Cowen on Optimal Marginal Tax Rates on CEOs
By David Henderson
A week ago, I sent to my editor my review of Tyler Cowen’s latest book, Big Business: A Love Letter to an American Anti-Hero. The book is outstanding. There are valuable facts and/or bits of economic reasoning on virtually every page. To give you an idea of how much I liked the book, I titled my review “A Love Letter to Tyler Cowen.” I’m not sure the editor will use that title. 🙂
The consistently great content meant that I had to leave a lot out of my review.
Here’s one part I left out but found striking. It’s on page 50, in the chapter on whether CEOs are paid too much. (Tyler says they’re not.):
Another factor is that, on average, top CEO talent helps larger firms proportionally more than smaller firms, and high salaries can be useful as a means of allocating the best talent to their most important uses. If Mark Zuckerberg had been running a midsized financial services company rather than Facebook, that would have been a waste of his talents, and likely Facebook would not have taken off as it did. A study found that when we take such “matching” factors into account, the optimal highest marginal tax rate on CEOs probably should be in the range of 27 to 34 percent. If the tax is much higher, the returns on making the right CEO-to-firm match will be smaller and productivity will be lower, and some of the star performers will end up at insufficiently important firms.
The study he refers to above is Laurence Ales and Christopher Sleet, “Taxing Top CEO Incomes,” American Economic Review, Vol. 106 (11): 3331-3366.
That reminds me of a point I made in a special seminar for some hand-picked, up-and-coming U.S. Navy and Marine Corps officers about 15 years ago. I had been asked to comment on some of their thoughts. One of their questions was “How can we get the best people into the Navy and Marine Corps?” I said that I thought it would be a mistake to get the best people in the Navy and Marine Corps. I pointed out that if the Navy had managed to persuade Bill Gates to become an officer in the mid-to-late 1970s, the PC revolution would have probably slowed by at least a few months. “Think,” I said, “of the enormous loss of productivity and consumer surplus from even a few months’ loss.”