The cyclically adjusted price-to-earnings (CAPE) ratio for the S&P 500 represents the inflation adjusted price of the S&P 500 divided by the rolling 10-year average of inflation adjusted earnings of the S&P 500. This ratio is also known as the Shiller P/E ratio, as it was made popular by Robert Shiller in his book Irrational Exuberance.
Although popularized by Dr. Shiller, similar ratios have been utilized by professional analysts and investors for decades. The similarity arising from dividing price by an average of periodic earnings. The difference typically coming from the actual time period utilized to average earnings. Differences aside, the ratio in its various forms is a useful tool to examine the relative valuation of a security or index across time while smoothing periodic earnings.
Great! Who cares?
Well, as of late, it seems that the answer to that question is uncertain. The CAPE is mentioned rather frequently across media outlets and among professionals, but it is increasingly being dismissed as a relic.
Any number of reasons can be recited for why this measure is no longer relevant. Some say that investors have simply excepted permanently higher multiples of price to earnings as the new normal. Others say that a ten-year average is too long or too short, or that the method of inflation is flawed.
To be sure, any of these criticisms might be accurate. It seems, however, that the primary reason that it is so frequently dismissed is that the CAPE has a rotten record of predicting imminent changes in market direction. In today’s world of up to the second dissemination of purportedly actionable investment advice, if it can’t predict what’s going to happen tomorrow, next month, or – gasp – even next year, then it is hard to see what use it has.
The fact is, the CAPE doesn’t wield much predictive power over the very near-term, and struggles over even two, three, and five-year periods. But where the CAPE starts to shine is over a ten-year period.
In particular, regressing observed forward 10 year compounding annual growth rates (CAGR) of the S&P 500 on the CAPE ratio depicts an interesting relationship with seemingly not insignificant explanatory power:
Source: Robert Shiller
The above chart examines data for the composite that ultimately became the S&P 500 dating back to 1941. This time frame provides for the inclusion of laws and regulations that have shaped modern markets.
With a coefficient of determination of .53, the relationship between the CAPE ratio and subsequent index performance is obviously less than perfect. However, it is also not difficult to determine that there is something going on here. As the CAPE ratio increases, subsequent 10 year compounding annual growth rates have tended to decline on average, historically.
Interestingly, if Dr. Shiller’s data is utilized back to 1900 to examine the same relationship, R2 decreases dramatically:
Source: Robert Shiller
Perhaps more interestingly, the extreme negative CAGR observations resulting from the years surrounding the Great Depression seem to have an outsized impact on explanatory power. But many of the severely negative CAGR observations that correspond with sub-30 readings of the CAPE were achieved just prior to, or on the way down from 1929 highs. In other words, the effects of overly rich valuation were felt long after prices and earnings realigned with historical averages.
This gets to the crux of why anyone is talking about the CAPE ratio at all right now. The CAPE ratio is presently hovering around 30. If the CAPE were at 17, it is doubtful that many would be asking what information could be extracted from it. That is, the CAPE is higher now than it has been in all but two other points in the last 100 plus years: just prior to the Wall Street Crash of 1929, and prior to the Dot Com Bust (although the Dot Com Boom would go on to lift the CAPE into the 40s). It is understandable that, due to the infrequency of such high readings, some might apply greater significance to readings far outside of the average.
With so few observations north of 30, and with the apparent decrease in explanatory power when the only other historical instance of ratios over 30 is included in the regression, it might be easy to dismiss the CAPE as irrelevant, particularly at elevated levels. Some might also draw the conclusion that, perhaps, multiples really have permanently expanded, considering the relatively mild knock-on effects of extremely high CAPE ratios leading up the Dot Com Bust compared with those just prior to the Great Depression.
So, does the CAPE ratio matter? It seems it would depend on what is trying to be understood from it.
Will it precisely predict anything? Probably not.
What it does do is provide us with an opportunity to understand where valuations are at, on a relative basis. With this information, we can begin to determine whether we can construct any historical parallels that might prove insightful.
Should investors be mindful that the CAPE ratio for the S&P 500 is approaching infrequently scaled heights? Probably.
The CAPE ratio, like so many other frequently ignored fundamental measures, should ultimately prove to demonstrate that when assets are more expensive on a relative basis, they tend to subsequently underperform on a relative basis. The current CAPE ratio of the S&P 500 indicates that, relative to historical readings, the price of the S&P 500 is expensive compared to the ten-year average of its inflation adjusted earnings. That’s about it.
So yes, we should care about the CAPE. But not necessarily any more than the practice of examining fundamental data and relative valuation in general. Numbers and history are impartial. If investors begin to discount the significance of either, they do so at their own peril…unless, of course, it’s different this time.