Avoid Stock Comp Expenses: The Paul Atkins Diet?


private investment trust Commissioner Paul Atkins gave a speech on Wednesday, June 8 before the Semiconductor Industry Association Leadership Luncheon in San Francisco. In it, he gave a couple of sobering insights into the current thought processes at the SEC – even if the views that he expressed were his own. And it sounds like Dr. Atkins recommends a corporate diet of low-carb stock options picked from a scrawny, patched-together market, instead of ones with estimated values raised in a calculator.

He offered praise for the FASB’s independence, but there’s heavy irony in his note that, since Sarbanes-Oxley, the relationship between the SEC and the FASB has changed, in particular that “the SEC now has more responsibility over the FASB, especially how the FASB funds itself. I have focused on increasing the transparency of FASB’s budget and processes and its overall accountability.”

Maybe that’s not a change for the better when it comes to setting standards for stock option accounting, it would seem.

Atkins went on to point out that “I have not met many people, either inside or outside the Commission, who are truly confident the FASB Standard 123-R models, Lattice or Black-Scholes, provide good estimates of employee stock option value, especially for options distributed in a broad-based plan. That has been and continues to be a concern for me, particularly if these estimations are material and can be subject to management of the assumptions and outcome.”

Well, given that he was speaking at an SIA function, it’s not hard to meet many people “outside the Commission” who are truly confident that the FASB standard-endorsed models will provide good estimates of option value. But he hasn’t met many “inside the Commission” who believe it’s workable?

Maybe he hasn’t read the analysis done by the SEC’s own Office of Economic Analysis, reviewed in this April posting. Perhaps he missed the bullet point on page 3 that said “There is evidence that the modified Black-Scholes-Merton approach provides reliable estimates of option value, even while the lattice, binomial and Monte Carlo approaches have important structural advantages that make them better suited for use by many companies.”

It didn’t sound like there was disagreement from a reputable group within the Commission in that statement.

There was a flurry of news a few weeks ago about the Cisco proposal to develop a “market” for employee options that would “solve” the valuation problems surrounding these compensation instruments. In recent weeks, the story has gone cold: there’s been no news from Cisco, and there’s been no comment from the SEC. Mr. Atkins let a few things out of the bag:

“I am encouraged that market participants and the SEC are working together to develop a market-based options pricing model. Without getting into too much detail, the idea is to try to establish a security that mimics the characteristics of the option and tries to find a willing buyer and a willing seller for this security. There is no better way to determine what something is worth than to get people to negotiate to buy it. I am following this closely and encouraged to see that we are using the power of the marketplace to guide us to make reasonable, informed regulatory choices.”

Market participants and the SEC are working together to develop a market-based options pricing model? That sounds out of character for Paul Atkins: he’s usually a fan of keeping government out of involvement with the private sector. Should the SEC be taking pains to develop a market-based options pricing model? Sounds like an intrusion into the private sector to me. Markets always develop when participants want them, I thought. Not when the government tries to make them happen.

And another thing: the SEC is fond of stating “We are the investor’s advocate.” If “market participants and the SEC are working together” on this – where’s the sunshine? Where’s the due process and the proposals and the comments? Investors would hate to miss this chance to see what’s going on, I think.

When the Cisco plan was hot news, there was one very interesting press report from Dow Jones Newswires, an interview with Myron Scholes, in which the co-inventor of the famed Black-Scholes option pricing model took some issue with the Cisco proposal:

“But the Cisco plan has potential pitfalls, suggested Myron Scholes, the economist who shared a Nobel prize in 1997 for his contributions to the Black-Scholes options pricing model. In order to bring “a very idiosyncratic contract to the market,” he said, it would have to be sold at a discount “in order to encourage the market to take it.”

While Cisco is billing the alternative market price as an attempt to find an “accurate” valuation for stock-option expensing, Schools said the market auction proposal is one that by its nature will not draw a true price. For one thing, a small market placement and little liquidity means the issue would likely sell at a discount.

“I think that the cost of the Cisco employee stock options would actually be greater than the price in the market place of the contracts that Cisco plans to issue,” he said. “

Mr. Atkins seemed to take a shot at the article in his speech:

“I would like to make one more point about this market-based valuation model. Some recent press reports have characterized this idea as a scaling back or a “discount” to Black-Scholes or Lattice. These characterizations are flawed. The FASB standard specifically allows market-based models to determine value of options. If a functioning market determines a value, then we need not compare that value to a number that is derived from any other existing model. Market value is real; a model is just an estimate. If you have any thoughts about this approach, I would welcome the opportunity to discuss it with you.”

It’s hard to theorize about the Cisco thought process when there’s so little sunlight on it. But based on Mr. Atkins comments, it seems to rest on the idea that any market price is superior to an estimate, and that’s what the FASB literature requires. It’s true that the FASB literature endorses fair values as a measurement yardstick, even in thin markets. It sounds as if these markets would indeed be thin: they might be non-visible markets made up of private institutional investors, and trade only once, at the issuance of the employee options. But what still is troubling is that this contrived market is for derivatives on the options to the employees, not the employee options themselves. It sounds like there would still need to be some reverse engineering done to get at the value of the security (the employee options) underlying the derivative (what’s being traded). And how might that employee option be valued? As Mr. Scholes indicates, “somewhere along the line the options would still need to be valued through an established options-pricing model.”

Interesting. A commissioner of the SEC “welcomes the opportunity to discuss [the model]” with members of the SIA. No mention of welcoming comments from investors. “We are the investor’s advocate?”