by Paul B.W. Miller and Paul R. Bahnson
In our preceding column, we once again went after those who persist in their pathetic arguments that stock options expense should not be recognized.
The sound bites warning of dire economic consequences have been resurrected by some managers to counteract the Financial Accounting Standards Board’s current efforts to mandate option expensing. Now that the board has gone public with a proposed standard, it will be interesting to see whether these same managers are successful in bribing, oops, persuading members of Congress to decide that expensing is a bad thing. The supposed calamitous consequences served as the rationalization behind the congressional flanking maneuver of a decade ago that stopped FASB from compelling expense recognition.
As we noted in our previous column, these arguments are nothing more than a transparent scheme to use incomplete financial statements to try to fool some investors into paying too much for stock. Of course, as a public policy tool, attempts to systematically overstate generally accepted accounting principles earnings would inevitably put sand in the otherwise well-oiled and smooth-running U.S. capital markets. We believe that no one could possibly be proud of that economic policy, yet many keep advocating it openly and without hesitation, when they should be ashamed.
In putting together that column, we didn’t go much beyond the typical rebuttal that’s been repeated over and over. However, we are now going to explain how this familiar response is inadequate, because it misses a very crucial point. Specifically, it focuses so intently on whether management should recognize a debit to expense that it completely overlooks the credit half of the entry.
As we see it, this conventional framing of the issue is anachronistic, dating back to the pre-Conceptual Framework era and the now-discarded “matching theory.” Specifically, matching focuses on the income statement item (in this case, the expense) without considering the balance sheet implications. In contrast, the CFW adopted the “asset-liability” theory that requires balance sheet account balances to be nailed down with real (relevant and reliable) numbers before going on to recognize the observed changes in assets and liabilities as revenues and expenses.
FASB has applied this theory many times since describing it in the CFW in 1980 and earlier. Because of intense political pressure, the board has apparently retreated back into matching and thus produced a remarkably shallow and incomplete method of accounting.
When we look at options through the asset-liability lens, we see that (a) options must be expensed, but (b) the FASB and International Accounting Standards Board methods of expensing are dishearteningly inadequate. We’re writing this column to expose that thinking to the sunlight, inviting our readers who already advocate expense recognition to reconsider their views on how best to accomplish that result.
To see the shortcomings of matching, let’s look first at APBO 25. As long as the strike price equals or exceeds the stock’s market value at the grant date, the issuer never recognizes any expense for fixed options, no matter what subsequently happens to the options’ market value and the underlying shares. Further, the issuer puts nothing on the balance sheet that reveals the value of the options overhang, leaving that relevant fact to be described only in the footnotes as the number of outstanding options, irrespective of their likelihood of exercise or their potentially dilutive effect.
When exercise occurs, the issuer records the new stock at the strike price, without regard to its current market value, thus grossly understating the event’s impact. In addition, a little-known requirement in APBO 25 forbids income statement recognition of the tax savings (for exercised non-qualified options) provided by deducting the discount between the strike price and the stock’s value at the exercise date on the tax return. As a result, management reports income tax expense that is usually well in excess of the taxes actually paid. All these defects are the direct consequence of determining up front that no expense should be recognized for stock options.
Because FASB has looked at this issue so carefully over two decades, one is tempted to say, “Aha! Surely they have figured out a much better way to describe the full effects of options.” Alas, this confidence would be poorly placed. Indeed, the intense political scrutiny and heat have forced the board to keep approaching the options issue using the obsolete matching theory instead of the asset-liability theory. What does this mean?
Under the method recommended in SFAS 123 (which is fundamentally the same as the one applied in the IASB standard and FASB’s current exposure draft), management must estimate the options’ market value as of the grant date. This amount is then treated like an advance payment, except that nothing gets recorded at issuance. Instead, an old-fashioned off-balance-sheet deferred charge (or dangling debit) is slowly and systematically “matched” against future years’ revenues. The annual expense is computed by amortizing this amount over the vesting period, based completely on initial measurements and expectations, not on subsequent reality.
When the options expense account is debited, the credit is thrown into a balance sheet equity account for outstanding options, although there is
no connection between the recorded amount and any real attribute of the options. By the end of the vesting period, the options account balance equals the options’ value as of the years-earlier grant date and has no semblance of their current value.
This result happens even if the options disappear under water or soar sky high. Thus, the method does not produce complete or comparable information. We believe that the board came to this compromise erroneously, either because the members actually clung to the matching theory or because FASB was unwilling to fight the political battle that would have surely followed an effort to apply another model.
In any case, we are certain that FASB’s expensing method is only a slight improvement over APBO 25. While management does recognize an expense for options, the reported amount deliberately ignores changes in the options’ value after the grant date. When options are exercised, the FASB method does not report the full cost incurred by selling the shares at a discount below their market value.
Alternatively, when options lapse, management does not recognize the gain that occurred when their value declined subsequent to their grant. Although the method does provide a meager balance sheet recognition of the overhang, it is there only to the extent that the options’ initial value has been amortized, thus providing absolutely no acknowledgement of the near-certain appreciation or depreciation of the options’ real value as of the balance sheet date.
Further, any stock issued upon exercise is recorded at the sum of the options’ old grant date value plus the strike price, an amount that is certain to be less, even much less, than the shares’ market value on the exercise date.
Finally, the SFAS 123 method continues to grossly misrepresent income tax savings realized when non-qualified options are exercised. If options appreciate above their initial value, additional tax savings occur; however, FASB’s method, like the one in APBO 25, does not reduce tax expense, but increases other paid-in capital. Thus, the issuer’s reported tax expense is permanently greater than the taxes actually paid.
So, there you have the huge irony inherent in the debate over stock options. All that incredible energy has been spent attacking and defending FASB’s effort just to get some expense on the income statement, even though the outcome is inconsistent with the Conceptual Framework, even though the balance sheet is void of useful information, and even though the preferred method doesn’t begin to tell the whole story of what happens when compensatory options are first issued and then remain outstanding.
While we prefer some progress over none (and thus favor mandatory expense recognition), we are thoroughly disappointed that the progress that would be created by the exposure draft is so minimal, when a complete method is right there in front of everyone, including FASB and the IASB.
We’ll tackle that method in our next column. As a hint, it involves acknowledging that options are derivative liabilities that expose the issuer and its stockholders to significant risks. Any accounting method must fully reflect those risks if it’s going to be fully useful.
Paul B.W. Miller is a professor at the University of Colorado at Colorado Springs, and Paul R. Bahnson is a professor at Boise State University. The authors’ views are not necessarily those of their institutions. Reach them at
