A better way to pay CEOs

Smarter incentives could reduce the risks they pursue.

Opinion

For Wall Street CEOs, December was the best of times, and it was the worst of times. Goldman Sachs gave Lloyd Blankfein a $68 million bonus. Rival Wall Street firm Morgan Stanley, however, gave John Mack (at his request) zip. As elsewhere, the best compensated are now leaving everybody else far behind, even within the ultraelite club of Wall Street chiefs.

No, I'm not calling for limits on CEO pay. Capitalism thrives on incentives, and compensation ceilings distort those incentives. But there is a better way to pay CEOs, by using incentives to refocus them on building longer-term, sustainable company value.

Mr. Blankfein's pay rewarded him for Goldman's outstanding performance, and for his good judgment in supporting prescient colleagues who sensed the risk of meltdown in the subprime mortgage lending market. Indeed, Blankfein may even have been undercompensated, given the quality of his decisions relative to those of many of his peers who have taken it on the chin with multiple billion-dollar writedowns. The spectacular industry-wide failure in risk management relating to subprime mortgage securities has thus far taken down the CEOs of Citigroup and Merrill Lynch, and vaporized the bonuses of Bear Stearns' chief, as well as Morgan's Mack.

It's hard to find a silver lining in the subprime mortgage securities cloud, though not for lack of trying. Some see, in the unprecedented stratification of incomes at the pinnacle of Wall Street, a validation of corporate America's approach to executive compensation. The "pay for performance" system worked, it's claimed. After all, each of the banks' CEOs got what he deserved: extravagant reward for the few who orchestrated the creation of substantial new value for shareholders and zero bonus (or even pink slips) for those who presided over massive value destruction.

"Ultimately, accountability for our results rests with me, and I believe in pay for performance, so I've told our compensation committee that I will not accept a bonus for 2007," Mack said in a statement.

That salutary message was sent out loud and clear throughout the industry. Nobody is entitled to a seven- (let alone eight-) figure payday – a CEO has to earn it.

But this emerging conventional wisdom is off the mark. Far from being silver lining, it was precisely the prevailing executive compensation model that drove Wall Street to place such big bets in the subprime mortgage market. The underlying loans were made on Main Street. But Wall Street took huge positions in this high-risk market mainly because its CEOs had (and still have) extraordinarily high incentives to roll the dice in their efforts to drive up company stock price.

Since the mid-1990s, American CEOs have been paid primarily in megagrants of stock and stock options – especially high-octane options – on the theory that having lots of "skin" in the game better aligns CEO interests with those of the firm's shareholders than would a large base salary. As is well known by now, these high-powered incentives have, in recent years, prompted a small minority of corporate CEOs to cheat by falsifying their accounting. But far more broadly, they have enticed honest CEOs to gamble imprudently with (mostly) other people's money, because it makes sense within these pay structures to play for unlimited upside while – at least in terms of compensation – there is a floor to downside risk.

Investing huge sums in subprime mortgage securities held out the prospect of massive returns if CEOs bet right. True, as Morgan's Mack and many of his peers have learned the hard way, betting wrong isn't cost-free. But with sky-high upside if the gamble pays off, and downside risk limited to a foregone bonus – or even, in extreme cases, a firing softened by the lucrative exit packages now taken for granted – there isn't enough deterrent to reckless investments.

A better compensation system would set CEO pay based on current-year company performance, but would hold back the bulk of it (in a good year) on paper as only a tentative payable. That credit balance would be subject to meaningful reduction in the event of a subsequent poor year. The Harvard Management Company, which has had enormous success investing a university endowment that has now grown to $35 billion, has compensated its executives in this manner for many years.

It may sound a tad cynical to presume that corporate leaders weigh the likely impact on their own wallets when making decisions about the best course of action for the company. But without that premise, we wouldn't need incentive-based compensation in the first place.

• Andy Zelleke is a lecturer in public policy at Harvard University's Kennedy School of Government and codirector of its Center for Public Leadership.