One of the most tired defenses against criticism of executive pay gone wild is the comment that athletes and pop stars earn outrageous sums of money, so why not executives? It's exactly that argument that Marc Hodak uses in a column for Forbes: "When rock stars make big bucks, we can look at the ticket and album sales and understand where it comes from. But when a CEO makes rock star income, we figure he must be scamming the shareholders."

Here's why that analogy doesn't work: when an owner of a sports team decides to spend $100 million on a superstar shortstop, that's his choice. It's his money. He owns the team, and he's entitled to do whatever he wants with it. Similarly, if 12-year old girls want to collectively spend $100 million on Menudo posters (And who can blame them!), that's their prerogative.

But what if the owner of a sports team was forced to spend money on players based on the whims of retired politicians and economics professors who serve on 15 other boards, collecting $100,000 from each company in exchange for going to a couple meetings a year and have no particular stake in the outcome of the investment? And what if these decision-makers could only be voted out once every few years, but the ownership of the team was so fragmented and most of the owners were so inattentive that it was nearly impossible to get them replaced?

That is, in effect, the situation we have in executive compensation. If the owners of public companies actually had any meaningful say in how much CEOs were paid, the sports star analogy would work. Since they don't, it doesn't. Executive pay is a classic principal-agent problem, and it's one that can only be solved through improved corporate governance.