Executive pay and the Prisoner's Dilemma

Excessive executive salaries are in the news, especially with the moves by the president and congress to limit salaries in companies receiving bail-out money. Devilstower at Daily Kos writes that CEO pay is the problem - echoing comments by Roger Lowenstein in an NPR interview. The basic argument is that, when somebody makes a lifetime's worth of money in a year or two, they have little incentive to do things that are of long-term value. Actions that are high-risk and of short-term benefit are taken regardless of possible long-term calamities they may cause.

This fits in with my understanding of Game Theory, particularly the old iterated Prisoner's Dilemma problem. In physics it seems like almost every theory of some complex mode of behavior boils down to a variant of the "harmonic oscillator" problem; I have a feeling that the Prisoner's Dilemma may serve the same role for economics and human behavior. It captures a fundamental non-intuitive situation: where cooperation is clearly beneficial to the group as a whole, but individual incentive always works against cooperating.

That's the case for the single instance of a Prisoner's Dilemma problem: defection is always the logical choice, producing the worst outcome for both players. But the iterated problem assumes the same actors play repeatedly, and consideration of both past and future cooperation enters into the logic of decision-making. It turns out the best iterated strategy is a simple Tit for Tat response: start by cooperating, and then do whatever the other player did the previous turn. The result is the evolution of overall largely cooperative responses among players, bringing about the best possible collective results for the group.

The lesson here is that the need for consideration of consequences over the long term must be an essential part of human motivation to produce good, cooperative outcomes. If incentives are only short-term or one-time, and not long-term, we end up with destructive, competitive, antagonistic cycles that do little good for society as a whole. Ponzi schemes are examples of completely short-term thinking: they fundamentally have no long-term plan to restore funds to investors. But similar issues arise with the incentives for executives of more normal companies, relative to those who invest in them.

On executive compensation, what would it take to shift that focus from short-term to long-term success? At minimum wage, a US worker would make close to $20,000 in a year; poverty level for a family of 4 is considered right now to be about $21,000. Successful companies obviously should be paying their executives at least that much! But do we need an upper limit, and what should it be?

A lifetime of work is about 40 years; receiving a lifetime's worth of minimum-wage or poverty-level salary in one year is likely not enough to remove an executive's personal concern about the long-term health of their company, but a lifetime of double minimum might be close. That's $1.6 million, at current rates. If you want to prevent that kind of attitude, you need a cap that's under that amount. The present proposals of around $500,000 for bailout recipients should certainly do the trick. For public companies it might be time to look into setting a $1 million cap. How would you enforce that? Returning to the old 94% tax rates for income (and capital gains!) at the high end could make a lot of sense...

But of course the argument on the other side is that high compensation motivates people to take necessary risks, the entrepreneurial spirit, and all that. I'm not terribly convinced that, once you get over the $1 million/year level, there's much motivation left in the extra money flowing in. What do you do with it all? But there's certainly legitimate grounds for concern that imposing a limit like that could have negative effects that might outweigh the speculated positive effects from imposing a more long-term view on executives.

Well, sounds like what we really need are some studies on the issues. A comparison of the long-term success of various corporations with different compensation levels for executives might tell us something. I'll be on the lookout for such studies and expect to post further thoughts on the issue. Meanwhile, it'll be interesting to see how the current mandate proposals fare, and what effect they have...