Abstract
We empirically examine two competing views of CEO pay. In the contracting view, pay is used to solve an agency problem: the compensation committee optimally chooses pay contracts which give the CEO incentives to maximize shareholder wealth. In the skimming view, pay is the result of an agency problem: CEOs have managed to capture the pay process so that they set their own pay, constrained somewhat by the availability of cash or by a fear of drawing shareholders' attention. To distinguish these views, we first examine how CEO pay responds to luck, observable shocks to performance beyond the CEO's control. Using several measures of luck, such as changes in oil price for the oil industry, we find substantial pay for luck. Pay responds about as much to a "lucky" dollar as to a general dollar. Most importantly, we find that better governed firms pay their CEOs less for luck. Our second test examines how much CEOs are charged for the options they are granted. Since options never appear on balance sheets, they might offer an appealing way to skim. Here again we find a crucial role for governance: CEOs in better governed firms are charged more for the options they are given. These results suggest that both views of CEO pay matter. In poorly governed firms, the skimming view fits better (pay for luck and little charge for options) while in well governed firms, the contracting view fits better (filtering out of luck and charging for options).
Introduction
In the traditional economic model, executive compensation is a finely-tuned instrument. Shareholders (the principals) design "optimal contracts" to ensure that CEOs (the agents) work in the company's best interest. This "contracting view" suggests that every dollar of pay is a calculated incentive to solve the moral hazard problem.
However, Marianne Bertrand and Sendhil Mullainathan’s seminal 2000 study challenges this equilibrium. They propose a "skimming view," where CEOs in firms with weak oversight effectively capture the board and set their own pay. In this model, CEOs extract as much wealth as possible, limited only by the "outrage constraint"—the fear of attracting too much negative attention from shareholders or the public. To distinguish between these two views, the authors look at how CEOs are rewarded for "luck": performance gains driven by factors entirely outside of their control.
The "Luck" Test: Methodology
The researchers employ an Instrumental Variables (IV) approach to isolate luck from skill. If the contracting view holds, an efficient board should filter out exogenous shocks and only reward the CEO for performance they actually influenced. If the skimming view holds, CEOs will take credit (and cash) for any positive performance, regardless of the source.
To measure luck, the authors identify three specific exogenous shocks:
- Oil Price Movements: For firms in the oil industry, global price fluctuations significantly impact returns regardless of management quality.
- Exchange Rates: Industry-specific movements in currency values.
- Industry Benchmarks: Year-to-year differences in mean industry performance that reflect broader economic cycles rather than individual firm success.
By interacting these luck measures with corporate governance indicators—such as the presence of large shareholders (5%+ owners), board size, and CEO tenure—the authors can see if better oversight acts as a filter against skimming.
Key Insights and Results
1. The Reward for Luck
The study finds that, on average, CEOs are rewarded for luck just as much as they are for "earned" performance. In the oil industry, for instance, a 1% increase in accounting return due to oil price luck increases CEO pay by approximately 2.15%. This is statistically indistinguishable from the 0.82% pay increase associated with general performance. Essentially, a "lucky dollar" is worth just as much to a CEO as a dollar generated through strategic brilliance.
2. Governance as a Filter
The "pay for luck" phenomenon is not universal; it is a function of governance. In firms with at least one large shareholder on the board, pay for luck drops by 23% to 33%. This suggests that when a principal is actually watching the agent, the board is much more likely to filter out external factors and reward only genuine performance.
3. The Entrenchment Effect
The data reveals a troubling trend regarding CEO tenure. As CEOs stay at a firm longer, they tend to receive more pay for luck. However, this only happens in firms without a large shareholder. This supports the skimming view: over time, CEOs "capture" their boards and influence the compensation process, but a strong external monitor can prevent this entrenchment.
4. Options as "Invisible" Skimming
At the time of the study, accounting rules did not require stock options to be expensed on the balance sheet. This made options an ideal tool for skimming, as they provided significant value to the CEO without appearing as a cost to the firm.
In an optimal contract, a CEO should be "charged" for the value of the options they receive (i.e., their cash salary or bonus should decrease proportionally). The authors found that in well-governed firms, CEOs are indeed charged for these grants. In poorly governed firms, however, options are essentially "free money," with no corresponding reduction in other pay components. Adding a single large shareholder to the board increases the "charge" for a $1 million option grant by approximately $50,000.
Conclusion and Policy Implications
The findings suggest that both the contracting and skimming views are valid, but their relevance depends on the firm's governance structure. In well-governed firms, the contracting view prevails: luck is filtered out, and CEOs are charged for their options. In poorly governed firms, the skimming view is a better fit: CEOs are rewarded for external shocks and receive "invisible" pay through un-expensed options.
This research had significant policy implications, particularly regarding the transparency of executive pay. The authors noted that forcing stock options onto balance sheets (a change that eventually occurred with FAS 123R in 2004) would likely reduce the ability of CEOs to skim by making the true cost of their compensation visible to shareholders. Ultimately, the paper serves as a reminder that without a strong principal to oversee the agent, the agent will eventually start writing their own paycheck.
Summary updated