“I am definitely concerned. When was [the cyclically adjusted P/E ratio or CAPE] higher than it is now? I can tell you: 1929, 2000 and 2007. Very low interest rates help to explain the high CAPE. That doesn’t mean that the high CAPE isn’t a forecast of bad performance. When I look at interest rates in a forecasting regression with the CAPE, I don’t get much additional benefit from looking at interest rates… We don’t know what it’s going to do. There could be a massive crash, like we saw in 2000 and 2007, the last two times it looked like this. But I don’t know. I think, realistically, stocks should be in someone’s portfolio. Maybe lighten up… One thing though, I don’t know how many people look at plots of the market. If you just look at a plot of one of the major averages in the U.S., you’ll see what look like three peaks – 2000, 2007 and now – it just looks to me like a peak. I’m not saying it is. I would think that there are people thinking – way – it’s gone way up since 2009. It’s likely to turn down again, just like it did the last two times.”
Professor Robert Shiller, June 25, 2014, The Daily Ticker
.....
On a historical basis, the CAPE of over 26 is already quite enough to expect more than a decade of negative real total returns for the S&P 500. Aside from the crashes that followed the 1929, 2000 and 2007 peaks, a very long period of negative real returns also followed the other historical peak in the CAPE near 24 in the mid-1960’s. As noted above, one adjustment to the CAPE that significantly improves its relationship with actual subsequent market returns – as it does for numerous other measures – is to correct for the implied profit margin embedded into the multiple. This is true even though the denominator of the CAPE is based on 10-year averaging. At present, the margin embedded in the Shiller CAPE is more than 20% above the historical average. Adjusting for that embedded profit margin – which, again, produces a historically more reliable indication of actual subsequent S&P 500 total returns – the Shiller CAPE would presently be over 32. That level might make even Professor Shiller question whether stocks should be a material component of portfolios (at least for investors with horizons much shorter than the 50-year average duration of S&P 500 stocks). In any event, even the phrase “lighten up” is problematic for the market if more than a few investors heed that advice.
The ratio of non-financial equity market capitalization to GDP (which has maintained a tight correlation with subsequent 10-year S&P 500 total returns even in recent times) is now about 134%, compared with a pre-bubble norm of 55%. The median price/revenue ratio S&P 500 components easily exceeds, and the average rivals, the levels observed at the 2000 peak. All of this suggests that investors may not appreciate the extent of present overvaluation, lulled once again by the assumption that cyclically-elevated earnings are permanent. Benjamin Graham warned long ago that this assumption is probably the chief source of losses to investors: “The purchasers view the good current earnings as equivalent to ‘earning power’ and assume that prosperity is equivalent to safety.”