Should we be worried about CAPE, the popular Shiller Cyclically Adjusted Price Earnings ratio? Experts tell us it is trading at exceedingly high levels and, the experts also say, those high levels portend poor stock returns, according to CAPE. We’ve thought about that and have come to the following conclusion:
First, CAPE is a notoriously bad prognosticator. Second, CAPE has been high for the past 20 years, and has moved steadily higher since 2010. Anyone complaining about stock market returns this decade? Finally, what about the “E” part of CAPE? What do earnings imply about current levels?
For those unaware, CAPE is a price earnings ratio based on average inflation-adjusted earnings from the previous 10 years, sometimes called the Shiller PE Ratio, or PE 10. As you can see from the chart below, CAPE valuations are at historically high CAPE levels relative to the average.
But are CAPE valuations truly stretched? This is a ten-year average calculation. In the last ten years, adjusted SPY earnings included two years (2008-2009) that could easily be considered trough earnings based on recent history ($17.31 and $51.71 per share, respectively versus $95.00 for the 2010-2016 average). When the 2008-2009 earnings fall out of CAPE starting next year, and IF those earnings are replaced by anything remotely close to recent history, is today’s CAPE multiple truly rich?
When we project CAPE by replacing trough earnings, we consistently see about a 15% decline in the multiple which puts CAPE on an adjusted basis today between 24-25, below the 27.12 average seen between 1996 and 2016. Even if you kick out the goofy 1999-2001 period, CAPE’s average historical multiple drops only to 24 for the twenty-year period, again in line with our adjusted CAPE today.
What about earnings? How reasonable is our assumption that 2018-2019 earnings will stay within recent historical averages? Google and Facebook account for about 8% of the S&P’s earnings today. Facebook wasn’t even part of the S&P prior to 2012. Ten years ago, Apple was a company in the index, not THE company. Would you rather own 2017’s earnings quality or 2007’s?
We’ve said it before and it’s worth repeating: from 2001 through 2005, the S&P 500’s p/e multiple DECLINED from 29.55 to 16.33, and the index traded up 41% mostly because the “e” in the p/e ratio materialized. Over the next two years, it seems entirely plausible that earnings driving CAPE ratios can hold. You could identify a handful of reasons for the market to trade down from here (China, North Korea, interest rates), but it is hard to argue that an over-valued CAPE is one of them.