Monstrous Pay.

A lot of people wondered why Bob Rubin, who had already gone through brilliant careers as head of Goldman Sachs Group Inc. and Secretary of the Treasury, would want to become the consigliere of Sandy Weill at Citigroup.

The answer came last week, when Citigroup disclosed the pay of its top executives.

On an annualized basis, I estimate Rubin will be pulling down $47 million a year for his first two years as chairman of Citigroup's executive committee. That's more than all but a handful of CEOs earn, with Weill himself taking up much of the space in that proverbial hand.

Rubin came into Citigroup with a nominal fixed salary of $1 million. But he also was given a guaranteed bonus of at least $14 million for 2000 and another $14 million for 2001. A guaranteed bonus has the look and feel of base salary, so we can safely see Rubin earning fixed compensation of $15 million a year.

That's 50 times more than Henry Paulson's fixed compensation at Rubin's old firm of Goldman Sachs. No wonder Rubin didn't want to go home again. That would be roughly akin to Sandy Weill moving back into the apartment in Brooklyn where he and his brilliant career were born.

Pay-for-existence

To earn fixed compensation of $15 million a year makes a mockery of the concept of pay-for-performance. If disaster strikes at Goldman, Paulson's pay presumably can take on the aerodynamic characteristics of a Sub Zero refrigerator dropped from an airplane at 35,000 feet. But Bob Rubin? He'll be down to his last $15 million a year.

Rubin was also granted options on 1.5 million shares of Citigroup stock. And in 2000, he will receive another option covering another 1.5 million shares. I estimate his 1999 option grant had a present value of $30.9 million when it was granted.

(Citigroup estimated a much lower figure, $18.5 million, but we'll come back to the company's methods of valuing options later.)

To illustrate the size of this grant, if Citigroup stock rose just 10 percent from its price of $47.81 on Oct. 25, 1999, the day before the option was granted, the paper profit would be $7.2 million.

Looking at all the future scenarios produced by the Black-Scholes option pricing model, his average possible gain on the option after seven years would be $99 million. That would mean a 13.2 percent annual rise in the stock over that period. And that, friends, is from just one of his two 1.5 million-share grants, and for merely average future performance.

In sum, it appears that Rubin has learned a great deal from his two most recent jobs. He has managed to engineer for himself a pay package that combines the very best of civil service pay protection and investment banking upside.

Pales next to Weill

Then there is the Numero Uno himself, Sandy Weill. When writing about Weill's pay each year, I feel like a sportscaster who has to consult a thesaurus to keep finding new synonyms for winning. In earlier writings, I have used huge, gargantuan, monstrous and Godzilla-like to describe Weill's pay. What's left?

By my reckoning, Weill's total pay package for 1999 was worth almost $115 million. But hey, that's lower than the $136 million I figured he earned in 1998. Even by Citigroup's tortuous methodology, which I describe below, his package weighs in at $59.5 million. That includes a salary of $1 million, a cash bonus of $8.7 million and restricted shares with a market value of $4.4 million when they were granted.

If he weren't such a brilliant CEO, Weill would have had a great career as a political spinmeister. Because when it comes to telling his shareholders how much he is earning, he spins better than Rumpelstilskin.

Under the Weill mythology, the 6.9 million option shares he was granted in 1999 were not really new option grants at all. Citigroup and some other companies allow an executive to turn in shares to satisfy the cost of exercising the option and paying the tax bill. In return, the company gives them a new option grant for the same number of shares they turned in, with the new grant carrying a strike price equal to that day's market price and with a term equal to the remaining term of the original option.

In 1999, Weill exercised 7.7 million shares for a gain of $76 million. In return, he was granted new options on 6.9 million shares. To Weill and to other CEOs who get these "reload" option grants, these are not new options but merely extensions of existing options.

To paraphrase Ecclesiastes, an option is born when it is granted, and it dies when it is exercised or expires underwater. But a "reload" option takes a page from another part of the Bible. It ought to be called a "Lazarns" option, because it keeps rising and rising from the dead. To my knowledge, there is no real-world option out there that does that. So, call it what you will, but I see a "reload" option as simply a new option grant.

Citigroup's methodology

In valuing these reload options, Citigroup assumes they will be exercised in approximately one year. The shorter the life of an option, the lower the value. By plugging a one-year assumption into the Black-Scholes model for valuing options, the company is able to report to shareholders a much lower figure for the present value than if it used a longer expected term. That is the primary reason that year after year, my estimates of Weill's pay are far higher than what the company reports.

Look at the 397,195 shares Weill reloaded in 1999 from the "Founders Grant" made to him in 1998. According to my valuation methodology, those shares have an estimated present value of $8.8 million, based on an assumed exercise point that is 70 percent along the way to their nominal expiration date in nine years. But Citigroup says those options are worth only $2.4 million.

Citigroup gets to its reload value by assuming the option will be exercised well before its nine-year nominal term. The Securities and Exchange Commission permits companies to use a shorter average term based on past experience in place of the nominal term.

At Citigroup, the "reload" option maybe exercised after six months if the stock price rises 20 percent or more. So, given Weill's fine performance, there is a certain sense in arguing he will not be using anywhere near the entire term. But suppose he loses his magic touch, or the entire market falls? Then the true length of time between date of grant and date of exercise could be a lot closer to nine years than one.

To me, that lowball figure of $2.4 million is a joke. These options were reloaded on Nov. 2, 1999. On that day it would have cost an investor not $2.4 million but rather $5.5 million to buy a Citigroup publicly traded option with a strike price that was higher than Weill's strike price ($55 vs. Weill's $53.38) and that had a term remaining, not of Weill's nine years, but rather of a mere 2.2 years. Go figure.

Do shareholders complain about Weill's pay? Not a word. They love him, because even after the last forklift leaves the Citigroup Treasury with his latest haul, there is plenty left for them.

So I give up. That's not to say that I won't be writing about Weill every year until he retires. But my heart's no longer in it. I'm actually beginning to think of him as an old friend.

Graef Crystal is a columnist on executive compensation for Bloomberg News.