Oct. 17 (Bloomberg) -- A large stock option may motivate a chief executive officer to such great risks that he ends up making decisions that can backfire for shareholders.

That's the conclusion of two business school professors in their paper published by the Academy of Management Journal: ``Swinging for the Fences: The Effects of CEO Stock Options on Company Risk-Taking and Performance.''

Wm. Gerard Sanders, the lead author, teaches at Brigham Young University in Provo, Utah. Co-author Donald Hambrick is at Pennsylvania State University in State College, Pennsylvania, and a former professor at Columbia University's Graduate School of Business in New York.

Their principal thesis is: Load up a CEO's pay package with too many options, and he will make risky decisions that promise outsized rewards for shareholders and, of course, for himself.

That those decisions can backfire and cause substantial losses may worry shareholders, but hardly bothers the CEO because his worst case is that his options will expire underwater. He won't lose any real dollars.

According to the professors' research -- and this is the big news here -- the possibilities of outsized gain or outsized loss aren't equal. Their analysis of companies where 50 percent or more of the pay package comes from option grants (measured at the grant date using the Black-Scholes model) shows that 6.8 percent of the cases produced large gains for shareholders while 10.1 percent produced large losses.

No Explanation

Sanders and Hambrick have no sure explanation for the lopsided results they obtained from studying data from 950 companies from 1993 through 2001.

``Our data do not allow any insights as to why this inferior performance comes about, but we anticipate that it is because option-loaded CEOs are riveted on upside possibilities, with little concern for downside,'' they write.

My initial suspicion was that the lopsidedness had something to do with the high-tech companies included in their study, which generally have the largest ratio of options to other pay and which underwent a boom in 1999 and a crash in 2000.

Not so, Sanders says.

``The results hold if we drop 2000, or both 1999 and 2000,'' he responded to me in an e-mail. ``In addition, we have looked at the identity of the firms that comprise the lopsided negative outcomes and they range from retail drug stores, health care companies, industrial companies to a couple of high-tech companies.''

With a finding that the side effects of the stock option drug seem to overwhelm its medicinal value, it's not surprising that the professors, although not claiming that options are useless, believe they shouldn't be so prominent in CEO pay packages. They seem to favor other forms of compensation, such as free stock awards.

Free Shares

Their rationale is that, with free shares, there is a real downside for the CEO, i.e., loss in share value. That negative, they figure, will cause him to be less risky in his decision- making.

As they put it: ``CEO shareholdings seemed to promote a more prudent type of risk-taking than was generated by stock options.''

In other words, the imbalance between big gains and big losses was much less.

There are side effects to this medicine, too. Using free shares might well reduce extremes of performance, but the thought that a CEO could nonetheless earn millions for negative shareholder returns is troubling to say the least.

I grant that a CEO who comes in from the outside may aim for the fences if he doesn't own any of his new company's shares and he gets a mega-option grant. In most cases, though, CEOs are holding a bunch of real shares. And they are also holding plenty of past-granted options that may be substantially in-the-money.

Median CEO

In a study I did this year of 542 U.S. companies with market caps of at least $3 billion during 2006, I found that the median CEO, who had 4.8 years' tenure at the end of 2006, was holding $18 million of real shares (i.e., not option shares). He was also holding unexercised options granted in past years that contained a paper profit of another $14 million. (Data were obtained from Equilar Inc.)

So, at least for the median CEO in my study, you can't effectively argue that he can go for the fences with impunity, knowing he will lose nothing if he fails. He stands to lose as much as $32 million in actual shareholdings and paper profits from past option grants.

The foundation of Sanders's and Hambrick's arguments rest on the ``economic man,'' whose sole goal in life is to maximize his income while turning his back, if necessary, on spouse, children, co-workers -- and even sleep.

There has to be an effective limit to monetary motivation. If you are sitting on 50 million option shares granted in the past 10 years and you have just been given options on 10 million more shares, will you really work harder or smarter because of more options?

To contact the writer of this column: Graef Crystal in Las Vegas at at graefc@bloomberg.net.

Last Updated: October 17, 2007 00:11 EDT