The Perfect Payday --- Some CEOs reap millions by landing stock options when they are most valuable; Luck -- or something else?
By Charles Forelle and James Bandler
18 March 2006
The Wall Street Journal
(Copyright (c) 2006, Dow Jones & Company, Inc.)
ON A SUMMER DAY IN 2002, shares of Affiliated Computer Services Inc. sank to their lowest level in a year. Oddly, that was good news for Chief Executive Jeffrey Rich.
His annual grant of stock options was dated that day, entitling him to buy stock at that price for years. Had they been dated a week later, when the stock was 27% higher, they'd have been far less rewarding. It was the same through much of Mr. Rich's tenure: In a striking pattern, all six of his stock-option grants from 1995 to 2002 were dated just before a rise in the stock price, often at the bottom of a steep drop.
Just lucky? A Wall Street Journal analysis suggests the odds of this happening by chance are extraordinarily remote -- around one in 300 billion. The odds of winning the multistate Powerball lottery with a $1 ticket are one in 146 million.
Suspecting such patterns aren't due to chance, the Securities and Exchange Commission is examining whether some option grants carry favorable grant dates for a different reason: They were backdated. The SEC is understood to be looking at about a dozen companies' option grants with this in mind.
The Journal's analysis of grant dates and stock movements suggests the problem may be broader. It identified several companies with wildly improbable option-grant patterns. While this doesn't prove chicanery, it shows something very odd: Year after year, some companies' top executives received options on unusually propitious dates. (An explanation of the methodology is below.)
The analysis bolsters recent academic work suggesting that backdating was widespread, particularly from the start of the tech-stock boom in the 1990s through the Sarbanes-Oxley corporate reform act of 2002. If so, it was another way some executives enriched themselves during the boom at shareholders' expense. And because options grants are long-lived, some executives holding backdated grants from the late 1990s could still profit from them today.
Mr. Rich called his repeated favorable option-grant dates at ACS "blind luck." He said there was no backdating, a practice he termed "absolutely wrong." A spokeswoman for ACS, Lesley Pool, disputed the Journal's analysis of the likelihood of Mr. Rich's grants all falling on such favorable dates. But Ms. Pool added that the timing wasn't purely happenstance: "We did grant options when there was a natural dip in the stock price," she said. On March 6, ACS said that the SEC is examining its option grants.
Stock options give recipients a right to buy company stock at a set price, called the exercise price or strike price. The right usually doesn't vest for a year or more, but then it continues for several years. The exercise price is usually the stock's 4 p.m. price on the date of the grant, an average of the day's high and low, or the 4 p.m. price the day before. Naturally, the lower it is, the more money the recipient can potentially make someday by exercising the options.
Which day's price the options carry makes a big difference. Suppose an executive gets 100,000 options on a day when the stock is at $30. Exercising them after it has reached $50 would bring a profit of $20 times 100,000, or $2 million. But if the grant date was a month earlier and the stock then was at, say, $20, the options would bring in an extra $1 million.
A key purpose of stock options is to give recipients an incentive to improve their employer's performance, including its stock price. No stock gain, no profit on the options. Backdating them so they carry a lower price would run counter to this goal, by giving the recipient a paper gain right from the start.
Companies have a right to give executives lavish compensation if they choose to, but they can't mislead shareholders about it. Granting an option at a price below the current market value, while not illegal in itself, could result in false disclosure. That's because companies grant their options under a shareholder-approved "option plan" on file with the SEC. The plans typically say options will carry the stock price of the day the company awards them or the day before. If it turns out they carry some other price, the company could be in violation of its options plan, and potentially vulnerable to an allegation of securities fraud.
It could even face accounting issues. Options priced below the stock's fair market value when they're awarded bring the recipient an instant paper gain. Under accounting rules, that's equivalent to extra pay and thus is a cost to the company. A company that failed to include such a cost in its books may have overstated its profits, and might need to restate past financial results.
The Journal's analysis raises questions about one of the most lucrative stock-option grants ever. On Oct. 13, 1999, William W. McGuire, CEO of giant insurer UnitedHealth Group Inc., got an enormous grant in three parts that -- after adjustment for later stock splits -- came to 14.6 million options. So far, he has exercised about 5% of them, for a profit of about $39 million. As of late February he had 13.87 million unexercised options left from the October 1999 tranche. His profit on those, if he exercised them today, would be about $717 million more.
The 1999 grant was dated the very day UnitedHealth stock hit its low for the year. Grants to Dr. McGuire in 1997 and 2000 were also dated on the day with those years' single lowest closing price. A grant in 2001 came near the bottom of a sharp stock dip. In all, the odds of such a favorable pattern occurring by chance would be one in 200 million or greater. Odds such as those are "astronomical," said David Yermack, an associate professor of finance at New York University, who reviewed the Journal's methodology and has studied options-timing issues.
Options grants are made by directors, with details often handled by a compensation committee. Many companies make their grants at the same time each year, a policy that limits the potential for date fudging. But no law requires this.
Until last year, UnitedHealth had a very unusual policy: It let Dr. McGuire choose the day of his own option grants. According to his 1999 employment agreement, he is supposed to choose dates by giving "oral notification" to the chairman of the company's compensation committee. The agreement says the exercise price shall be the stock's closing price on the date the grants are issued.
Arthur Meyers, an executive-compensation attorney with Seyfarth Shaw LLP in Boston, said a contract such as that sounded "like a thinly disguised attempt to pick the lowest grant price possible." Mr. Meyers said such a pact could pose several legal issues, possibly violating Internal Revenue Service and stock-exchange listing rules that require directors to set a CEO's compensation. "If he picks the date of his grant, he has arguably set a portion of his pay. It's just not good corporate governance."
UnitedHealth called the process by which its grants were awarded "appropriate." It declined to answer specific questions about grant dates but noted that on all but two of them, grants were made to a b road group of employees.
William Spears, a member of UnitedHealth's compensation committee, said the October 1999 grant wasn't backdated but was awarded concurrently with the signing of Dr. McGuire's employment contract. Mr. Spears said a depressed stock price spurred directors to wrap up negotiations and get options to management. The board revised terms of the employment contract last year and will start making stock-option grants at a regular time each year, Mr. Spears added.
The SEC's look at options timing was largely prompted by academic research that examined thousands of companies and found odd patterns of stock movement around the dates of grants. One study was by Erik Lie of the University of Iowa. He found that share prices generally fell before option grants and rose afterward, with the result that recipients got options at favorable times. He concluded this was so unlikely to happen by chance that at least some grant dates had to have been filled in retroactively.
Another possible explanation for big rises in stock prices following grants is that executives knew favorable company news was coming and timed the grants just before it. But academics think timing for company news is a less likely explanation for the patterns, given the consistency of the stock climbs after grant dates. Also, for many of the companies the Journal examined, no obvious company news followed closely upon the option grants.
It's also possible companies sometimes award options after their stock has taken a fall and seems to them to be undervalued. In point of fact, the companies can't possibly know what the stock will do next, but that doesn't mean they might not feel confident enough about a recovery to think they are hitting a favorable time to grant options.
The use of stock options surged in the late 1990s as young firms that had bright prospects but little revenue used them to attract and pay executives. As dot-com and telecom shares exploded, stock options became a source of vast wealth.
They also grew controversial. Critics worried that big options grants tempted executives to do whatever it took to get the stock price up, at least long enough to cash in their options. At the same time, during a general bull market, the options sometimes richly rewarded executives for stock buoyancy that had little to do with their own efforts.
At Mercury Interactive Corp., a Mountain View, Calif., software maker, the chief executive and two others resigned late last year. Mercury said an internal probe found 49 cases where the reported date of options grants differed from the date when the options appeared to have been awarded. The company said it will have to restate financial results. The SEC is still looking at Mercury, said someone familiar with the situation.
Analog Devices Inc. says it reached a tentative settlement with the SEC last fall. It neither admitted nor denied that it had misdated options or had made grants just before releasing good news that would tend to push up the stock. The Norwood, Mass., computer-chip maker tentatively agreed to pay a $3 million civil penalty and re-price some options. CEO Jerald Fishman tentatively agreed to pay a $1 million penalty and disgorge some profits. Analog didn't make him available for comment. The company said it will not restate its financial records.
In some instances, backdating wouldn't be possible without inattentive directors, securities lawyers say. At one company the SEC is looking at, lawyers say, it appears that someone picked a favorable past date for an option grant and gave it to directors for retroactive approval, perhaps counting on them not to notice. In another case, the lawyers say, a space for the grant date appears to have been left blank on paperwork approved by directors, or dates were later altered.
Until 2002, companies didn't have to report option grants until months later. The Sarbanes-Oxley law, by forcing them to report grants within two days, left less leeway to retroactively date a grant.
The new rule reduced stock patterns suggestive of backdating, but didn't eliminate these altogether, according to a study by M.P. Narayanan and H. Nejat Seyhun of the University of Michigan. They found that companies report about a quarter of option grants later than the two-day deadline -- and that such delayed reporting is associated with big price gains after the grant dates. It is a pattern Mr. Narayanan calls "consistent with backdating."
Before the stricter rules, Brooks Automation Inc., a semiconductor-equipment maker in Chelmsford, Mass., gave 233,000 options to its CEO, Robert Therrien, in 2000. The stated grant date was May 31. That was a great day to have options priced. Brooks's stock plunged over 20% that day, to $39.75. And the very next day it surged more than 30%.
A June 7 Brooks report to the SEC covering Mr. Therrien's May options activity made no mention of his having gotten a grant on May 31, even though the report -- which Mr. Therrien signed -- did cite other options-related actions he took on May 31. Not until August was the May 31 grant reported to the SEC.
It wasn't the only well-timed option grant he got. One in October 2001 came at Brooks stock's lowest closing price that year, once again at the nadir of a sharp plunge. The Journal analysis puts the odds of such a consistent pattern occurring by chance at about 1 in nine million.
Mr. Therrien, who stepped down as CEO in 2004 and retired as chairman this month, didn't return messages seeking comment. Chief Financial Officer Robert Woodbury said Brooks is "in the process of revamping" practices so grants come at about the same time each year. Mr. Woodbury, who joined in 2003, said no one at Brooks would be able to explain the timing of Mr. Therrien's grants.
The highly favorable 2000 grant also benefited two others at Brooks -- the compensation-committee members who oversaw the CEO's grants. Although Brooks directors typically got options only in July, that year a special grant was awarded just to these two directors, Roger Emerick and Amin J. Khoury. Each got 20,000 options at the low $39.75 price. By the time of their regular July option-grant date, the stock was way up to $61.75, a price far less favorable to options recipients.
Mr. Emerick, a retired CEO of Lam Research Corp., declined to be interviewed. Mr. Khoury, the CEO of BE Aerospace Inc. in Wellington, Fla., didn't return messages left at his office.
Another company, Comverse Technology Inc., said Tuesday that its board had started a review of its past stock-option practices, including "the accuracy of the stated dates of options grants," following questions about the dates from the Journal. The announcement reversed a prior Comverse statement -- given a week earlier in response to Journal inquiries -- saying all grants were made in accordance with applicable laws and regulations.
The Journal's analysis spotlighted an unusual pattern of grants to Kobi Alexander, chief executive of the New York maker of telecom systems and software. One grant was dated July 15, 1996, and carried an exercise price of $7.9167, adjusted for stock splits. It was priced at the bottom of a sharp one-day drop in the stock, which fell 13% the day of the grant and then rebounded 13% the next day.
Another grant, on Oct. 22, 2001, caught the second-lowest closing price of 2001. Others also corresponded to price dips. The odds of such a pattern occurring by chance are around 1 in six billion, according to the Journal's analysis.
Before Comverse announced its internal probe, John Friedman, a member of the board's compensation committee, said directors had noticed the scattered nature of the grants -- eight between 1994 and 2001 fell in six different months -- but management assured them there were valid reasons. Mr. Alexander, the CEO, didn't return phone calls.
This week, Comverse said that, as a result of the board's review of its options grants, it expects it will need to restate past financial results.
Propitious option timing can help build fortunes even at companies where the stock doesn't steadily rise. Shares of Vitesse Semiconductor Corp., although they zoomed in the late 1990s, now rest at about the level of a decade ago. But Louis R. Tomasetta, chief executive of the Camarillo, Calif., chip maker, reaped tens of millions of dollars from stock options.
Mr. Tomasetta got a grant in March 1997 that, adjusted for later stock splits, gave him the right to buy 600,000 shares at $5.625 each. The date they were priced coincided with a steep fall in Vitesse's stock, to what turned out to be its low for the year. He pocketed $23.1 million in profit when he exercised most of these options between 1998 and 2001. Had the grant come 10 days earlier, when the stock price was much stronger, he would have made $1.4 million less.
In eight of Mr. Tomasetta's nine option grants from 1994 to 2001, the grants were dated just before double-digit price surges in the next 20 trading days. The odds of such a pattern occurring by chance are about one in 26 billion.
Alex Daly, a member of the Vitesse board's compensation committee, said a review of the grants found "nothing extraordinary" about their timing, and "absolutely no grants have been made to anyone, least of all the CEO, that are out of sequence with our normal grant policy." Vitesse's finance chief, Yatin Mody, said the grants were "reviewed and approved" by the compensation committee, "and the exercise price set as of the date of the approval, as documented by the related minutes." He declined to provide a copy of those minutes. Mr. Tomasetta said the grants were "approved by the board and the price set at the close of the day of approval."
At ACS in Dallas, Mr. Rich helped turn a small technology firm into one with more than $4.4 billion in annual revenue and about 55,000 employees. ACS handles paperwork, accounting and data for businesses and government agencies. It is a major outsourcer, relying on global labor. "It is a pretty boring business," Mr. Rich told the University of Michigan business school in 2004, "but there is a lot of money in boring."
While most of Mr. Rich's stock-option gains were due to rises in ACS stock, the exceptional timing of grants enhanced his take. If his grants from 1995 through 2002 had come at each year's average share price, rather than the favorable dates, he'd have made about 15% less.
An especially well-timed grant, in which Mr. Rich received 500,000 options at $11.53, adjusted for stock splits, was dated Oct. 8, 1998. This happened to be the bottom of a steep plunge in the price. The shares fell 28% in the 20 trading days prior to Oct. 8, and rose 60% in the succeeding 20 trading days.
ACS's Ms. Pool said the grant was for Mr. Rich's promotion to CEO. He wasn't promoted until February 1999. Ms. Pool said there was a "six-month transition plan," and the Oct. 8 option grant was "in anticipation" of his promotion.
Mr. Rich would have fared far worse had his grant come on the day ACS announced his promotion. The stock by then was more than twice as high. The grant wasn't reported to the SEC until 10 months after the stated grant date. Ms. Pool said that was proper under regulations in place at the time.
A special board committee oversaw Mr. Rich's grants. Most years, its sole members were directors Frank Rossi and Joseph O'Neill. Mr. Rossi declined to comment. Mr. O'Neill said, "We had ups and downs in our stock price like any publicly traded stock. If there were perceived low points, would we grant options at that point? Yes."
Mr. Rich said grants were made on the day the compensation committee authorized them, or within a day or so of that. He said he or Chairman Darwin Deason made recommendations to the special board committee about option dates.
Mr. Rich, who is 45 years old, resigned abruptly as ACS's chief executive on a Thursday in September to "pursue other business interests." Again, his timing was advantageous. In an unusual separation agreement, the company agreed to make a special payment of $18.4 million, which was equal to the difference between the exercise price of 610,000 of his outstanding stock options and the closing ACS stock price on the day of his resignation.
But the company didn't announce the resignation that day. On the news the next Monday that its CEO was departing suddenly, the stock fell 6%. Mr. Rich netted an extra $2 million by cashing in the options before the announcement, rather than on the day of it.
Mr. Rich said ACS signed his separation agreement on Friday, using Thursday's price for the options payout. He said it waited till Monday to release the news because it didn't want to seem "evasive" by putting the news out late Friday.
George Anders contributed to this article.
How the Journal Analyzed Stock-Option Grants
The Wall Street Journal asked Erik Lie, an associate professor of finance at the University of Iowa who has studied backdating, to generate a list of companies that made stock-option grants that were followed by large gains in the stock price.
The Journal examined a number of the companies, looking at all of their option grants to their top executive from roughly 1995 through mid-2002. Securities-law changes in 2002 curtailed the potential for backdating a grant. Executives typically receive option grants annually.
Mr. Lie and other academics say a pattern of sharp stock appreciation after grant dates is an indication of backdating; by chance alone, grants ought to be followed by a mixed bag of stock performance -- some rises, some declines.
To quantify how unusual a particular pattern of grants is, the Journal calculated how much each company's stock rose in the 20 trading days following each grant date. The analysis then ranked that appreciation against the stock performance in the 20 days following all other trading days of the year. It ranked all 252 or so trading days in a given year according to how much the stock rose or fell following them.
For instance, Affiliated Computer Services Inc. reported an option grant to its then-president, Jeffrey Rich, dated Oct. 8, 1998. In the succeeding 20 trading days -- equal to roughly a month -- ACS stock rose 60.2%. That huge gain was the best 20-trading-day performance all year for ACS. So the Journal ranked Oct. 8 No. 1 for ACS for 1998.
It is very unlikely that several grants spread over a number of years would all fall on high-ranked days.
But all six of Mr. Rich's did. Another of his option grants also fell on the No. 1-ranked day of a year, March 9, 1995. Two grants fell on the second-ranked day, those in 1996 and 1997. In 20 02, his options grant was on the third-ranked day of the year, and in 2000, his grant came on the fourth-ranked day.
If a year has 252 trading days, the probability of a single options grant coming on the top-ranked day of that year would be one in 252. The chance of it coming on a day ranked No. 8 or better would be eight in 252.
The analysis then used the probability of each grant to figure how likely it is that an executive's overall multiyear grant pattern, or one more extreme than the actual pattern, occurred merely by chance. The more high-ranked days in the pattern, the longer the odds and the more likely it is that some factor other than chance influenced those dates. Two companies said they did use something other than chance -- they made grants on days when they thought the stock was temporarily low. This could explain results that differ somewhat from chance, but it wouldn't account for the extreme patterns of consistent post-grant rises.
John Emerson, an assistant professor of statistics at Yale, reviewed the methodology and developed a computer program to calculate the probabilities for all of the executives' grants except those to UnitedHealth CEO William McGuire. Because the number of his grants and complexity of his pattern made a computational method infeasible, the Journal used an estimate for his probability that Mr. Emerson said is conservative. Mr. Emerson said the figures for all six executives surpass a standard threshold statisticians use to assess the significance of a result.
For Mr. Rich's grants, the Journal's methodology puts the overall odds of a chance occurre nce at about one in 300 billion -- less likely than flipping a coin 38 times and having it come up "heads" every time.
Exceedingly long odds also turned up in the Journal's analysis of grant-date patterns at several other companies. "It's very, very, very unlikely that they could have produced such patterns just by choosing random dates," said Mr. Lie.
David Yermack, an associate professor of finance at New York University, reviewed the Journal's methodology and said it was a reasonable way to identify suspicious patterns of grants. But Mr. Yermack also said the odds shouldn't be thought of as precise figures, largely because they depend on assumptions in the method used to determine which grant dates are more favorable than others.
Because nobody actually authorizing the grant on a given day could have known how the stock would do in the future, the Journal's analysis used post-grant price surges as an indication of possible backdating. Academics theorize that the most effective way to consistently capture low-price days for option grants is to wait until after a stock has risen, then backdate a grant to a day prior to that rise.
The decision to look at 20 trading days after each grant was arbitrary. But Messrs. Yermack and Lie said it was a reasonable yardstick to detect possible backdating. Using a longer period, such as a year, wouldn't be a good way to spot backdating of a few days or weeks because the longer-term trading would overwhelm any backdating effect.
The 20-day price rises don't present an immediate opportunity to profit, since options can't usually be exercised until held a year or more. But when the options do become exercisable, they'll be more valuable if they were priced when the stock was low.