Bruce Bartlett writes (from behind the NYTimeSelect firewall, unfortunately),

[Alan] Reynolds has been criticized for making too much of inherent limitations in the data that have been known for many years and for ignoring evidence that conflicts with his thesis. Most economists would probably agree with Gary Burtless of the Brookings Institution…there are too many different sources all showing a rise in income inequality…No matter how you slice it, the distribution of income has become more unequal over the last 20 years or so…

If it were costless to play Robin Hood and take from the rich and give to the poor, it would be hard to oppose. But there are costs. We really don’t want the Gateses of the world sitting around clipping coupons. We want them out there thinking of new products and businesses

I agree with Bartlett that we might as well concede that inequality has increased, and instead focus on the costs and benefits of going after high CEO pay. To me, the cost is the loss of market signal.

Suppose you set a limit on CEO compensation of, say $2 million a year (leaving aside the enforcement challenges that Bartlett highlights in his post). Then, once a CEO has hit the limit, she has no incentive to take a job with a different company. By the same token, if she is incompetent, her Board may hesitate to fire her, because they may have no way of luring anyone else to take her job. If the compensation ceiling works, you’ll wind up with a lot more situations where a company is run by the wrong CEO.