Stocks Still Don't Look Very Expensive
I recently argued that today’s stock market is not at all like the 1997-2001 dot-com bubble and that, in fact, the simple benchmark developed by John Burr Williams, the original value investor, indicates that investors can anticipate a long-term return on the S&P 500 that will be well above the return on Treasury bonds. Today, I look at another investment benchmark that suggests that stocks are not at all bubbly.
Robert Schiller, Yale’s famed bubble bellwether, has cautioned that his cyclically adjusted price-earnings ratio (CAPE) is at an unusually high level, which is often a prelude to a downturn in stock prices. However, he also notes that it is impossible to predict when or how far prices will fall. As Keynes famously observed, “Markets can remain irrational longer than you can remain solvent.”
But is the stock market really irrational now? One reason Shiller’s CAPE may be high is that corporate earnings over the past decade were depressed by the Great Recession. If the average unemployment rate over the next decade is lower than over the past decade, CAPE may return to its historical average because earnings increase, not because stock prices fall.
Another reason that CAPE is relatively high is that, as Shiller notes, yields on 10-year Treasury bonds are barely above 2 percent. Lower interest rates increase the present value of any cash flow. Therefore, when interest rates are low, price-earnings ratios tend to be high. This inverse relationship between interest rates and P/E implies a positive relationship between interest rates and the earnings yield, E/P. The cyclically adjusted earnings yield, CAEP, is the inverse of Shiller’s cyclically adjusted price-earnings ratio, CAPE.
This figure shows that the cyclically adjusted earnings yield for the S&P 500 and the ten-year Treasury rate have tended to move up and down together:
The earnings yield is not a perfect measure of the return from owning stocks. Investors get dividend checks, not earnings checks, and the stockholders’ eventual returns depend on how successfully companies reinvest their retaining earnings. If retained earnings are squandered on failed ventures, they may as well not have been earned in the first place. If, on the other hand, a company earns a high return on its investments, a dollar of retained earnings is worth more than a dollar to shareholders. This is presumably why Berkshire Hathaway has never paid a dividend—they are confident that they can earn a higher return than shareholders could earn on their own.
Despite these limitations, the correlation between the earnings yield and bond yield in the above figure is striking. So, too, is the fact that since 2009, the earnings yield has consistently been above the 10-year Treasury rate, suggesting that stocks have been attractively priced relative to Treasury bonds.
However, matters are not quite this simple. The earnings yield is an estimate of the real, inflation-adjusted return from stocks. It should consequently be compared to the real, inflation-adjusted return from bonds. In 1979 Franco Modigliani and Richard Cohn argued that, during the inflationary 1970s, investors mistakenly compared earnings yields to dollar interest rates when they should have been looking at inflation-adjusted interest rates. They concluded that, because of this market-wide mistake, earnings yields were about twice what they should have been. So, stock prices were about half what they should have been. Stocks were a bargain.
I calculated real inflation-adjusted interest rates each year by using the change in the consumer price index over the next year. For example, I adjusted the dollar interest rate in January 2000 by using the change in the consumer price index between January 2000 and January 2001. For the past year, I used a 2 percent rate of inflation, the average over the past several years. My calculations are an imperfect measure of investors’ inflation expectations, but they may not be systematically biased.
The next figure compares these real interest rates with Shiller’s cyclically adjusted earnings yield. What immediately leaps out is how close these have been since 1981, and how far apart they were in the 1970s. A comparison of the two figures confirms the Modigliani-Cohn argument. In the 1970s, investors evidently priced stocks so that earnings yields would be close to dollar interest rates rather than real interest rates. If this was a mistake, stock prices were much too low in the 1970s.
As it turned out, the annual return on the S&P 500 between 1979 (when Modigliani and Cohn made their argument) and 1989 was an eye-popping 18 percent. Part of the reason was the drop in interest rates in the 1980s, but part of the reason may well have been that investors made a colossal mistake in the 1970s.
There is a story about two finance professors who see a $100 bill on the sidewalk. As one professor reaches for it, the other says, “Don’t bother; if it was real, someone would have picked it up by now.” Finance professors are fond of saying that the stock market doesn’t leave $100 bills on the sidewalk, meaning that if there was an easy way to make money, someone would have figured it out by now.
There is truth in that, but it is not completely true. Stock prices are wacky during speculative bubbles and panics. The 1970s were another time when the glib assumption that markets know best was battered and bruised. It was not a slight mispricing. It was an expensive car wreck that left truckloads of $100 bills on the sidewalk.
Today, if the low level of real interest rates is taken into account, stocks do not look particularly expensive. The cyclically adjusted earnings yield is 3.3 percent, which looks pretty good compared to a real ten-year Treasury rate of essentially zero percent. We are led to the same conclusion as the John Burr Williams benchmark calculation. Nobody knows whether stock prices will be higher or low tomorrow, or next week, or next year. But for value investors who don’t try to predict short-term price fluctuations, stocks do not look expensive. A long-term investment in stocks is likely to be more rewarding than a long-term investment in bonds.