Still Bullish After All These Years

Jeremy Siegel has never lacked enthusiasm for the market.

Even nine years into one of the strongest stock market recoveries in history, “this bull market is not over yet,” said Siegel, who has been a University of Pennsylvania finance professor since 1976.

That should not come as a surprise. His 1994 book, “Stocks for the Long Run,” helped cement the conventional wisdom that stocks are a better bet for long-term investors than other popular investments such as bonds and gold.

Siegel’s timing couldn’t have been better. A year after his book appeared, U.S. stocks went on an epic run. The S&P 500 Index more than tripled from 1995 to 1999.

As the market marched higher, a growing chorus of observers warned that the rally was overdone. One of them was Yale professor Robert Shiller, Siegel’s friend and MIT classmate, who argued in his 2000 book, “Irrational Exuberance,” that stocks had become dangerously rich. Shiller’s cyclically adjusted price-to-earnings, or CAPE, ratio for U.S. stocks climbed to 44.2 by the end of 1999, well above the historical average CAPE ratio at that time of 15.5 since 1881. To this day that’s the highest CAPE ever recorded.

Shiller wasn’t the only one concerned about the market. Siegel, too, warned in March 2000 that technology stocks were a “sucker bet,” days before the tech-heavy Nasdaq 100 Index tumbled. But Siegel also thought that many nontech stocks were reasonably priced. Armed with two centuries of stock market data, he predicted two months later that stocks would continue to deliver their historical average real return of 7 percent a year over the medium and longer term.

It hasn’t quite worked out that way. The S&P 500 has returned 5.4 percent annually since 2000 through November, including dividends, or a real return of 3.2 percent. During the same period, the Bloomberg Barclays U.S. Aggregate Bond Index and gold have returned 5.1 percent and 8.7 percent, respectively.

Two decades after its record peak, the CAPE ratio is flashing warning signs again. The S&P 500 has tripled since 2009. The CAPE ratio stands at 31.3 as of November, nearly double its long-term average of 16.8. It was only higher on the eve of the Great Depression in 1929 and during the dot-com mania of the late 1990s.

Nevertheless, Siegel said the “stock market is fairly valued now.” He expects it to deliver a 5 percent real annual return from its current level, well above the CAPE’s implied earnings yield of 3.2 percent. The CAPE ratio, in other words, is a false prophet.

The 72-year-old Siegel is one of few finance thinkers with the gravitas to take on the vaunted CAPE ratio. Siegel, who was born in Chicago, excelled with numbers from the beginning. He graduated with a B.A. in math and economics summa cum laude from Columbia University in 1967 and received a Ph.D. in economics from MIT in 1971. He went on to a distinguished academic career, first as an assistant professor at the University of Chicago from 1972 to 1976 and since then as a professor at the Wharton School.

Over the years, he has written on a dizzying array of subjects such as monetary policy, indexing, inflation, taxation, interest rates, bank regulation, valuation and the business cycle. He has won myriad awards for his academic research. But “Stocks for the Long Run” brought Siegel fame outside the ivory tower and, in the minds of many investors, made him the archetypal bull.

Underlying his disagreement with the CAPE ratio is one of the oldest debates in accounting. For decades, generally accepted accounting principles, or GAAP, required public companies to account for assets based on their acquisition cost. Critics complained, however, that ignoring assets’ market prices misstated their true value.

In response, the Financial Accounting Standards Board issued the first of several “mark-to-market” rules in 1993, which require companies to account for some assets based on their market value rather than historical cost. The basic idea is that companies must adjust the value of those assets when their prices change and recognize a gain or loss, as appropriate.

The problem is that asset prices tend to decline during recessions, and marking them to market pulls down earnings when companies are struggling to remain profitable. As Siegel sees it, “when GAAP changed to mark-to-market accounting, it changed the character of earnings during recessions, making them much more severe.”

That’s been true of earnings recessions since the rule change in 1993. The S&P 500’s GAAP earnings declined 54 percent from their peak in September 2000 to trough in December 2001, according to Shiller’s data. Earnings also declined by a breathtaking 92 percent from June 2007 to March 2009. Siegel points out that the 2008 financial crisis saw a “much worse drop in earnings than the Great Depression, which doesn’t make sense because the Depression was five times worse.” In fact, those are by far the worst two earnings declines since the Committee on Accounting Procedure, a precursor to FASB, began issuing formal accounting rules in 1939.

And lower earnings, of course, translate into higher price-to-earnings ratios. According to Siegel, that’s why the CAPE ratio has been perplexingly high since the change to mark-to-market accounting. The CAPE was below its long-term average during 56 percent of months from 1881 to 1992. But since 1993, that number has dropped to just 2.3 percent.

“Using operating earnings,” Siegel said, “you don’t see the same overvaluation as with GAAP earnings.” Operating earnings exclude the impact of mark-to-market accounting and other nonrecurring items.

According to my calculations, the average CAPE ratio using operating earnings has been 23 since December 1998 — the earliest date for which it can be calculated — compared with 26.5 based on GAAP earnings. And the CAPE ratio is 27.3 as of November using operating earnings, compared with 31.3 for GAAP earnings.

Siegel concedes that stocks aren’t cheap by either measure, but he said that the fair value of U.S. stocks is higher than the CAPE’s historical average of 16.8 implies. “Historical CAPEs would be higher if mark-to-market applied to the entire series because earnings would be lower due to write-downs,” he said. He also thinks that with the cost of owning an S&P 500 index fund approaching zero, investors are willing to pay more for stocks than they once were.

There are reasons to wonder whether the CAPE has lost its predictive power, as Siegel suspects. Consider that the S&P 500 returned 10.3 percent annually from February 1926 to December 1992. The CAPE ratio averaged 14.7 during that period. Since then, the average CAPE has jumped to 26.4, and yet the S&P 500 still managed to return 9.7 percent a year through November.

Also, the CAPE has been useless in recent years. It climbed above its long-term average in June 2009 and kept climbing. Meanwhile, the S&P 500 has returned a whopping 15.7 percent a year since then. In fairness, the CAPE has never been perfectly prescient, and I think it’s too soon to throw it in the dustbin of financial models. Still, Siegel has been right to ignore the CAPE’s warnings in recent years, the right man with a timely message.

Source: Bloomberg Gadfly,