Stocks Are Still an Oasis
By JEREMY J. SIEGEL WSJ Op-Ed 7.25.02
Frank J. Williams, a great stock trader of the early 20th century, said, "The market is most dangerous when it looks best; it is most inviting when it looks worst." Well, certainly the market cannot look much worse than it has over the past several weeks, Yesterday morning, the decline of the Standard & Poor's 500 Index surpassed the horrible 1973-74 bear market, before posting a strong afternoon rally. All this has occurred despite the fact that the real economy has done far better than almost anyone had expected following the Sept. 11 terrorist attacks. Why is this happening?
In a nutshell, the lack of dividends and disbelief that the earnings reported by firms are real and can substitute for this cash return. In past bear markets, the dividend yield served as a natural cushion for falling stocks. As long as investors believed that the dividend payment was safe (and cash dividends declined very little during bear markets), they would buy when stocks fell to levels where the dividend yield became attractive.
Hard Landing
But as dividend yield fell towards zero during the bull market of the 1990s, stock prices became solely dependent on their earnings. And as long as faith in those earnings could be sustained, stock prices continued to rise. But this rise was very precarious. As the faith in earnings faded, there was nothing that investors could fall back on to support prices. The traditional cushion has been taken away, and the landing was bound to be very hard.
The roots of the current malaise stem from two sources: first, management-defined earnings that differed from reported earnings and could not be adequately monitored; second, managements' exclusive focus on delivering capital gains, instead of cash dividends.
Earnings manipulations actually began in the 1990-91 recession when firms realized that huge write-offs connected with discontinued operations, which caused reported earnings to fall sharply, were greeted with joy by Wall Street analysts. "Operating earnings," which excluded all sorts of special "one-time" charges and which had no formal definitions in accounting standards, became management's new target. Investors sent stock prices higher if management exceeded earnings estimates and sold mercilessly if earnings fell short (or only "met") management's targets.
During the latest bull market, firms also found that paying dividends bestowed little benefit to the stock price. This was fine with managers since they didn't have to turn operating earnings into hard cash. The inability to deduct dividends from corporate taxes encouraged management to use debt financing, where interest was deductible. Furthermore, tax-sensitive investors greatly preferred the lower-taxed (and often long-deferred) capital gains instead of receiving fully taxable dividends.
Stock options worsened the trend towards substituting capital gains for dividends. Options became valuable only when the price of the stock rose; dividends were of no use to the option holder.
The focus on earnings instead of dividends stood in sharp contrast to the entire history of stock returns. Until the last 10 years, about three-quarters of stocks' real return has come from dividends and not capital gains. In fact, dividends were the primary means that firms had to show that earnings were real. Investors care little about how firms defined their earnings as long as hard cash flowed to investors.
Federal tax law played a large role in creating the wrong incentives for both management and investors. Allowing deductibility of debt interest, but not of dividends, rationalizes the heavy use of debt. Furthermore, options grants were encouraged because they could be deducted from taxes, but not reported as expenses. I do not wish to exonerate executives who looted their companies, auditors who committed fraud, and analysts who hyped stocks because of banking fees paid to their firms, but many of the problems that we have today have their roots in our Byzantine tax code that distorts incentives and invites fraud.
Yet amid all the gloom, there is reason for optimism. To restore credibility of earnings reports, Standard & Poor's has proposed a new standard for measuring earnings in a white paper released in May entitled "Measures of Corporate Earnings." S&P, with the help of Warren Buffet, Prof. Eli Bartov of New York University and top accountants from major Wall Street firms, has defined a new tough measure called "core earnings" that indicates the ongoing profitability of a company's core businesses. Core earnings subtracts options expenses, portfolio gains and most restructuring charges from reported earnings, while eliminating the "write down" of asset values, which the committee rightfully regarded as incidental to ongoing profitability.
The core earnings standard is extremely demanding. A review of all S&P 500 firms showed that core earnings in 2001 were only 83% of reported earnings (under generally accepted accounting principles) and only 66% of operating earnings (assuming JDS Uniphase, which took a whopping $56 billion write down, is excluded). Options expenses comprised almost all the 17% difference between core and reported earnings, as the net effect of other adjustments (including exclusion of the gains from pension portfolios) almost exactly offset each other.
The core earnings concept allows us to take a tough look at real earnings. "Top down" estimates for reported earnings for the S&P 500 Index for 2003 (estimates made by macroeconomists, not the "bottom-up" estimates for operating earnings that are more than one-third higher and are made by analysts with rose-colored glasses), are $45.00. If we subtract 17% from these reported earnings for option grants, we are left with $37.35 core earnings. With the S&P 500 now below 800, stocks are trading at about 21 times core earnings.
But core earnings next year are apt to be higher than $37.35. I believe that option grants will go the way of the bear market and will be far lower in future years. Options grants were necessary for many firms to attract employees, especially in the technology sector, during the height of bull market. Today, many of those same employees should be thankful to have a paying job. If options expenses decline by one-half next year, the price/earnings ratio of the market is 18.6, and if the tech sector is excluded (which, despite the beating it has taken, is still in many cases too richly priced), the P/E ratio of the 420 nontech firms falls below 18.
The bears will say a P/E ratio of 18 is still too high since the average is 15 over the past 50 years and P/E's often drop below average in a bear market. But that average is not measured against this gold standard "core earnings" definition I have used. Furthermore, my research suggests that the proper P/E level for today's market, with its low transactions costs, low inflation, and favorable tax rates for capital gains, should be in the low 20s, not 15.
What about underestimated pensions costs? Bears maintain that the market decline confirms that the return estimates used by most firms on their pension portfolios are far too high. But there is good news and bad news in the market decline. Yes, the value of the pension portfolios has fallen. But from these depressed levels, the return estimates made by many pension funds are finally close to being correct. As ridiculous as a 12% return estimate was at the market peak in March 2000, a 6% to 7% return estimate is far too low from current stock market values, which I believe will approach, if not exceed, the 7%-plus-inflation that has been realized over all long-term periods.
The bears say that history shows that there are long bull markets followed by extended periods of stagnant or even declining markets. Even if market history can be divided into bull and bear segments (and I doubt they are as regular and predictable as some historians believe), it is patently wrong to warn investors about stocks now that they have declined by more than 50%.
There have been six major stock market peaks (including the 1929-32 stock market crash) in the past 100 years. To be sure, investors would have suffered significant negative returns if they invested at these market peaks. But after the market has dropped by 40%, subsequent five year returns have averaged 8.6% per year above inflation and none has been negative. And all subsequent 15-, 20-, and 30-year returns have not only been positive, but have also been above the 7% long-run average real return on stocks.
Stocks Over Bonds
But will stocks necessarily be more rewarding than alternative assets? I can hear the bond enthusiasts reminding investors that the 8.7% real returns on U.S. Treasury bonds since 1982, when the great bull market began, have now topped stocks' 8.4% return over the same period. But that is because interest rates on long bonds fell from 14% to just below 4%. It is mathematically impossible for bonds to repeat those returns over the next decade, even if interest rates fall (God forbid!) to Japanese levels of 1.5%. I dare any bond enthusiast to look me in the eye and maintain that an investor who buys the 10-year U.S. Treasury bond at the current yield of 4.4% will outperform an investor who buys a diversified portfolio of common stocks over the next decade.
None of the above analysis guarantees that we are at the bottom of this bear market. But many of the current fears about earnings quality are overdone. By the toughest definitions of earnings, the prices of stocks are now more than reasonable. History is definitive that once investors have suffered this much pain, subsequent stock returns will be very rewarding. Amid all this gloom, the market looks very inviting to me indeed.
Mr. Siegel is a professor of finance at the Wharton School and author of "Stocks for the Long Run" (McGraw Hill, 1998).
Updated July 25, 2002 12:30 a.m. EDT
online.wsj.com |
