We think multiples are even higher than they appear....

Goldfarb: When I think about the current market, I am reminded of a comment made in 1997 by Warren Buffett.  Repeating an assessment made by his mentor Benjamin Graham more than four decades earlier, Mr. Buffett said, "Common stocks look high and are high, but they are not as high as they look."

        By contrast, we believe that today, common stocks look high and are high, but they are higher than they look - at least in terms of the popular benchmark S&P 500 Index, which I believe encompasses about 80% of the market value of publicly-traded companies in the U.S. The primary reason for our belief is the widespread overstatement of earnings created by two accounting practices:

Understated compensation expense = overstated earnings.

Goldfarb: First, the fact that the cost of extravagant stock option packages are not expensed through the Income statement.  Generally Accepted Accounting Principles do not require companies to recognize an expense on their income statements associated with what is clearly a value transfer coming out of the pockets of shareholders.  As Warren Buffett, in his inimitable capacity to concisely get to the heart of the matter, asked several years ago: "If options aren't compensation, then what are they?  If compensation isn't an expense, then what is it?"

        Note that many companies have options programs which, if expensed, would have only a modest impact on reported earnings.  However, while high levels of equity-type awards and profit skimming were once the exclusive province of either start-ups or companies run by corporate renegades, today it is not uncommon to see annual option awards equal to 3% of shares outstanding or worse at established and respectable companies.

And restructuring charges front load future expenses.

Goldfarb: Our second reason for believing that the market is higher than it seems is that restructuring charges in one fell swoop cast doubt on the historical earnings record, on the current year's so-called operating earnings and, to the extent that these charges "front load" certain future expenses, on prospective reported earnings as well.

        The 1998 Berkshire Hathaway annual report contains an excellent discussion of these two items and we would encourage you to read it.  However, we thought it might be useful today to elaborate on the options issue, as we have become increasingly concerned about this troubling phenomenon.

Don't hold your breath for CEOs to reform their accounting.

Goldfarb: While companies are permitted by current accounting guidelines to expense options directly through their income statement, to date none has volunteered to switch to this stricter reporting standard.  And who can blame them?  The widespread overstatement of earnings due to the non-recognition of option expense is equivalent to grade inflation in schools.  Is it at all surprising that executives, now heavily incented by stock price appreciation due to their large options stakes, have not stormed the gates to join the class of the professor who uses tougher grading standards?  Unfortunately, without strict mandates from the SEC or FASB, the lavish use of options and underreporting of the associated cost will almost certainly continue.

And Wall Street hangs its hat on reported earnings...

Goldfarb: Even more unfortunate, although perhaps as understandable, is the fact that there is little discussion of option expense on Wall Street among the people who are supposed to be valuing corporate America's earnings.  Prominent market strategists who make public pronouncements regarding market valuations are basing their assessment strictly on reported earnings, without much regard to the large but unrecorded options expense borne by virtually every company, not to mention so-called restructuring charges, special items, or even-more hidden adjustments to shareholders equity.
        And the analysts who cover individual stocks don't do any better.  We find it both remarkable and ironic that when a company's reported quarterly earnings fall even a penny short of analyst estimates, its stock price can drop by 50%, while another company that meets analyst estimates but engages in a massive unreported earnings dilution through lavish stock option grants can enjoy a flood of buy recommendations and a soaring stock price.

Of course, disregarding legitimate expenses is convenient.

Goldfarb: Perhaps the complexity inherent in estimating the true economic cost of stock options, combined with a lack of interest by their clients, is why we have yet to see any Wall Street analyst produce an estimate of options-adjusted earnings for a publicly- traded company.  We also suspect that In an era in which analysts' EPS estimates are compared to the so-called "consensus" estimate, analysts' devotion to uniform standards of reporting also becomes understandable, if not especially commendable.

In fact, seeing no evil has been REWARDING lately.

Goldfarb: Regrettably, the overstatement of earnings constitutes an invisible tax on investors. The payment of this tax will be delayed as long as the majority of the investment community continues to accept inflated earnings at face value.  It is interesting to note that until now, most of America has benefited from this delusion.  Perhaps this is the reason investors have passively accepted and voted for so many option programs that frankly rob them of much of their wealth.
 

Valuing stock options is much like valuing common stocks.

Goldberg: Admittedly, the exercise of trying to calculate the true expense of stock options for any individual company is complex.  The value of a stock option is equal to the present value of a company's expected future per share free cash flows, net of the present value of the exercise proceeds.  In this respect, valuing an option is similar conceptually to valuing a company's common stock, although the structure of an option does deliver some additional leverage both on the upside and the downside relative to the common.
 
        The necessary qualitative analysis that goes into valuing future cash flows cannot be captured in any one-size-fits-all formula, such as the Black-Scholes model that is the convention in financial statement footnotes regarding stock options.  In this respect, valuing option grants, like valuing common stocks, is more of an art than a science.

Lower stock price= a HIGHER option value, not a lower one.

Goldfarb: The basic problem with Black-Scholes and other mathematical option-pricing models is that they assume the current market values accurately reflect intrinsic value - in essence, they assume the stock market is efficient.

        For example, consider a company that grants stock options when its stock is selling at 5 times earnings.  Using Black-Scholes, a stock option grant issued when the underlying common stock is selling at 5 times earnings is valued at a price 10 times less than an equal-sized grant from the same company issued when the stock price is selling for 50 times earnings.  But we think it could conceivably be worth at least 10 times more.

And it's not like the concept of intrinsic value is unknown.

Goldfarb: In determining the level of aggregate annual stock option grants and the size of individual employee awards, the vast majority of compensation committees, by using the Black-Scholes model, accepts this flawed assumption.  These committees abdicate their responsibility, by relying on compensation consultants armed with seemingly scientific options valuation models and impressive peer group analyses.

        This pass-the-buck mentality is ironic, when you consider that senior management and Board members frequently make estimates of intrinsic value for decisions such as whether to issue stock or pay cash in an acquisition or whether to repurchase existing shares or to sell new ones, not to mention whether to buy or sell shares for their own personal accounts.

Thus, current valuations are frequently off by light years.

Goldfarb: While one could debate some of the finer points of valuing the cost of option grants, any reasonable estimate suggests that the cost is much too large to be ignored.  In fact, the conventional option valuation models have, over the last fifteen years, been off by hundreds of country miles in the aggregate, and by light years in many individual cases.
        The fatal flaw of these models is their inability to differentiate between real diamonds and cubic zirconium between undervalued and fully-priced merchandise.  In other words, these models do not [even attempt to] capture differences in capital appreciation potential.
 

If it was a problem before, it's a whole new ballgame today.

Goldfarb: Any method of valuing options has drawbacks.  However, an indisputable point is that the number of options issued matters greatly.  There was a time not too long ago when an annual option grant equal to 1% of shares outstanding raised eyebrows.  However, today, yearly grants equal to 2% or 3% of shares outstanding are not uncommon.

For S&P leaders, earnings may be off by a factor of three! 

Goldfarb: One method of option valuation that is both observable and quantifiable is to impute an expense based on a retrospective calculation of the actual value transferred - that is, the difference between the grant price at the time of issuance and today's stock price.
        We thus utilized this methodology to calculate the options- adjusted 1995 earnings of the 10 stocks which made the greatest contribution to the increase in the S&P 500 Index in 1998.  We excluded Lucent and MCI Worldcom which didn't exist in their current form in 1995 and replaced them with numbers 11 and 12 on the list, Merck and Home Depot.
        Using this method, the option-adjusted earnings for this group of companies in 1995 would have been a staggering 65% lower than what those companies reported at the end of 1995.  In fact, this adjustment to earnings wipes out the reported earnings of four of the 10.  Incredibly, in the cases of Microsoft and Cisco, the derived options expense was roughly 10 times reported earnings.

That the current method understates the expense is clear....

Goldfarb: Admittedly, current stock prices do seem frothy.  And it could be argued that they presently overstate the 1995 options expense and earnings dilution.  But remember - there are six years to go before these options expire!

        Additionally, we would point out that any broad-based basket of option grants awarded since the inception of the Black-Scholes model had ultimate values significantly higher than the Black-Scholes model indicated at the date of grant - even when you apply a discount rate to the stock price on the date of expiration.

Adiusted for options grants, Microsoft's never earned a penny!

Goldfarb: Applying the retrospective valuation methodology to Microsoft's past and current employee base is a quite interesting exercise.  At the current market price, we estimate that the value of options granted since Microsoft went public in 1986 is many times the company's cumulative net income since its founding.

        From this case study, we would invite you to consider the following hypotheses: First, after adjusting for the ultimate value of stock options granted during its corporate life, Microsoft has never earned a penny for its nonemployee owners.  Indeed, when it comes to distributing profits, this global high-tech powerhouse has operated more like an agrarian cooperative than like a publiclyowned corporation.  Secondly, Bill Gates, the icon of late20th century capitalism, could instead be history's most successful communist!
 

These aren't some pie-in-the-sky analyses....

Goldfarb: Even if the current stock prices of companies like Microsoft overstate their intrinsic value, as these companies scramble to buy in shares ... in order to cover past option grants, there is often a huge disparity between the exercise price and the price at which shares are repurchased - resulting in a negative adjustment to shareholders equity as well as intrinsic value.

        There are of course examples of companies both confident and nimble enough to avoid this potential impairment by repurchasing shares as soon as the options are granted and at prices roughly similar to the exercise price of the options.  However, if the P/E is high enough and the number of option grants is great enough, there is no salvation....

        I would ask you: What is the economic value of an enterprise that dazzles Wall Street with reported earnings growth of 20-25% or more quarter after quarter, year after year, ad infinitum, but all of whose earnings are consumed to fund and cap employee compensation expense?

        The economics of such an enterprise are analogous to those of a highly successful sports team whose players' compensation exceeds the team's revenues.  This team may be well in the black in its league standings, but deeply in the red in profitability.