The Shiller CAPE ratio.
For more than a century, from 1880 through 1990, the stocks making up the U.S. market could be purchased for roughly 15 times the amount of their average annual corporate earnings, as measured over the previous decade. That is, if the companies making up the stock market generated an aggregate of $2 trillion in net profits over the past 10 years, thereby posting average annual earnings of $200 billion, then stock market capitalization was probably around $3 trillion. The ratio wasn't fixed and immutable--it might be higher so that stocks cost $4 trillion, or lower so that they were $2 trillion. But they were very unlikely to be as high as $5 trillion or as low as $1 trillion.
This ratio, that of stock market capitalization to companies' average earnings, became known as the Shiller CAPE ratio. Since the mid-1990s, when it was introduced by Shiller, it has served as an informal measure of stock market value. When the Shiller CAPE figure is well above its historic norm, stocks are said to be expensive. Stock returns over the next few years will likely be low. If, on the other hand, the Shiller CAPE ratio is unusually low, stocks are thought of as cheap and therefore poised for good performance.
It's easy to see why people believed this when looking at the picture, particularly when considering that the big blip to the right (more on that shortly) occurred afterthe publication of the thesis.
In the first 110 years of the data set, only six times did the Shiller CAPE ratio exceed 23 or fall below 7. The indicator went high in 1900, in the late 1920s, and then again in the early 1960s. Each period was followed by rocky stock market performances. It went low on three instances as well, in 1920, 1932, and in the early 1980s. The 1932 signal was premature, but 1920 and the early 1980s were terrific buying opportunities, being the start of huge market rallies.
These insights, however, came in hindsight. Even as professor Shiller first released his research, in 1996, the 110-year pattern was changing. At that time, the CAPE ratio had hit 25--a level that previously had led almost immediately to a market slide. On this occasion, though, stocks kept rising, more than doubling over the next four years.
Eventually, from 2000-02, the bear market did arrive, as predicted several years previously by the Shiller CAPE ratio. That wasn't a helpful call. Yes, from the time that the indicator crossed its high band in 1996 to the time of the 2002 market trough, stocks had below-average results. However, moving the end date by one year in either direction increases the stock return to average, and moving the end date by two years leads to strong results. There was nothing in the CAPE ratio to indicate when end points should be used.
The CAPE's informativeness did not improve. After showing stocks to be overpriced at the 2002 market bottom, the CAPE ratio was not particularly prescient in foreseeing the 2008 market crash. True, the ratio was elevated entering 2008--but it had been similarly elevated for the previous five years.
Then, after showing a brief neutral signal in 2009, the CAPE measure promptly went negative again, early in the current bull market.
As Livermore points out, since 1990 the Shiller CAPE measure has spent 98% of the time above its alleged norm and only 2% below. Apparently, he writes, stock valuations have reached a "permanently high plateau." I prefer a different allusion: The Schiller CAPE ratio has achieved a New Normal.
Livermore suggests two reasons why. The larger of the reasons is an argument that Wharton professor Jeremy Siegel has previously made, that in 2001 the Financial Accounting Standards Board changed how companies account for balance-sheet "goodwill" that is created when they pay above book value to purchase another company. Unlike in the past, companies must now immediately write down goodwill if the acquired company's worth is judged to be permanently impaired. That approach depresses current earnings when compared with those of the past.
The goodwill effect equals four points of the Schiller CAPE ratio, which drops from 24.5 under the traditional calculation to 20.6 for Livermore's version. Effectively, then, that single accounting adjustment alters the ratio's signal so that instead of suggesting that stocks are abnormally high in price, they are instead only on the higher side of normal--with plenty of history suggesting that they could move higher yet.
A second, smaller issue is that of dividend-payout ratios. It turns out that if companies distribute more of their profits as dividends (as opposed to reinvesting those monies back into their businesses) then the Shiller CAPE ratio declines. As dividend-payout rates have declined over the years, the Shiller CAPE ratio has gradually been inflated. Livermore figures this effect to be 1 percentage point per year.
Unfortunately, these items don't save the Shiller CAPE ratio. They don't eliminate the unprecedentedly long climb of the 1990s, when the ratio showed stocks at dangerously high levels for year after year. Nor do they correct the five-year negative indication during the mid-2000s, when stocks were posting healthy returns.
The picture is better, to be sure. The goodwill-adjusted chart does much better post-2008. Unlike the original CAPE ratio, the new version suggests pretty strongly to get into stocks near the bottom and to stay in them throughout the rally.
The exercise is forced, though. Such is the story as told by Shiller's CAPE ratio, as adjusted by Livermore. The ratio could also be adjusted in many, many other ways, reflecting not only many other changes in accounting practices but also broader changes in the business world. Many more earnings now come from health care and other consumer services, for example, and fewer from manufacturing. That change surely should affect CAPE ratios. So, too, might modern items like technology-network effects.
Also, selecting a different time frame leads to different conclusions. When calculating what he calls the Pro-Forma version of the CAPE ratio, Livermore changes the time of the data set. Rather than run from 1881-1994, as does Shiller's ratio, Livermore looks at the 1954-94 average. This is not a nefarious change; Livermore like Shiller before him starts the exercise when the data permit. However, altering the time frame does affect the results.
As Livermore's goodwill adjustment does not address the large 1990s' anomaly, I am not fully convinced. It's well researched, and it's a part of the story, but it is not a fix. The struggles of the CAPE ratio's signal will not end by making that change. There remains something else that appears be different since 1990, something that sent the CAPE ratio to a different regime.
What that is, I do not know. What I do know is that when an empirical measure begins to sputter, it's probably best to move on. It's not as if the ratio's use is supported by theory.
John Rekenthaler has been researching the fund industry since 1988. He is now a columnist for Morningstar.com and a member of Morningstar's investment research department. John is quick to point out that while Morningstar typically agrees with the views of the Rekenthaler Report, his views are his own.