Expensing Options Solves Nothing

The use of stock options for executive compensation has become a lightning rod for public anger in the wake of the series of corporate disasters and frauds reported in the press over the past year. One can easily see why: Top executives at many of yesterday’s star companies grew hugely rich on the gains they made on their options, profits they’ve been able to keep even as the value these people were supposed to have created disappeared.

What makes the windfalls particularly galling is that it looks as if the system was set up to guarantee executives the money. The supposed scam works like this: Current accounting regulations allow companies to ignore the cost of option grants on their income statements, which means that they can award valuable option packages without affecting reported earnings. Not charging the cost of the grants supposedly leads to overstated earnings. Overstated earnings purportedly translate into unrealistically high share prices, permitting top executives to realize the value of their options purely on the basis of an accounting anomaly.

Presented in this light, the treatment of executive share options in financial reports certainly looks like a scheme to transfer money from shareholders to greedy, undeserving executives, secure in the cover of a booming stock market. And it also seems egregiously simple to fix. If the problem is caused by distortions created when companies don’t charge options against income, all you’d have to do is treat executive share options the same way you would any other component of an executive’s pay package—as an expense to be written off against the current year’s revenues. It’s not as if there’s any mystery to calculating the value of an option. A widely accepted methodology exists for doing just that: the famous Black-Scholes model. It lets you calculate the value of an option from a few easily estimated inputs: a company’s current stock price, the option’s exercise price, the time period during which the option can be exercised, the dividend yield, the volatility of the underlying stock, and the level of interest rates.

The great and the good have lined up behind proposals to “expense” options. Corporate behemoths like GE and Coca-Cola have already declared that they will start reporting vested option grants at their estimated Black-Scholes value as an expense each year, and even entrepreneurial upstarts like Amazon have followed suit. The market’s guiding lights—Warren Buffett and Alan Green-span among them—advocate this approach. Academic heavyweights concur, including two of my colleagues at Harvard Business School, Robert Kaplan of Balanced Scorecard fame and Nobel prize winner Robert Merton. Politicians have been quick to jump on the bandwagon; last February, Senator John McCain, the champion of clean government, sponsored a measure to enforce the expensing of executives’ options.

Now I’m not about to take issue with those people. When it comes down to it, I don’t really care if companies expense their option grants. On balance, I think they shouldn’t, particularly if it entails disclosing less information in their footnotes. And I’ll show you why in a moment. But what’s really driven me to put pen to paper is an altogether different concern: The current focus on the expensing issue is deferring a much broader discussion that politicians, regulators, investors, corporate executives, and accountants need to have.

To begin with, reporting an executive option grant as a cost item on the income statement does not add any information that’s not already included in the financial statements. If anything, expensing options may lead to an even more distorted picture of a company’s economic condition and cash flows than financial statements currently paint. More seriously, focusing on the theoretical value of stock options is likely to distract people from thinking more broadly about whether compensation (options or otherwise) does what it’s supposed to do—namely, recruit the right people, retain them, and provide appropriate performance incentives.

But what worries me most of all is the likelihood that once laws and regulations are put in place to enforce options expensing, the public will conclude that corporate America’s bookkeeping problems have been solved. If that happens, we will have missed a golden opportunity to engage in a serious discussion about the really big problems afflicting corporate accounting.

It’s Probably in the Footnotes

Financial Accounting Standards Board regulation 123, introduced in October 1995, already requires companies to disclose data about option grants in the footnotes to their financial statements. A quick look at any publicly traded company’s reports shows just what a wealth of information those footnotes provide. Microsoft, for example, reveals in a footnote on pages 36 to 38 of its 10(k) form for 2001 that, by June of that year, 898 million options were outstanding. Of these, 331 million could be exercised; the rest were not yet fully vested. It had authorized but not granted options on a further 550 million shares. During the six years preceding June 30, 2001, it had granted new options on almost 1 billion shares.

The footnote goes on to describe the option terms in some detail. We learn that the weighted average of the exercise prices assigned to the options granted in 2001 was $60.84, that they expired in ten years, and that the average vesting period was 4.5 years. Microsoft estimates the total value of the options granted in 2001 to have been $3.4 billion ($2.3 billion after taxes).

Hence, if Microsoft had expensed the options in 2001, reported net income (before a one-time accounting charge) would have dropped from $7.7 billion to $5.5 billion, or almost 30%. Had Microsoft expensed options for the six years ending in June 2001, cumulative net income would have fallen from $35.1 billion to $29.6 billion, or slightly less than 16%.

Pundits like Alan Greenspan and Warren Buffett would argue that Microsoft overstated its profitability by failing to include the options charge in its income statement. Investors must therefore have been misled into believing Microsoft was doing better than it really was. But is that really true? Did Microsoft actually earn less? Did investors receive the wrong signal?

To answer these questions, you must take into account a wider range of issues than just the impact of deducting the estimated value of the options from current income. Consider, for example, how the use of stock options affects cash flows. Companies that compensate executives with options, particularly at the senior level, usually don’t have to pay as much to recruit them. If management fails to improve performance, then the company’s share price will not rise, and the options will not be exercised. In that case, shareholders will be left with the same claim on future cash flows that they had at the outset, but they will have saved the cash they might otherwise have had to spend on recruitment and retention.

In the event that the management team performs well and the options are exercised, the company can expect an inflow of exercise money along with a tax saving. Here’s how that works:

Microsoft’s 2001 10(k) provides a wealth of details on the actual cash flows arising from the exercise of previously issued options. In that year, employees exercised options on 123 million shares at an average cost to them of $11.13, representing a total payment to the company of about $1.4 billion. Since companies can deduct from taxes approximately 35% of the difference between the money that grantees actually paid for the shares at the time they exercised their options and the market value of the shares at that time, we can work out that the total cash inflows from the exercise of those options was $3.5 billion.

For the six years that ended in June 2001, the positive cash flows from option exercises at Microsoft and their associated tax savings totaled $20.8 billion.¹ What is most interesting, and curious, is that the proposed expensing of Microsoft’s options depresses income while increasing cash flows. One must question a policy that keeps the two from rising and falling in tandem.

What Options Can Do for You

What investors care about most is their claims on the after-tax cash flows of their company. A grant of stock options potentially transfers to other people, the grantees, some of the claims of existing shareholders on cash flows—that’s precisely why options have value. For shareholders to be compensated for that potential loss, the option grant must somehow increase cash flows, on the principle that a smaller piece of a larger pie can be better than a larger piece of a smaller pie. Regardless of the accounting treatment of the cost of options, accountants will have adjusted their estimation of the number of shares outstanding to reflect the likely impact of option grants and exercises.

So where does the increase in cash flows come from? We’ve already seen how the increased cash flows that option grants generate reduce their cost to shareholders. Those cash flow gains, though, fade into insignificance when compared with the potential benefits that good managers can bring to their company by exercising skill and good judgment. During the six years in which Micro-soft handed out the options on approximately 1 billion shares, the total market value of Microsoft increased by slightly over $300 billion, representing a jump in share price of $62.00 and a percentage rise of 413%. (In the same period the S&P 500 rose 83%.) Estimated free cash flows per share also increased, from $0.57 to $1.27. The reported value of the option grants made (using Black-Scholes) over the six years was $8.2 billion. The estimated value of all stock options awarded in each of those six years was under 1% of the market value of the company at the beginning of each respective year. (The six-year average was 0.6%.) If you were a shareholder during this time, you would have had precious little basis on which to complain about the compensation of your company’s management.²

Indeed, one of the chief arguments advanced for using options at all is precisely that they are an effective way of attracting and retaining the right kind of managerial talent—the kind that will increase earnings and drive up stock prices. Anyone joining Microsoft during most of its history profited handsomely by staying and missed out on significant appreciation by leaving. It is explicitly because of such outcomes that the management teams venture capitalists assemble for start-ups have been happy to accept lots of options in lieu of cash. People unwilling to accept such a package are shown the door or are not hired in the first place. Those who stay are powerfully motivated to increase their company’s value so that they can turn their options into cash. Looked at in this way, options have the earmarks of an investment in the future rather than an operating expense. This is of course one reason that venture capitalists are not in favor of the accounting reforms.

Unfortunately, expensing options by subtracting their full Black-Scholes value from income in the year of the grant does not take into account any of the possible benefits of using stock options as part of a human resource management strategy. Accountants would probably say that because it is difficult to predict many of the future benefits, they prefer to err on the side of caution by ignoring the benefits and taking the maximum hit. For this reason, the accounting profession has compiled an extensive record of treating many expenditures that have the potential to improve expected performance over the long term as current-period expenses. Examples include spending on R&D and staff training. Although companies make such investments only when they expect them to boost performance, conservative dogma demands that the financial statements recognize only their cost. But, again, such treatment ends up distorting the picture. And that is particularly true of options. At least with line items like R&D, there is a real cash outlay, ensuring that the impact on income is the same as it is on cash flows. But no such outlay is made when companies distribute stock options.

The Real Accounting Problem

Whatever the rights and wrongs of footnote disclosure, and whatever the merits of options as a form of compensation, it’s certainly a scandal that accounting is so muddled in its treatment of performance-linked compensation, which currently encourages companies to prefer options to potentially more effective methods of equity-linked compensation.

For instance, one problem with most traditional options is that they reward the holders even if the issuer’s stock goes up less than its competitors’ or its cost of capital goes up more. It seems a little strange to reward people for riding a rising market’s coattails, and for that reason many compensation experts advocate using indexed options instead, which tie the exercise price to the level of some benchmark. Yet under current accounting rules, indexed options must be expensed, penalizing income, whereas conventional options, arguably a less effective compensation method, need not be.

Another anomaly is the treatment of restricted stock grants, which, like indexed options, also have to be expensed, even though they, too, theoretically deliver more bang for each dollar cost to the company than conventional options do. Unlike options, stock grants afford the employee some downside protection because equity does not expire worthless if the exercise price is not exceeded by the expiration date. Consequently, employees value a dollar’s worth of stock more than they do a theoretical dollar’s worth of options, which means employers don’t have to be as generous with stock grants as with options. Again, it seems perverse to have an accounting treatment that discourages companies from experimenting with a potentially more cost-effective compensation method.

Obviously, expensing options would remove these inconsistencies, but it would only create other ones. Consider, for instance, the way many companies treat long-term performance-related cash bonus programs. Under such programs, an executive can earn a cash payout two or three times his or her base salary if the company’s reported income increases by more than a given percentage during some time period. At present, companies are required to record an expense for such bonus plans only when it becomes apparent that a payout is likely.

This method of accounting is very different from that proposed for options, which would impose in the year of the grant the full, multiyear cost of the contingent liability. Yet bonus plans are very similar to options in three crucial respects. First, bonus plans give the recipient a contingent claim on extra payments triggered when certain financial targets are reached, just as an executive with options gets a claim on the company’s future cash flows if the shares rise to a certain price (above the exercise price). Second, even though the bonuses do not increase the number of shares outstanding, they do transfer wealth from shareholders to employees because employees have a prior claim on future cash flows. Finally, as with options, bonus plans do not penalize the manager in the event of poor performance (they can only be worthless). Even when the bonus plan seems unlikely to pay out, and no charge to income is deemed appropriate, the right to get the bonus still has value. There is always a chance income will rise fast enough to justify the bonus. The contract has value just as an option has value even when the current stock price is less than the exercise price, as long as there is still time left before expiration. Given those similarities, I have to wonder why no one is proposing a more “conservative” costing of bonuses.

The only reasonable argument for forcing companies to take a current charge for the estimated value of current stock option grants is to put all forms of incentive compensation on a similar footing. Then, if they were all accounted for in the same way, the impact on reported earnings would be the same whether at-the-money or performance-indexed stock options were granted or long-term cash payout plans were used. Alternatively, analysts would be equally well informed if all types of contingent compensation were adequately described in the footnotes, including a detailed listing of the underlying assumptions about things like volatility, exercise price or conditions, and duration. No self-respecting analyst would focus solely on reported income and ignore the balance sheets, the cash-flow statement, and the associated footnotes (see the sidebar “There’s Always Room for More Disclosure”). The fact that so many analysts failed to do so in the case of WorldCom, Enron, and Adelphia just shows that the problems go far beyond how any particular incentive is accounted for.

The big issue is not the accounting treatment of disclosed managerial decisions but whether those decisions make sense in the first place. Does the compensation program attract the right people? Does it provide incentives to increase long-term value? Does it send the right signals internally and externally? Are adequate controls in place to thwart malfeasance and damaging, self-interested behavior? These are the critical questions in today’s world.

What Matters More

It is fascinating to observe pundit after pundit come down strongly on the side of expensing stock options in the reported financial statements, as if that were the silver bullet for combating corporate malfeasance and resolving all our accounting problems. But the proposals under consideration can do no more than palliate public outrage. What we need is a much more comprehensive look at the recent scandals so that we can begin to figure out what the real issues are.

As a start, let’s consider one of corporate America’s biggest villains: Enron. Certainly the company was liberal with stock option grants, though not as liberal as many others. In fact, expensing options in Enron’s accounts would have changed reported profits by only about 10%, whereas the change would have been around 30% for Microsoft, which has received no criticism for its options programs. (A comparison of the two companies’ programs is shown in the exhibit “Microsoft Versus Enron.”)

Microsoft Versus Enron It is instructive to compare Microsoft’s situation with Enron’s, where it appears to some that stock options drove unethical or even illegal behavior. As this table shows, moving the reported value of options from the footnotes to the financial statements would have had more impact on Microsoft’s numbers than on Enron’s. Enron’s problems ran much deeper than options.

The real accounting scandal at Enron had nothing to do with the failure to expense options. Rather, it related to a failure to disclose something else entirely on both the income statement and the balance sheet. Enron had taken advantage of some very liberal (and economically nonsensical) accounting rules that allowed the company to transfer assets and liabilities to certain so-called special purpose entities (SPEs). According to the Powers report, which was published by a special committee of Enron’s board after the company entered bankruptcy protection proceedings, Enron’s management used the SPEs simultaneously to overstate income and understate debt. For example, Enron would sell certain assets to new SPEs, booking a gain on the sale. Then, in quite a few of the transactions, Enron would repurchase the very same assets within months at a slightly higher price. These were not legitimate sales; they were instead short-term, unrecorded loans to Enron. What’s more, several of Enron’s officers were partners in some of the SPEs. Because these officers had more to gain from their SPE ownership than from their ownership of Enron (an obvious conflict of interest), they might have been tempted to structure transactions that were favorable to the SPE but not to Enron.

Were some of these issues disclosed in Enron’s financial statements and related footnotes? Yes, they were, but even the special committee of the Enron board of directors later described the disclosures as “obtuse” and woefully inadequate. A careful and skilled analyst could never have figured out all of the possible problems at Enron from its reported financial statements. In this regard, current accounting for stock options actually serves as a model for disclosure, in some respects. Investors are given lots of information about stock option plans, including some that can help them assign a value to the options granted. In sharp contrast, investors in Enron could not judge the value—on the basis of the information they were given—of the contingent liabilities that Enron had incurred either for itself or for its complex SPEs.

But even much fuller disclosure would not have saved Enron or, for that matter, WorldCom or Adelphia. The failures at those companies were more likely caused by a combination of fraud committed by individuals, inadequate control and governance systems that tolerated clear conflicts of interest, and a frothy market in which analysts failed to do even the simplest reality checks on reported cash flows. Those analysts who took WorldCom’s reported income as proof that it was doing well would have come closer to the truth if they had simply calculated free cash flows. Then they would have seen the capital expenditures that the company was reporting falsely in order to conceal the true level of its operating expenses. All the required information was there: It just was ignored by the investment community.

Even if the proposed rules for stock option accounting end up discouraging the use of stock options, the potential for fraud, and grotesquely outsized gains, will not be reduced. Any compensation system that is based on performance has the potential to encourage cheating. Only ethical management, sensible governance, adequate internal control systems, and comprehensive disclosure will protect the investor against disaster.

The tensions in the American business model surrounding the way companies measure and track their performance are much less black-and-white than the popular press would have us believe. The furor over expensing is, if anything, a sideshow distracting us from deeper flaws in accounting standards, compensation philosophy, and professional standards in the financial services industry. If the advocates of expensing win their small point and the spotlight on corporate America fades away as a result, I fear that we will end up having done nothing at all to prevent unscrupulous executives from yet again stealing their investors’ money.