I asked Cornell economist Robert Frank to comment on my earlier post about limiting Wall Street salaries. He wrote back this afternoon, arguing that finance is a special case requiring more aggressive interventions than would normally be wise. In particular, he argued that there's another reason to disproportionately high pay in finance: It steers people away from more productive endeavors:
The problem with high pay in the financial industry isn't just that it tempts people to do illegal or unethical things. It also sucks a lot of talent out of other occupations in which it would do vastly more good. (I explain this point in some detail in the attached file.) In general I think we should let markets sort out pretax pay, and if we don't like the resulting distribution, we should then use the tax system to change it. That's because we generally benefit from a price mechanism that assigns the most talented people to the most important jobs. IBM has annual earnings of about $10 billion. If it can hire someone who will improve its bottom line by just 3 percent, that's $300 million better for both the company and for society as a whole. Executives would obviously be willing to work hard for a tiny fraction of today's salaries (they always used to and still do in many other countries). Market-determined pay is important not because it elicits greater effort but only because it steers the right people to the right jobs. IBM will be less likely to be able to attract the most talented executives if it pays the same salary as smaller companies. But because it's primarily relative pay that people care about it, the current allocation mechanism would work equally well even if the top marginal tax rates on income were 80 or 90 percent. With a tax system like that in place, everyone would have reason to celebrate when an executive signed a lucrative contract. As far as allocative efficiency is concerned, the finance industry is the glaring exception to all. Unlike most other industries, where company profits are closely linked to social value added, there is virtually no such link in the financial industry. When we send smarter people to financial firms, they just figure out more devious ways to increase leverage and unload hidden risk on the rest of the system. They make the total economic pie smaller, not bigger. So an incentive system that steers society's brightest to the financial industry is actually counterproductive. We really don't need smart people to do the essential service that this industry needs to perform, which is to make capital available to worthwhile investment projects. There is almost never a worthwhile investment that fails to happen for lack of finance. Most projects can be operated initially on a small scale (Jobs and Wozniak, for example, started in their parents' garages). If you have a killer idea and a bank won't lend to you, a group of friends can, say, by taking second mortgages on their houses. Then once the idea's viability is clearly established, traditional lending institutions will stumble all over themselves to provide additional capital. Geniuses aren't required here. So in the financial industry, we actually have the luxury of entertaining the possibility of adopting one of the populist pay cap proposals that would be folly in almost any other industry.
In general, I read Frank as arguing for a pretty explicit tradeoff here: Interventions into market pay are bad. Letting financial compensation continue to inflate the relative attractiveness of the industry, however, is worse. I don't know that this really gets to the question of how, precisely, you go at these earnings. But it's a fair argument for why you want to get at them.
