The CAPE ratio, or Cyclically Adjusted Price Earnings ratio, is making headlines again. The stock market valuation indicator developed by Nobel laureate Robert Shiller is giving many investors headaches. But, what does the CAPE ratio really tell us about future equity market returns?
The Shiller PE, as the CAPE ratio is often affectionately called after its creator, is the price-earnings ratio based on the earnings of the last 10 years (cyclically adjusted). The idea behind the 10-year period is that recessions and/or accounting ‘tricks’ don’t nullify the usefulness of this valuation measure, as can be the case with more traditional valuation measure like the standard PE ratio.
What makes the CAPE ratio attractive is that it helps forecast long-term equity market returns. This is shown in the graph below. For every calendar month since 1947, I have linked the CAPE ratio, as published by Shiller, with the average annual real return on the S&P 500 Index in the 10 years thereafter. Then, the calendar months are ranked according to their CAPE ratio and split into five quintiles. The first quintile represents the 20% of all months with the lowest CAPE ratios and their subsequent return, while the fifth quintile holds the return data of the 20% most expensive months.
The graph reveals that a higher valuation is accompanied by lower future return. In the ten years after the calendar months with the lowest CAPE ratios the real return on the S&P 500 Index averaged a healthy 6.7%. However, ten years after the months with the highest CAPE levels, you hardly would have made any return at all. Corrected for inflation, just 0.2% return remained. The graph also shows that the more extreme the Shiller PE gets, the more extreme the subsequent return. In the few instances that the CAPE fell below 7, you realized, on average, almost 10% real return per year. But in months the CAPE topped 40, you lost, again on average, more than 5% per year!
The usefulness of the CAPE ratio concerning forecasting returns is not limited to the US. Norbert Keimling from Star Capital calculated the CAPE ratios for all countries included in the MSCI World Index, with historical data going back at least thirty years. The scatter plot that results (shown below) confirms the relationship: higher valuation implies lower future return.
So much for the good news. Keimling’s graph also demonstrates that the CAPE ratio isn’t the holy grail of return forecasting. It explains roughly half of the variation of future returns. Impressive, but at the same time it confirms that valuation is not the sole factor explaining future equity returns. Obviously, economics, sentiment and politics also play part. And let’s not forget, the sometimes inimitable, behavior of investors.
But there are other, more technical, factors that limit the forecasting power of the CAPE ratio as well. For example, take the stock market of Greece. What do earnings realized in the last 10-year really say about the Greek stock market now? A large part of the companies that generated these earnings has vanished as a result of the Greek debt crisis. This leads to the notion that the forecasting skills of the CAPE ratio is reduced whenever there are structural changes in the composition or structure of the underlying stock market.
Structural changes in other areas constrain the predictive power of the Shiller PE, too. For example, when investors structurally demand a lower risk premium on equities. Take Japan for instance, where, for years on end, the Bank of Japan has been aggressively buying Japanese equities. The BoJ is directly supporting equity investors with its relentless demand. Theoretically, this ‘BoJ-put’ should push up the valuation of Japanese equity, without lowering returns going forward. And what about the historically low bond yields around the globe? These could be matched by a structural higher CAPE, as equities are attractively valued against the incredibly expensive bonds.
Jeremy Siegel, author of ‘Stocks for the Long Run’, offers yet another CAPE shortcoming. In determining the CAPE, Shiller uses the reported earnings of the S&P 500 companies. However, Siegel states that current accounting rules push down reported earnings much further during recessions than they did before. Hence, the current CAPE is artificially too high because of the massive earnings collapse following the Great Recession of 2009. To circumvent this issue, Siegel comes up with an alternative earnings definition, which, not surprisingly, lowers the CAPE ratio. This last criticism of the CAPE ratio is less convincing, though, as Siegel uses the NIPA earnings, or all profits made throughout the US economy, including non-listed companies. This is not necessarily a fair representation of the earnings realized by ‘just’ the S&P 500 companies.
Now, for the really scary part. The latest Shiller PE equals 29.2, which is the highest level since 2002. Obviously, that’s not great news. Based on history, we have to assume that the real return on the S&P 500 Index won’t be much above zero in the coming decade (see first chart again). Perhaps Keimling’s scatter plot offers some solace as it points out that the actual returns can deviate quite a bit from the average. There have been some occasions where the CAPE topped 30 while the subsequent return still topped 10%.
I should not bet on this happening for the S&P 500 Index, however. For most part these results are linked to countries where structural changes in the composition of the equity market took place. This argument is not valid concerning the S&P 500 Index. In addition, historical return data for the S&P 500 Index don’t provide much room for optimism either. In just one out of the 23 months, in which the US CAPE ratio stood between 27 and 31, hence comparable to the current CAPE level, did the subsequent average 10-year real return manage to top 4%. Contrary, the real return was negative in eight of those 23 months. And what if Siegel’s earnings definition is used to push down the CAPE ratio? Even then historical data suggest your return probably won’t top 2% in the coming ten years. Therefore, it seems fair to conclude that a high valuation of US stocks, as measured by the Shiller PE, does imply low returns going forward. Fortunately, the CAPE ratio for other regions is significantly lower. In fact, in Europe, Japan and emerging countries valuations are below average. There is still return to be made, but just not on US equities.
Norbert Keimling, StarCapitalAG, 2016, Predicting Stock Market Returns Using the Shiller PE.
Jeremy. J. Siegel, 2016, The Shiller CAPE Ratio: A New Look.