by Maggie Mahar
An extraordinary assumption, as everyone on the Advisory Council knew—or at least had been told. The number was based, in part, on a report by Joel Dickson, a financial analyst at the Vanguard mutual fund group. The Council had charged Dickson with figuring out how much stocks had averaged, after inflation, for the longest period he could measure. Going back to 1900, he came up with 7 percent. A reasonable fellow, Dickson provided no guarantees: in his report to the Council he noted clearly that that there was at least a 50 percent chance his numbers wouldn’t be right.19
Nevertheless, the Council took the 7 percent before-inflation figure, guessed that, in the future, inflation would average roughly 4 percent a year, and came up with 11.28 percent as the likely total return from stocks in the years ahead. (On that basis, the policy makers reckoned, if 40 percent of the money was invested in stocks, everything would work out just fine.)
Trouble is, what was generally considered the most reliable benchmark at the time showed equities had averaged just 10.71 percent a year over the 71-year period from 1926 through 1996. “In a triumph of statistics over common sense, the Council’s plans all assume that stock prices from here on will rise more quickly than they have in the past,” Sloan noted. “Stocks have risen about 1,000 percent since the bull market started in August of 1982. But no tree grows to the sky. Except, of course, for simulated trees in computer models.”
The difference between 11.28 percent and 10.71 percent might not sound enormous, but while a dollar invested in 1926 at 11.28 percent would have become $1,975 by the end of ’96, if that same dollar earned just 10.71 percent it would have grown to only $1,372.
But the real point here is not that the Council should have used 10.71 percent instead of 11.28 percent—but rather that absolutely no one knows how much the stock market will return over the next 10, 25, 50, or 75 years.
“Betting that stock prices will keep rising rapidly because they have been rising rapidly ‘is like the guys on Noah’s ark projecting six more weeks of rain on the 39th day,’ said Joseph Rosenberg, chief investment strategist at Loews Corp. and one of Wall Street’s most respected investors: ‘You can’t believe how dumb a government can be.’”20
Yes you can. At the beginning of 2003, despite the market’s dismal decline, the Bush administration continued to favor diverting some portion of Social Security savings into private accounts, giving individuals the opportunity to gamble the money on stocks. No surprise, the mutual fund industry stood foursquare in favor of the proposal.21
What the plan’s proponents seem to have forgotten is that the Social Security Trust Fund is not an investment club, it is a safety net—and not just for the elderly but for the nation. If that safety net is rent, and retirees have trouble making ends meet, inevitably, taxpayers will wind up bailing them out—either directly, by supplementing their income, or indirectly, by footing the bill for the extra medical care that a larger population of indigent elderly will need.
This is why they call it a Trust Fund.
THE FINAL RUN-UP (1998–2000)
—16—
“FULLY DELUDED EARNINGS”
[There has been] much loose talk about “value creation.” We readily acknowledge that there has been a huge amount of true value created in the past decade by new or young businesses and that there is much more to come. But value is destroyed, not created, by any business that loses money over its lifetime, no matter how high its interim valuation may get. What actually occurs in these cases is wealth transfer, often on a massive scale.
—Warren Buffett, 2000 letter to
Berkshire Hathaway’s shareholders
In July of 1998, Jim Chanos happened upon an intriguing item tucked away in the most recent issue of Business Week. As a short seller, Chanos made his living by paying attention to detail. On this particular day, he was reading a Business Week advertising supplement. There, he found the results of a poll the magazine had taken at its seventh annual forum for chief financial officers three months earlier.
The magazine asked the 160 chief financial officers who attended the forum to respond to the following proposition:
“As CFO, I have fought off other executives’ requests that I mispresent corporate results.” Using “audience response electronic keypads,” they chose from the following three answers:
Yes, I fought them off.
I yielded to the requests.
Have never received such a request.
What was refreshing was the honesty of the answers. One might think that a prudent CFO would be tempted to claim that no one at his company had ever suggested such malfeasance. In fact, only 33 percent responded: “Have never received such a request.”
The majority—55 percent—acknowledged that their colleagues had at least suggested cooking the books: “Yes, I fought them off.”
Incredibly, fully 12 percent went a step further, admitting, “I yielded to the requests.”1
Why Business Week chose to publish the results in its advertising supplement was not entirely clear. What was certain was that by 1998, financial chicanery had become commonplace throughout corporate America. That two-thirds of these CFOs freely admitted that they had been asked to goose the numbers suggested that the type of person who might blanch at such a suggestion had probably fallen off the corporate ladder early on in the bull market.
Creative bookkeeping was not confined to the late nineties—and it was not limited to technology companies. Well-known names such as Oxford Health, Green Tree Financial, Boston Chicken, Sunbeam, and Cendant fell under the weight of their own bad numbers. Abuse was widespread. From 1997 to 2002, roughly 1,000 companies would be forced to admit that the earnings that they had reported were not quite correct.2
The market’s spiral not only pushed share prices to unsustainable heights, it also fostered a corrupt corporate culture hooked on high growth. “Delivering double-digit earnings growth year after year is no longer simply what corporate re-engineers call a ‘stretch goal’ for an organization, or a rare achievement to be celebrated. It’s become a mandate, a benchmark, a test of corporate manhood, an expectation hard-wired into the culture,” wrote The Washington Post’s Steven Pearlstein as he looked back on the bull market. “The addiction to double-digit growth has spread across the corporate landscape to firms in older, mature industries desperate for the ‘growth company’ moniker that qualifies them for Wall Street’s highest reward: a stock price equal to 20, 30, even 40 times earnings.”3
The newspapers reported on the Sunbeams and the Cendants. Nevertheless, the media tended to embrace the theory that these were what The Wall Street Journal called “notable exceptions.” Just a few days before Chanos read the Business Week poll, The Wall Street Journal ’s “Abreast of the Market” column reassured readers: “With accounting questions playing a big part in some spectacular recent stock blowups, investors might wonder just how believable are the earnings now fueling record stock prices. Rest easy: They are more believable than in previous decades.
“Some high profile disasters aside, U.S. corporate accounting has been getting steadily more conservative in recent years, not less so, many experts believe—a view backed up by new research….” The column cited research by several accounting professors and wound up quoting Abby Cohen’s assistant.4
But while the accounting professors provided theory, Business Week had, however inadvertently, done field research. Time would prove its poll correct.
THE WHOLE BUSHEL
Ultimately, the Jerome Levy Forecasting Center, a highly respected, non-partisan, independently funded consulting firm, would expose just how far and how deep the rot had spread in a seminal study of corporate accounting during the bull market: “Two Decades of Overstated Corporate Earnings.”
“Over the past ten years, the financial media have spotlighted case after case of earnings misrepresentations…[but] the focus of public concern remained on finding the bad apples; little attention was paid to the quality of the entire bushel,” the study’
s authors wrote.
“Just how widespread and serious was the overstatement of aggregate corporate profits?” they asked. “The answer is startling. The evidence indicates that corporate operating earnings for the Standard & Poor’s 500 have been significantly exaggerated for nearly two decades—by about 10 percent or more early in this period and by over 20 percent in recent years [emphasis mine]. These figures are conservative—the magnitude of the overstatement may be considerably larger.”5
Both the press and the public were reluctant to face the fact that the problem was systemic. It was easier to acknowledge that a few corrupt CEOs were puffing up earnings statements. To admit that the entire system had been gamed—and not just in the late nineties but throughout much of the bull market—meant questioning the underpinnings of an “efficient” market. If that market was operating on tainted information, just how efficient could it be?
SEC Commissioner Arthur Levitt knew the answer, and by 1998 he was more outspoken than he had been when he first came to Washington. That September, in a speech that he delivered at the NYU Center for Law and Business, Levitt was blunt: “Too many corporate managers, auditors, and analysts are participants in a game of nods and winks. In the zeal to satisfy consensus earnings estimates and project a smooth earnings path, wishful thinking may be winning the day over faithful representation. As a result, I fear that we are witnessing an erosion in the quality of earnings, and therefore, the quality of financial reporting. Managing may be giving way to manipulation. Integrity may be losing out to illusion.”6
Companies like GE boasted of “managing” their earnings so that they rose, consistently and smoothly, quarter after quarter, year in and year out. In this way, GE was able to always meet, if not beat, analysts’ estimates. Of course, in the real world of business some years are better than others; profits do not rise in a straight line. This is why Jim Grant called managed earnings “fully deluded earnings.” Grant quoted Jim Chanos: “The trouble with smoothing a naturally jagged pattern of earnings is that the underlying problems of the business itself are likely to be obscured—until the day when they can’t be any longer.”7
To mask problems, some companies created virtual revenues. One of the simplest ways to do this is to pay customers to buy your goods. And by the late nineties this is precisely what Cisco was doing. George Noble, a Boston-based portfolio manager best known for a successful stint running Fidelity’s Overseas Fund, recalled stumbling onto one of Cisco’s virtual customers. At the time, Noble was attending a road show for B2, a small broadband company that was trying to drum up interest in an IPO:
“To lend credibility to the whole thing, they were pointing out all the big investors they had, like Morgan Stanley—who by the way was also their underwriter. Then they said, ‘We buy all of our equipment from Cisco. We gave Cisco an order for $330 million of equipment and we got 135% vendor financing.’”
Noble perked up. He decided to play dumb. “How does that work? What is vendor financing?”
“Well, the order was for $330 million of equipment and they gave us $450 million of financing at 9% interest and with no payment and principal for the first year,” the company spokesperson explained.
“Is that good?” Noble asked, still playing the innocent.
“The Morgan Stanley banker shot up out of his chair because he wanted to dazzle us with how great the deal was for B2,” Noble recalled. “Everyone knew where I was going with that question except the banker who went on to explain what a great deal it was.”8
At least for B2. The company received $330 million of free equipment plus $80 million in cash. In return, Cisco received $330 million of revenues—on paper. Maybe B2 would be able to repay the loan. Maybe not.
Cisco was hardly alone. Telecom-equipment suppliers were particularly generous with their loans: by the end of 2000, they were collectively owed as much as $15 billion by customers, a 25 percent increase in a single year.9
“EXTRAORDINARY” ITEMS BECOME ORDINARY
All of this was, of course, perfectly legal. Contrary to the conventional wisdom, outright fraud was not the most pervasive accounting problem of the nineties. Most creative accountants played by—and with—the rules.
In the Levy Center report, Walter M. Cadette, David A. Levy, and Srinivas Thiruvadanthai outlined the two major ways that corporations used the rules to inflate their earnings:
by focusing investor attention on operating earnings rather than net income; and
by paying executives in stock options rather than cash, and so masking the expense.
In theory, what Standard & Poor’s calls “operating earnings” is a fairly clean concept. These are the profits that a company makes in the ordinary course of doing business: the revenues it takes in by selling its product or service, minus taxes and expenses. Operating income does not include extraordinary one-time gains or expenses. For instance, if a company sells a division, this is a nonrecurring gain that will not appear the next year, and so should not be included under operating income. Similarly, if it lays off 1,000 employees and gives them generous severance packages, this is a one-time expense, outside of the ordinary cost of doing business, and so it’s not supposed to be deducted from operating income. By comparing operating income year over year, an investor can see how the core business is faring.
That is the theory. But, as the Levy Center’s report explained, over the course of the bull market companies found various ways to goose operating income. Sometimes, a company would mislabel one-time (and nonrecurring) revenue as operating revenue—as if it had sold products. This is precisely what IBM did in 1999 when it used a one-time $4 billion gain from its sale of its global network business to offset that year’s normal expenses.
On other occasions, a company would sweep true operating expenses into the “one-time expense column.” Once tucked into that column, the expense did not have to be subtracted from operating income.
In 1999, Warren Buffett elaborated on how this bit of legerdemain works: “A large chunk of costs that should properly be attributed to a number of years is dumped into a single quarter, typically one already fated to disappoint investors. In some cases, the purpose of the charge is to clean up earnings misrepresentations of the past, and in others it is to prepare the ground for future misrepresentations. In either case, the size and timing of these charges is dictated by the cynical proposition that Wall Street will not mind if earnings fall short by $5 per share in a given quarter, just as long as this deficiency insures that quarterly earnings in the future will consistently exceed expectations by five cents per share.”10
Because investors were so focused on the notion of quarterly earnings growth, the ruse worked. “If you take something as a [one-time] restructuring charge, investors will forgive you immediately,” explained Robert S. Miller, an executive brought in to clean up Waste Management. “We’ve almost lost the notion of what are earnings and what are one-time charges.”11
Operating earnings have a place. The only way an investor can see whether a company’s profits are growing, year by year, is if he has a clean snapshot of annual revenues minus annual expenses. Occasional one-time restructuring charges should not blur the picture too much. Trouble is, over the course of the bull market one-time extraordinary charges became “all-the-time” extraordinary charges.
Kodak took the trend to an extreme. From 1991 to 1998, Kodak took six extraordinary write-offs for restructuring costs that totaled $4.5 billion—more than all of its net profits for the preceding nine years. Granted, over that time Kodak had been in a major transition period, exiting five major business lines while sales dropped 25 percent.12 Nevertheless, over a period of years, these “extraordinary” expenses had become a regular part of its business. Yet they were not subtracted from operating income. “Just how ‘one-time’ are restructuring costs if they occur every year?” asked Jim Chanos. “If opening and closing factories and hiring and firing employees are integral to a manufacturer’s business, shouldn’t they be t
reated that way in its earnings reports?”
Others agreed. In 2003, Gail Dudack, SunGard’s chief market strategist, pointed out just how ordinary “extraordinary items” had become (see chart “Extraordinary Items as a % of Reported S&P EPS,” Appendix, page 465). Up until the mid-eighties, “extraordinary charges were minuscule,” she pointed out. “But then they began to grow. This year, Standard & Poor’s forecasts that these one-time items will equal 38% of earnings per share on the S&P 500. Accounting gimmicks are making it almost impossible to analyze or compare income statements.”13
Meanwhile, the media helped Wall Street keep all eyes focused on the numbers that corporations wanted investors to see. “Perhaps nowhere is the symbiosis between the media and the market more evident than in the reporting of operating results,” the authors of the Levy Center report observed. “When a company announces its earnings, it typically issues a press release including operating earnings and management’s comments on its results. The media pick up and broadcast these figures even though, unbeknownst to most consumers of the information, the company may file quite different results later with the SEC in its 10Q financial statements. The casual definition of operating results a company employs when issuing a splashy announcement for the public may not pass muster with the SEC. Yet firms do not publicize nor do the media cover the more regulated results filed with the authorities.
