How Should Performance Signals Affect Contracts?

Pierre Chaigneau is Associate Professor of Finance at Queen’s University Smith School of Business; Alex Edmans is Professor of Finance at London Business School; and Daniel Gottlieb is Associate Professor of Managerial Economics and Strategy at the London School of Economics. This post is based on their recent paper, forthcoming in the Review of Financial Studies. Related research from the Program on Corporate Governance includes Paying for Long-Term Performance by Lucian Bebchuk and Jesse Fried (discussed on the Forum here).

Incentive pay for top executives is increasingly based on performance measures (“signals”) other than the stock price—financial metrics such as earnings and sales, and sustainability metrics such as carbon emissions and safety. How should performance signals be incorporated into incentive pay contracts?

A simple solution is to give CEOs shares in the firm, because this perfectly aligns them with shareholder value. However, an important lesson from principal-agent models of incentive provision is that top executives should be rewarded not when shareholder value is high, but when they made the (ex ante) right decisions. This subtle difference can have major consequences. For example, an increase in shareholder value due to a booming economy should not be rewarded with higher CEO pay.

Executive pay should thus depend on shareholder value, but only insofar as it provides valuable information about the decisions made by top executives. Similarly, Nobel laureate Bengt Holmström showed that other signals should be used in incentive contracts if they provide incremental information about top executive decisions.

While Holmström tells us whether to incorporate a signal into a contract, he does not study how to do so. Our paper, How Should Performance Signals Affect Contracts? (forthcoming in the Review of Financial Studies) gives practical guidance for incorporating such signals into compensation contracts.

We point out that a signal can affect two dimensions of compensation contracts:

  • The minimum performance “threshold” that the executive must beat for pay to be positive. For a stock option, this is the strike price.
  • The sensitivity of pay to performance above this threshold. For a stock option, this is the number of options received by the CEO.

That is, a signal can either affect the minimum threshold for the executive to be paid, or how much pay she receives for beating the threshold (or both).

We show pay-performance sensitivity above the threshold, e.g. the number of options received by the CEO, should depend on the precision of the stock price as a measure of the quality of the CEO’s decisions. For example, if the stock price is a very noisy measure of performance in a recession, the CEO should receive fewer options in a recession. Intuitively, incentives are concentrated in states of the world where the stock price is a good measure of the CEO’s decisions.

In practice, however, the number of vesting options often depends on other signals, for example direct signals of CEO performance such as high earnings or low carbon emissions. We show that this practice may be suboptimal. Instead of affecting the pay-performance sensitivity above the threshold, these signals should affect the threshold itself—e.g. the strike price of options, not the number of vesting options. This is because a good signal of CEO performance, such as low carbon emissions, should increase pay by a fixed amount (as long as the stock price is not so low that the CEO should be paid nothing to begin with), and this is achieved by reducing the strike price. Changing the number of options would mean that the CEO’s “bonus” for low carbon emissions depends on the stock price rather than being a fixed amount.

We conclude that the current practice of performance vesting, where the number of options ultimately received by the manager depends on measures of managerial performance, is suboptimal. Most performance measures should instead affect the strike price of options.

In addition to options, our analysis applies to any other component of incentive pay where the executive receives zero if performance falls below a threshold. Other examples are bonuses, which are zero unless a performance threshold is hit, and performance shares where the executive receives no shares if performance is sufficiently poor. Most performance measures should affect the threshold required to trigger a payment, rather than how fast pay rises with performance if the threshold is hit.

The complete paper is available for download here.

Both comments and trackbacks are currently closed.