The mathematics of executive compensation.

In 2006-that was just two years ago, when Wall Street still existed-the fifth highest-paid CEO in the United States, according to Forbes' annual roundup, was a man named Richard Fuld. Fuld sold enough shares in his company to make himself $122 million that year. This wasn't unusual for him: In five years, he'd reaped close to $376 million in total pay.

The company that Fuld ran once was one of the leading investment banks in the country. It was called Lehman Bros. It is now defunct.

In the annals of outrage over executive pay, Richard's Fuld's nine-figure payouts will occupy a fat chapter. It's people like Fuld who make a bailout of financial companies so hard to swallow for many-and why the bailout bill comes with caps on CEO pay. How could someone who messed up so badly have gotten paid so much for doing that?

An even better question: How come this kind of thing happens so often? Scanning the ranks of the highest-paid CEOs, one discovers a surprising number of them come from the losingest big corporations. Look, there's Yahoo's Terry Semel, in the top 10 for both 2006 and 2007. Hey, over there-Countrywide's Angelo Mozilo! No. 10 in 2006 and, stunningly, the seventh best-paid CEO in 2007, well into the mortgage bust.

In fact, the very nature of CEO compensation almost guarantees that this pattern will repeat itself over and over. During a company's successful years, the chief executive's compensation will look fairly reasonable, his stock price will be rising, and he'll be hailed as a corporate hero. Then his pay will rise as his company tanks. Imagine a chart of share prices rising quickly and then turning down. Now superimpose on it a chart of chief executive pay, rising first slowly and then shooting up like a glorious hockey stick.

There's no official name yet for this oft-repeated phenomenon, so we'll call it the Fuld Curve. Naming it after him seems like the right sort of honor-one of the few he can't buy.