The following post comes to us from Peter Cebon of the University of Melbourne and Benjamin Hermalin, Professor of Economics at the University of California, Berkeley. Work from the Program on Corporate Governance about CEO pay includes: The CEO Pay Slice by Lucian Bebchuk, Martijn Cremers, and Urs Peyer (discussed on the Forum here); Paying for Long-Term Performance by Lucian Bebchuk and Jesse Fried (discussed on the Forum here); and Lucky CEOs and Lucky Directors by Lucian Bebchuk, Yaniv Grinstein and Urs Peyer (discussed on the Forum here).
Our paper, When Less Is More: The Benefits of Limits on Executive Pay, forthcoming in the Review of Financial Studies, addresses the question of whether limits on executive compensation harm or benefit shareholders. In particular, our model shows that if regulation limits executive compensation, this can make it possible for the board to give the CEO incentives that are both more effective and less costly, and for the two parties to create a relationship that is more collaborative. Among the implications—some of which we are exploring in a companion paper in progress—is this collaborative relationship makes it more attractive for the CEO to pursue long-run strategies (e.g., organic growth) that are more profitable than the short-run strategies (e.g., mergers and acquisitions) they would have pursued if firms had to rely on stock-based compensation for their executives.
The idea behind our results is the following: ideally, as in most agency models, the shareholders (or their representatives, the board of directors) would like to pay executives based on their actions, not those actions’ stochastic outcomes. We assume, however, an informational friction prevents that: although the directors can observe the executives’ actions, those observations cannot be used to enforce a formal contract. Yet, because the board and the executives play repeatedly, the firm may be able to overcome this problem via reputation: the board promises to honor the terms of an agreement and, even though not legally enforceable, that promise is credible because there is a net loss from reneging; a board that reneges today can’t enter into similar agreements, with their attendant benefits, in the future—future executives would not see offers of such agreements, known as relational or informal contracts, as credible.
The cost of losing credibility, and hence the deterrence from reneging, depends on the value of a series of formal contracts which tie executives’ compensation to firm performance. They are the next-best alternative to relational contracting. The greater the firm’s profits from formal contracts, the greater the board’s temptation to renege on a relational contract. If formal contracting becomes too attractive, a fully efficient or profit-maximizing relational contract becomes impossible. The temptation to renege will simply be too great and, because executives would anticipate the board will renege, such a contract fails to provide them incentives. In such a situation, state-imposed restrictions on formal contracts can be beneficial: by making formal contracting less profitable, the temptation to renege on a relational contract is reduced, which permits the use of relational contracts. It is important to emphasize that reneging often produces higher short-run profits but lower overall profits.
These results are robust. State-imposed limits can also increase profits if the board offers the executives a contract with both formal and relational components, rather than just a relational contract, and if the cap is on total compensation rather than just contingent pay. The results also hold if the shareholders possess all the bargaining power—i.e., our argument is unrelated to arguments to limit executive pay to prevent executives lining their pockets at shareholders’ expense.
Our basic findings have some interesting implications:
- The results depend on the relative value of relational and formal contracts. Because this is firm specific, our results are nuanced: in some circumstances a prohibition on contingent compensation would be beneficial; in others, caps, rather than outright bans, would be optimal; in others, limits have no effect, as formal contracting poses no “threat” to relational contracting.
- It helps explain the prevalence of formal incentive (e.g., stock-contingent) contracts despite complaints that they are not as effective as desired, or overly reward managers given what they achieve. It may be that boards cannot utilize the optimal relational contract absent state-imposed restrictions.
- Under the model, shareholders often prefer relational contracting and, therefore, favor caps on contingent compensation while executives prefer formal contracts, which allow them to earn a rent. These opposing preferences are reflected in the actual public policy debate over restrictions on pay, with shareholder advocates on one side and executives on the other. It is also reflected in the location of debate—in the legislatures rather than company meetings.
- The basic model rests on the assumption the board can perfectly monitor executive behavior. In an extension, we argue that it depends on how well the directors understand the processes by which the company creates value, which in turn, reflects an investment decision. We discuss three implications:
- In some companies, a skilled board will not be cost-effective, and so the shareholders optimally forgo having such a board and rely on formal contracts. We expect formal contracts to dominate in companies where the costs of a high quality board are high, and the benefits are low, such as with hedge funds. In contrast, we expect relational contracts to dominate when the costs are low and the benefits are high, as with an electric utility at the dawn of the solar age.
- Notwithstanding, the benefit of having a capable board can be a function of whether contingent compensation is capped. If it is capped, then the value of a capable board is greater than if it is not.
- If board quality is endogenous, we should expect to see higher board quality when contingent compensation is capped; or, somewhat conversely, a negative correlation between board quality and levels of contingent pay.
- If we draw these strands together, it points to CEO compensation caps as one of the levers that changes the mix in the economy between innovation-oriented long-run players with high-skilled boards and transaction-oriented short-run players with low-skilled boards.
The full paper is available for download here.