April 9 (Bloomberg) -- Richard Fairbank took a risk. Shortly after becoming chief executive officer of Capital One Financial Corp. in July 1994, he shed base salary, bonus and benefits. What he got in return was option grants. Lots of them.

Was the risk worth it? Until now, yes for Fairbank. Also yes for shareholders who bought in the first few years after Fairbank got the top job. Otherwise, no.

Fairbank's performance measured from Nov. 16, 1994, the date of the McLean, Virginia-based credit-card issuer's initial public offering, through the close on April 4, 2008, has been good. Capital One Financial beat the annual return on the Standard & Poor's 500 Index by 8.6 percentage points. Yet Fairbank's performance deteriorates as his tenure narrows to the present.

In nine time windows, all ending Dec. 31, 2007, and ranging from a nine-year window to a one-year window, Capital One Financial underperformed the S&P 500 Index each time. The underperformance became greater as the time window narrowed. In seven of the nine periods, total returns were negative.

Over his tenure and through Dec. 31, 2007, Fairbank exercised options with aggregate gains of $530 million. At the end of that period, he also had unexercised paper profits of an additional $16 million. That's a staggering amount of money for performing relatively poorly over the past nine years.

Shortly after becoming CEO, Fairbank and his board compensation committee began to rewrite the rules for his pay.

More Risk

The first stock option he received in place of salary, bonus and benefits carried more than the usual risk. Although the strike price was set at the market price at grant and the option had a term of 10 years, it was subject to full cancellation unless Fairbank delivered a minimum of 20 percent annual appreciation in stock price during the first three years following the grant.

In 1998, Fairbank took on even more risk. He gave up previously granted options that were substantially in the money for an even bigger grant of at-the-market options. That 20 percent per-year appreciation requirement was again attached.

Then Fairbank seemed to lose his appetite for taking on all that pay risk. For his 1999 option grant, the 20 percent feature was still there, but in the fine print there was an escape clause: If 20 percent appreciation couldn't be obtained over three years, the option would still become exercisable in the last year of its 10-year term.

In 2001, that escape clause was moved up to the end of the sixth year, and Fairbank also received a second large grant that simply vested over the first three years.

More Options

In 2003, all that 20 percent business was dropped, and Fairbank got a large option grant with three-year vesting. He also received for the first time the opportunity to earn thousands of free shares based on earnings-per-share growth. He eventually took title to 242,000 free shares.

In 2004 through 2006, all options vested over five years, with no performance requirements. And in 2007, the vesting period was dropped to three years.

While Fairbank was tweezing risk out of his pay package, there seemed to be no diminution in the size of his potential rewards. His option grant made in 2006 covered 595,000 shares, while the one made in 2007 covered 1.7 million shares.

That might have been alright if Fairbank had actually done something for his shareholders. But in 2007, diluted EPS sank to $3.97 from $7.62 in 2006, while total return for the year ended Dec. 31, 2007, was negative 38 percent.

That's not exactly the environment to justify an almost threefold increase in the number of option shares granted.

Too Much

Capital One Financial told shareholders that Fairbank's 2007 option was worth $17 million, $1 million less than it reported as the value of the grant made in 2006.

How can this be when the number of shares granted in 2007 was almost triple those granted a year earlier?

There are two reasons. First, very shortly after making the 2007 grant, Capital One Financial increased its quarterly dividend 14-fold. Other things being equal, a higher dividend yield results in a lower option present value.

Second, the company lowered the assumption of when Fairbank's option would be exercised to five years from seven years used in connection with his 2006 option grant. This also lowered the option's present value.

That change is curious because when a stock is performing poorly, if the option is ever exercised it's done very late in its 10-year term.

By my estimate, Fairbank's 2007 option was worth $21 million at its grant, not the $17 million reported by the company. But even granting an option with a $17 million valuation for a disastrous year is close to being $17 million too much.

(Graef Crystal is a columnist for Bloomberg News. The opinions expressed are his own.)

To contact the writer of this column: Graef Crystal in Santa Rosa, California, at graefc@bloomberg.net.

Last Updated: April 9, 2008 02:50 EDT