This classic paper from the year 2000 by Marianne Bertrand and Sendhil Mullainathan dives into the murky waters of executive compensation. It challenges the traditional economic view that CEO pay is always a finely-tuned instrument designed to align the interests of managers and shareholders. Instead, the authors suggest that in the absence of strong oversight, CEOs essentially "skim" wealth from the companies they lead.
TL;DR
CEO pay is often driven by "luck"—factors like oil prices or exchange rates that are entirely outside a manager's control. However, this "pay for luck" is significantly lower in firms with strong corporate governance, suggesting that CEOs in poorly-governed firms effectively set their own pay.
What Problem Does This Paper Address?
In economic theory, the Principal-Agent problem describes the tension between shareholders (principals) and CEOs (agents). Ideally, boards of directors create "optimal contracts" that reward CEOs for hard work and smart decisions.
However, there are two competing views:
- The Contracting View: Pay is an incentive tool to solve moral hazard.
- The Skimming View: CEOs have "captured" the pay-setting process. They use their influence over the board to extract as much cash as possible, limited only by the fear of drawing too much negative attention.
The authors seek to distinguish these views by looking at how CEOs are rewarded for performance they didn't actually cause.
Methods
The researchers employ an Instrumental Variables (IV) approach to isolate "luck" from "skill." * Defining Luck: They identify three exogenous shocks: movements in world oil prices (for oil companies), changes in industry-specific exchange rates, and year-to-year differences in mean industry performance. * The Two-Stage Test: First, they predict firm performance based on these "luck" variables. Second, they measure how sensitive CEO pay is to this luck-induced performance compared to general performance. * Governance Interactions: They interact these measures with governance indicators—such as the presence of large shareholders (5%+ owners), board size, and CEO tenure—to see if better oversight filters out the luck. * The "Charge for Options" Test: They examine whether other components of pay (salary/bonus) decrease when a CEO is granted stock options. In an optimal contract, a CEO should be "charged" for the value of the options; in a skimming model, options are "free money" because they don't appear on the company's balance sheet.
Datasets
- Oil Industry Data: 51 of the largest US oil companies (1977–1994).
- The "Yermack" Dataset: 792 large US corporations (1984–1991), which includes detailed information on stock option grants and board composition.
- COMPUSTAT: Used for industry-wide performance benchmarks and financial data.
Key Insights and Ideas
- Luck is Rewarded: On average, CEOs are rewarded for luck just as much as they are for "earned" performance. A "lucky dollar" is worth as much to a CEO as a dollar generated through hard work.
- Governance as a Filter: This is the "aha" moment. The "pay for luck" phenomenon isn't universal. In firms with a large shareholder on the board, pay for luck drops by 23% to 33%.
- The Entrenchment Effect: As CEOs stay at a firm longer (increasing tenure), they tend to get more pay for luck—but only if there is no large shareholder to keep them in check. This suggests CEOs "capture" their boards over time.
- Options are Skimming Tools: Because accounting rules (at the time) didn't require options to be expensed on the balance sheet, they were an "invisible" way to increase pay. The paper finds that poorly governed CEOs are charged much less for the options they receive than CEOs in well-governed firms.
Key Results
- Sensitivity: In the oil industry, a 1% increase in accounting return due to oil price "luck" increases CEO pay by about 2.15%. This is statistically indistinguishable from the pay increase for general performance (0.82%).
- Governance Impact: Adding a single large shareholder to the board increases the "charge" for a $1 million option grant by approximately **$50,000**.
- Board Size: Larger boards (often seen as less efficient) are associated with more pay for luck and lower charges for options.
Limitations and Future Work
- Data Constraints: The study could not account for the change in value of existing options held by CEOs, only new grants. This likely understates the total "pay for luck" in the system.
- Asymmetry: The authors find some evidence that CEOs are rewarded for good luck but not necessarily punished for bad luck (the "heads I win, tails you lose" scenario), though they suggest this requires more formal study.
- Policy Implications: The paper concludes that forcing stock options onto balance sheets (which eventually happened with FAS 123R in 2004) would likely reduce the ability of CEOs to skim.
Summary: This paper serves as a reminder that economic models of "optimal" behavior only work when the underlying power dynamics are balanced. Without a "principal" (like a large shareholder) to watch the "agent," the agent will eventually start writing their own paycheck.
Summary updated