Executive Pay: What Does Luck Have to Do With It?

Fail to see the logic of CEO compensation? You’re probably overlooking the economic effects of recruitment, retention, and corporate governance
money falling

By Pierre Chaigneau

For those inclined to view big business as irresponsible, poorly governed, and out of touch, CEO compensation provides the biggest and juiciest target. The average annual pay for the top 100 Canadian CEOs hit a record $10.4 million in 2017. Last year in the U.S., CEOs at the largest companies received the highest compensation increases since the Great Recession. The trend has drawn the ire of some shareholder activists who have staged rebellions against company boards approving huge pay days to senior managers.

For economists, CEO compensation has been a source of spirited debate. Some say that high executive pay is merely a reflection of supply and demand and the stock-based incentives built into compensation packages. They point to standard economic models in which compensation packages are designed to encourage CEOs to maximize company value.

But that’s been hard to square with the empirical evidence that CEO compensation has as much to do with luck as with the value CEOs create. It has been shown that, notwithstanding the deliberations of board compensation committees, chief executives are often rewarded for factors outside their control. Such factors may be a favourable business cycle or a change in the price of key commodities, such as oil, that can shape the fortunes of key sectors of the economy.

How Recruitment and Retention Factor In

In a recent paper I co-wrote with Nicolas Sahuguet (HEC Montréal), we presented a model that allowed for the fact that compensation packages are designed to not only provide incentives but also to attract and retain key executives. Indeed, for many large firms, attraction and retention are even more important than general incentives related to CEO performance.

When firms compete for top executive talent, the compensation level necessary to attract and retain a CEO depends on a CEO’s talent but also on a firm’s valuation of CEO talent. This in turn depends on the firm’s size and, thus indirectly, on business conditions or “luck.” In a booming economy, for example, cyclical industries do more business, so that managerial decisions affect a larger business base and have greater impact.

Seen in this light, some interesting insights are revealed. For one thing, firms with better corporate governance — for example firms with an independent board or with large blockholders — do a more thorough job of monitoring. They are able to get more and better information on CEO performance and ability, which means they have the confidence to hire CEOs of uncertain quality and to dismiss those that underachieve.

Well-Governed Firms Can Take Risks

This is important because inexperienced CEOs arrive with a lot of uncertainty about how they’ll perform once at the helm and have to make high-stakes decisions. A few good or bad months often go a long way towards determining their value going forward. So companies with strong board members who can closely monitor their performance will take a chance on them and pay them based more heavily on performance — how well they build firm value — rather than on the vagaries of prevailing business conditions. Indeed, their performance is very informative about their talent, which is a priori still very uncertain, so that the outside offers they can get strongly depend on their performance. A firm that wants to retain such top executives will then need to raise their pay a lot when their performance is stellar. 

By contrast, experienced CEOs — say former General Electric CEO Jack Welch or onetime Hewlett Packard CEO Meg Whitman — arrive with extensive media recognition. In their case, their talent is already well-established, but the unknown is the possibility of disruptive change in business conditions. Such a CEO will likely be hired by a firm with more diffuse ownership and relatively weak monitoring abilities. Since board members of these companies kowtow to senior executives and are loathe to dismiss a CEO, they look for a known quantity. As the performance of these CEOs is not very informative of their talent, which is already known, their variable pay will mostly be based on business conditions (or “luck”) rather than performance.

Firms bear the costs of having hired a bad CEO but they do not reap all the benefits of having hired a good CEO

The model can also contributes to explaining the rise in CEO pay over the last decades, which coincided with an improvement in corporate governance. Hiring a CEO with uncertain talent is costly for firms for two reasons: one, they will need to adjust CEO compensation following a good performance for retention purposes; and two, the dismissal and replacement process of a CEO who turns out to be incompetent is costly. Firms bear the costs of having hired a bad CEO but they do not reap all the benefits of having hired a good CEO.

In this context, improvements in corporate governance reduce the cost of CEO turnover by facilitating the identification and dismissal of bad CEOs. Thus, they increase the value of a CEO with uncertain ability from the firm’s perspective. We show that, even if the improvement in governance practices only occurs in firms with the worst governance (for example via a diffusion of best practices), this increases CEO pay in all firms due a market equilibrium effect. The diffusion of best governance practices can thus partly explain the rise in CEO compensation.

This may not make you feel any better about seemingly high CEO compensation or why CEOs can be rewarded for favourable business conditions yet go unpunished for poor performance. As in any sphere of life, of course, some top executives will abuse their positions and extract undeserved benefits. As a whole, however, trends in executive compensation are consistent with fundamental economic forces.

— With Alan Morantz

Smith School of Business

Goodes Hall, Queen's University
Kingston, Ontario
Canada K7L 3N6

Follow us on: