I look forward to fund manager John Hussman’s blog posts each week (hussmanfunds.com). John is an experienced market researcher, but in some of his articles last year, he reflected on his overvaluation concerns, which were similar to those expressed by Jeremy Grantham’s team (read GMO’s November 2013 newsletter here regarding asset overvaluation in 2013).

Hussman’s February 17 post contains some dramatic updates to his long-term market thesis that I will review below, and then comment on further. I recommend reading John’s full article: Topping Patterns and the Genuine Cause for Optimism. (First 6 exhibits that follow are from Hussman’s Feb. 17 article.)

First, John reproduces a graph he’s been updating frequently, using a methodology from Didier Sornette’s Why Stock Markets Crash:

The graph shows the S&P 500 converging on what Sornette terms a “log-periodic bubble,” which is essentially caused by investors “buying the dips” reflexively every time the stock market declines in value. “Buy the dips” results in corrections that are shorter and shallower each time. This pattern has prevailed immediately before each of the most serious historical stock market bubbles burst (1929, 1973 and 2000) — the fact that it’s repeating again in early 2014 is mildly nerve-wracking.

Next Hussman explains a technical pattern known as “3 peaks and a domed house,” attributed to George Lindsay’s Art of Technical Analysis, which was edited and updated by George Carlson in 2011. The pattern is shown below — Hussman asserts that we are most likely at point 27 in the pattern — the one just before a market crashes:

The similarities with 1929′s point 27 (S&P 500):

1973′s point 27 (S&P 500):

2000′s point 27 (Dow Jones Industrial Average):

and 2014′s point 27 (S&P 500) are impossible to miss:

Considered along with the Sornette log-periodic bubble analysis, I would say this additional perspective takes us into moderately nerve-wracking territory — especially after a 5-year bull market run that’s raised equity values 135% above the March 2009 lows.

The Shiller P/E10 (or “CAPE,” for cyclically-adjusted P/E ratio), is one of the most popular metrics analysts cite as signaling overvaluation. Some analysts have questioned if the P/E10 is as relevant as it once was, however. (This article by Chris Turner via Advisor Perspectives explains how analysts use the P/E10 to estimate the fair value of the S&P 500.)

I have written several articles on how the P/E10 forecasts long-term stock returns (including The Journal of Portfolio Management, June 2007). Below I’ve updated the exhibits from a 2012 article I published in Problems and Perspectives in Management, entitled Could US Stocks be Fairly-Valued Under the “New Normal” Paradigm? The graph shows that, as of year-end 2013, US stocks were overvalued based on historical values of the P/E10, but only by as much as 8.6% if investors are benchmarking “normal” to the post-1982 period (the start of the last secular bull market):

On the other hand, if the market’s memory is shorter, and investors benchmark fair value to a P/E multiple based on only a 3-year moving average, and use an exponential moving average of past earnings that places more weight on recent observations (which I call the P/EXP3):

US stocks could have been undervalued as of year-end 2013 by as much as 6.8%, if, once again, investors are benchmarking “normal” to the post-1982 period.

Conclusion: Both a Sornette log-period bubble and a Lindsay “3 peaks and a domed house” formation are evident in US equity values in late 2013 through early 2014. These patterns have preceded the bursting of stock market bubbles in 1929, 1973 and 2000. Additionally, US stocks are “overvalued” based on a traditional P/E10 CAPE, but are within the high range of normal valuation if the market no longer benchmarks to a 10-year moving average of earnings and/or a 130-year history of relative valuation.

Consistent with a large body of research (Shiller’s Irrational Exuberance included), it’s extremely difficult — and perhaps impossible — to see a bubble through the windshield. Only after the bubble bursts, and we can study it in the rear-view mirror, can we rationally deconstruct how we got there and why we couldn’t back away from the precipice with sufficient caution not to burst the bubble. I see current circumstances as another of these instances. Investor enthusiasm for “buying the dips” makes it difficult to forecast the imminent bursting of a bubble, unless there are further negative developments in the unfolding global credit market downturn. I would recommend assuming a strongly defensive position only if global volatility increases significantly. This week’s news flow bears close watching.