25 Oct 2006  Views on News

Fixing Executive Options: The Veil of Ignorance

Who says you can't rewrite history? Dozens of companies have been caught in the practice of backdating options for top executives. But this is only part of the problem with C-level compensation packages, which often motivate top executives to act in their own best interests rather than those of shareholders. Professors Mihir Desai and Joshua Margolis turn to philosopher John Rawls for a solution: Reward the execs, but don't give them the details. Key concepts include:

  • Too often executive incentive packages are not aligned with the best interests of shareholders. Why create long-term value if your bread is buttered by quarterly performance?
  • Option compensation could be restructured to ensure that managers were aware of the value of their compensation without any knowledge of the details of their compensation—a concept inspired by philosopher John Rawls' work on distributive justice.
  • These options may only be useful for CEOs, senior officers, and directors—not middle management.


The latest corporate governance crisis is buried in the details of executive compensation contracts. Don't like the timing of the stock option grant you got or the strike price of the contract? No worries! It turns out that this is nothing an eraser can't fix.

While the full scope of backdating option practices remains unknown, this most recent scandal has deepened the sense in many quarters that option contracts given to managers distort behavior in destructive ways. The ability to play with, and respond to, the many variables in an option contract—the timing of the grant, the strike price, the vesting period—appears to vitiate many of the benefits of incentive alignment. How might option compensation be refashioned to deliver the benefits without the distortions?

Perhaps the answer to these problems can be found with philosophers rather than financial or accounting experts. John Rawls developed an influential body of work about justice, starting from the premise that ignorance might just be bliss. His theories were designed to consider a monumental, if esoteric, question: If you were in charge of the world and had to live in that world, how would you decide who got what?

Rawls' insight was that the fair solution could be determined by creating uncertainty about the position the person choosing the outcome would find him- or herself in. In short, the person cutting the cake has to decide the size of the slices and where at the table to place each slice before knowing where he or she will be sitting. Without knowing where you'll end up, you have to really think through what's a fair allocation.

What in the world does this theory of distributive justice have to do with option compensation? Option compensation could be restructured to ensure that managers were aware of the value of their compensation without any knowledge of the details of their compensation. In effect, CEOs and senior management would be kept behind Rawls' renowned "veil of ignorance" with respect to the precise strike price of their contracts and the timing of vesting.

Compensation committees would provide senior management with the value of their incentive compensation and examples of combinations of prices and vesting periods that correspond to the value of their contracts. But, the precise timing of grants, strike prices, and vesting periods would be kept hidden until vesting, or possibly until they leave the firm.

With Rawlsian options, managers would retain their high-powered incentives and know that large fractions of their wealth were invested in their companies. But they would not know all the thresholds that are critical for the value creation. The temptation to adjust the details of grants (e.g., "I think now is a great time to load up on my option compensations so let's do a big grant now…") or to respond opportunistically to those details (e.g., "I really need an extra penny of earnings this quarter as this is when everything vests…") is tremendous given the amounts of wealth involved. Rawlsian options would keep the fox out of the hen house but keep the fox hungry.

Such a solution preserves the virtues of options and only sacrifices some information sharing. Options still function as powerful incentives, constituting a sizable chunk of potential managerial wealth, and shareholders can remain assured that incentives are well aligned. Managers will not, however, know if they should try to accelerate earnings or delay them. They will not know what stock price they need to beat.

Managers will simply have to think through what the right thing to do is for shareholders as a whole. This uncertainty is a burden on managers and one that will force them not to choose investments, mergers, and accounting decisions based on the details of their contracts. Rather, they will have to consider what they should be considering—how to maximize value for their shareholders as a whole in the most effective way.

Drawing on theories of distributive justice to inform CEO compensation will surely be met with skepticism. From an accounting perspective, Rawlsian options would reinforce the need to expense options at their value at the time of grant and this will surely dissatisfy some. These options are also likely not well suited to middle management and may only be useful for CEOs, senior officers, and directors. Mechanisms would also need to be developed to ensure that a compensation committee withheld these details from managers. But, in just this one case, we think that managerial ignorance might well translate into shareholder bliss.

About the authors

Mihir Desai is the MBA Class of 1961 Fellow, and an associate professor at Harvard Business School.

Joshua Margolis is the Thomas S. Murphy Associate Professor of Business Administration at Harvard Business School.