WASHINGTON, July 28 (UPI) -- Coming to the end of the second quarter earnings season, the naturally optimistic business media are caroling how second quarter earnings have "exceeded expectations." Maybe so, but much accounting chicanery has been used to get them to do so, and the "expectations" had themselves been manipulated.
Let's start with the Internet auction site E-Bay, one of the outstanding successes of the earnings season, with net income announced at $109.7 million for the second quarter, or 33 cents per share, well above the company's "guidance" of 30 cents per share. Of course, in an effort to make the results look even better, and justify the sky-high share price of around $110 at midday Monday, up almost 100 percent this year, and at a historic price-earning ratio of over 200, the company claimed "pro forma" earnings of $120.9 million.
However, these figures ignore the effect of stock options, which E-Bay is sufficiently old-fashioned not to include on its income statement and to value only in a note to the accounts. E-Bay's precise expense for stock option issuance will appear only in the company's 10-Q statement, not yet published -- heaven forfend that analysts should be able to make their own adjustments to the company's financial position at the time it is announced. But in the first quarter of 2003, E-Bay announced that its stock based compensation expense for the quarter, using a Black-Scholes calculation methodology, was $29.6 million, which, if deducted from net income, would have reduced that quarter's net income from $104.1 million to $74.5 million, a fall of 28.4 percent. Unpleasant, certainly, and it should be factored into the stock price, but not earth-shatteringly bad.
There's just one problem. The Black-Scholes valuation methodology makes a number of assumptions, most crucially the "volatility" of the underlying stock, which assumptions can of course be manipulated -- nobody really knows what a stock's likely volatility is. In E-Bay's case, we learn that in the first quarter, the company granted options on 7.33 million shares at an average exercise price of $78.51 per share. To get a Black-Scholes value of $29.6 million, each option must have had a value of 29.6/7.33, or $4.04 per option, i.e. 5.14 percent of the underlying share price.
But these are 10 year options, which I take to be exercisable at close to the market price on the date of grant (this is normal practice, and the average price of $78.51 corresponds fairly closely to E-Bay's market price in the first quarter.) 5.14 percent seems awfully low.
It is awfully low. In midday Monday trading (prices from Yahoo Finance) an E-Bay call option with a strike price of $110, close to the current price, would trade (mid price basis) at $11.40 for a January 2004 expiration (just under 6 months) $20.20 for a January 2005 expiration (18 months) or $26.80 for a January 2006 expiration (30 months.) 10.4 percent, 18.4 percent and 24.3 percent of the current share price, in other words. By all means take some kind of deduction for the non-tradability of a 10 year employee share option, but since the true "market" value of an unrestricted such option, by extrapolation, is more than 30 percent of the stock price, it surely cannot be valid to value an option that is only moderately restricted at less than 20 percent.
Take that 20 percent value, and E-Bay in the first quarter granted options worth not $29.6 million but $115.1 million.
Completely wiping out its net income. The second quarter had similar net income, apparently a similar number of shares issued, and certainly a higher average share price, so that quarter's net income must have been wiped out, too.
Alternatively, you can look at option exercises, and the value given away to employees by allowing them to buy shares cheaply. In the first quarter, 4.50 million share options were exercised, at an average exercise price of $46.30 per share, a $32.21 discount to the average issue price (which we can take as a proxy for the average market price during the quarter.) Thus 4.50 million multiplied by $32.21 was given away to management by stockholders, or $144.9 million.
Again, more than the company's net income for the quarter.
The truth is, that E-Bay is giving away to its management, quarter by quarter, more money than it's making. For these guys, it's still 1999!
For Citigroup, which also surprised the market positively by its second quarter results, the problem is different. Since 1982, the group has benefited enormously from a secular decline in dollar interest rates that lasted twenty years, but appears now to have reversed, with 10-year Treasury note yields up around 110 basis points (1.1 percent) since their low in mid June. Citigroup hedges its overall interest rate position to a large extent, but not completely; the May 10-Q statement remarked that a 100 basis point increase in the U.S. dollar yield curve would have an effect of reducing group earnings by $765 million. However, that does not take account of reductions in fee income from bond underwriting, nor of losses on the bank's trading portfolio. In the second quarter of 2003, debt underwriting produced income of $507 million, up 41 percent from the second quarter of 2002, while fixed income trading produced income of $1,257 million, up 10 percent from the second quarter of 2002. You can expect both those figures to drop by 50 percent or more in a market in which the trend for interest rates is generally upwards.
In addition, Citigroup will inevitably suffer as interest rates rise and mortgage refinancing becomes more difficult from a rise in delinquency rates on its credit card portfolio, net credit losses on which in the second quarter of 2003 were $1,887 million, or 6.08 percent of outstandings, at an annual rate. This is close to the all-time high in the loan loss ratio from credit cards; needless to say, a rise in interest rates will cause it to spike further.
Apart from any "creative accounting" in the figures themselves, therefore, you can reckon that at least $1 billion of Citigroup's $4.3 billion of Net Income was due to exceptionally favorable conditions in the U.S. dollar bond markets, conditions which appear now to have disappeared, possibly for a lengthy period. It also doesn't help that Citigroup was today fined $145 million in respect of its Enron activities, and has undetermined further outstanding possible liabilities for its underwritings during the "bubble" of 1996-2000. All this is entirely outside the possible effect of a renewed dip into U.S. recession, or a drop in the equity markets, of course.
General Electric, on the other hand, switched its stock options accounting to expense options (albeit on a Black-Scholes valuation basis) in 2002, so that problem, at least, has been removed from its income statements. In GE's case, unlike in that of E-Bay, the problem comes from being too old a company rather than too new -- it is the accounting for its pension plan.
During the 1990s, GE several times revised the accounting basis for its pension plan, so that the assumed return on plan assets was increased and the required contributions reduced, and then accrued to its income statement, not the actual returns on the plan, but an assumed return on the plan. In 2002, this all came back to haunt it, albeit not to the extent of disclosure in the company's profit figures. At the end of that year, GE's retiree health and other benefit plan was un-funded to the extent of $6.01 billion, of which only $2.98 billion had been recognized on the balance sheet, while the pension plan had an excess funding of $4.55 billion, but had recognized $14.07 billion in pension "earnings" on the balance sheet. Net, there was thus, at December 31, 2002, $12.55 billion on GE's balance sheet of "prepaid pension asset" that didn't actually represent anything.
It gets worse. In 2002, after deducting $750 million of retiree benefit costs, GE was still able to recognize $806 million of "net cost reduction from principal postretirement benefit plans" in its income statement. This in spite of the fact that in that year the plans lost a total of $5,476 million from their actual value, and accrued an additional $1,384 million in service costs for benefits earned. In other words, GE's profit benefited by $806 million from its pension plan, when if it was recording actual revenues and costs it should have booked costs totaling $6,860 million. The difference, $7,666 million, was more than half GE's recorded net income for 2002.
In addition, there must be some question as to whether the actuarial value of the liabilities for GE's pension funds is adequate. GE assumed in 2002 a rate of return of 8.5 percent on the funding of its pension and health obligations (down from 9.5 percent in 2001 -- that change alone resulted in a further $2,217 million in "actuarial loss" which did not show up on GE's income statement. However, the plans are now relatively short term in duration, with "benefits paid" from the pension plan at $2,197 million being nearly twice the "service cost" of $1,107 million, which tells you there is a high proportion of retirees in the plan.
If we assume that equity investments (56 percent of the fund) can no longer be relied upon to yield more than their long term average of about 9 percent per annum, and that the current level of 26 percent of the fund in fixed interest investments (yielding at most 5 percent with any degree of safety) is a minimum, not a maximum, given the fund's relatively short "duration" then the average return on 82 percent of the assets is only around 7.7 percent. If the remaining 18 percent of the assets is largely in real estate, then it too will currently be having its difficulties. In short, an assumed return of 7.5 percent is about the maximum that is actuarially respectable, and even that is a little high.
GE states proudly that it has made no pension contributions since 1987. If it has to make them in 2003, even without any "top-ups" to restore full funding of the plans, or any write-offs of the $12.55 billion in mythical assets on its balance sheet, then the net cost to GE's income statement will be $2.19 billion, about 15 percent of net income. That, as a minimum, is the decline we should expect going forward, from this one item alone.
These are just three of the companies recording earnings in the past few weeks that were positively received by the markets. Overall, recorded profits on S&P 500 companies so far are up about 8.3 percent in the second quarter compared to 2002 (before taking account of any little hiccups such as those above) compared with the 5.3 percent that the market, guided downwards by corporate investor relations staff, had been expecting at the beginning of July.
Of course, the third and fourth quarter of 2003 are supposed to see an economic "recovery. According to Thomson FirstCall, quoted in TheStreet.com, expectations are for an earnings gain of 13.6 percent in the third quarter, and 21.6 percent in the fourth quarter.
Does anybody, whose professional living does not depend on issuing, trading, holding or selling stocks, really believe that gains of that magnitude will occur, at least not without even further accounting chicanery?
On Wall Street, clocks have stopped; it is a financial Brigadoon, perpetually marooned in 1999!
(The Bear's Lair is a weekly column that is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that, in the long '90s boom, the proportion of "sell" recommendations put out by Wall Street houses declined from 9 percent of all research reports to 1 percent and has only modestly rebounded since. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)
