?

We exposit an integrated agency model of multi-period career concerns and labor market equilibrium with managerial reservation utility levels, and thus pay levels, determined endogenously for firms of different sizes. Based on observations from a long time-series of S&P 1500 companies, we estimate the stochastic production function describing the incremental wealth created by the manager as a function of “effort”, latent raw “talent”, idiosyncratic firm risk (asset volatility) and the opening value of assets employed. We show that CEO talent affects the marginal productivity of the firm at approximately twice the rate as effort. Since asset volatility is also more subject to scale effects than effort, risk per marginal product of effort is higher in larger firms. Due to the cost of compensating managers for risk, pay-performance sensitivity optimally declines with size. Furthermore, our talent estimates explain much of the increments to real CEO pay levels and income over recent decades as a response to increases in talent and as compensation for higher risk borne by executives, with firm size growth playing a negligible role. We also identify the most talented CEOs who earned enterprise returns 17 times higher than the CEOs of the largest firms.

Optimal Contracts under Moral Hazard and Adverse Selection:A Continuous-Time Approach

In spite of the importance of optimal contracting problems under moral hazard and adverse selection, current literature offers no optimal solutions to contracting problems under moral hazard and adverse selection with risk averse agents. The agents risk aversion, however, appears to be critical for understanding managerial compensation problems. We present a continuous-time agency model with a risk-averse agent and a risk-neutral principal to show that moral hazard and adverse selection can be optimally resolved with a menu of linear contracts. In application, we discuss a few managerial compensation problems involving managerial project selection and capital budgeting decisions, and show that a flat-wage contract is sometimes optimal.

Optimal Contracting under Mean-Volatility Ambiguity Uncertainties

We present a continuous-time agency model under mean-volatility joint ambiguity uncertainties, where both the principal and agent exhibit Gilboa-Schmeidlers extreme ambiguity aversion. For this, we extend the well-known martingale method for mean-volatility control problems in weak formulation, and show that the second-best contract in general consists of two sharing rules: one for realized outcome and the other for realized volatility. The outcome sharing is for uncertainty sharing and work incentives, and the volatility sharing is to align the agents worst prior over ambiguity uncertainties with the principals own. The structure of the optimal contract offers an alternative way of looking at managerial compensation. Unlike the existing standard contracting models, our model implies that the compensation increases with the realized volatility, and that pay-performance sensitivity to the outcome is unrelated to the realized volatility. The positive dependence on the realized volatility is consistent with the frequent executive compensation practices of granting stock options and the empirical evidence of the sharp rise in executive pays during 1990s.

A GENERAL EQUILIBRIUM MODEL OF A MULTIFIRM MORAL‐HAZARD ECONOMY WITH FINANCIAL MARKETS

We present a general equilibrium model of a moral‐hazard economy with many firms and financial markets, where stocks and bonds are traded. Contrary to the principal‐agent literature, we argue that optimal contracting in an infinite economy is not about a tradeoff between risk sharing and incentives, but it is all about incentives. Even when the economy is finite, optimal contracts do not depend on principals’ risk aversion, but on market prices of risks. We also show that optimal contracting does not require relative performance evaluation, that the second best risk‐free interest rate is lower than that of the first best, and that the second‐best equity premium can be higher or lower than that of the first best. Moral hazard can contribute to the resolution of the risk‐free rate puzzle. Its potential to explain the equity premium puzzle is examined.