Hussman: Topping Patterns And The Proper Cause For Optimism

by: John Hussman

Excerpt from the Hussman Funds' Weekly Market Comment 2/17/14:

Following a moderate decline from its recent highs, the market experienced a “reflex” advance last week. As I noted in the February 3 comment, “Even the shallow 3% retreat from the market’s all-time highs may be enough to prompt a reflexive ‘buy-the-dip’ response in the context of extreme bullish sentiment here, as the S&P 500 bounced off of a widely monitored and steeply ascending trendline last week that connects several short-term market lows over the past year. Regardless, the potential for short-term gains is overwhelmed by the risk of deep cyclical and secular losses. We presently estimate prospective 10-year S&P 500 nominal total returns averaging just 2.7% annually, with negative expected total returns on every horizon shorter than 7 years.”

Needless to say, our concerns are little changed by the last week’s advance, and with this low-volume reflex rally in place, we may observe a much deeper and uncorrected loss if the prior resolutions of severely overvalued, overbought, overbullish, rising-yield conditions are an indication. The expectation of impending market losses should certainly be tempered by the fact that similarly extreme conditions in February and May 2013 were largely uneventful, and were followed by further gains. Still, it’s important to recognize that extreme syndromes of overvalued, overbought, overbullish, rising-yield conditions have previously created risk and instability over a period longer than a few weeks or even months.

Likewise, in the context of log-periodic bubbles – as we’ve observed in U.S. equities since 2010 (see The Diva is Already Singing) – the “critical point” or “finite time singularity” is not a crash date, but the inflection point from self-reinforcing speculation to fragile instability. Didier Sornette observed this more than a decade ago in Why Stock Markets Crash. Our best estimate of that inflection point remains about January 13. It’s also worth remembering that the “catalysts” associated with sharp market losses have often been fully recognized only after the fact, if at all. As Sornette emphasized, “The collapse is fundamentally due to the unstable position; the instantaneous cause of the crash is secondary.”


A few months ago, Bill Hester examined a century of price data, and observed that the non-overlapping periods most closely correlated with the past 5-year trajectory of the S&P 500 were the advances preceding the 2007, 2000, 1987, 1937 and 1929 market peaks. This shouldn’t be a surprise – the past 5-years have largely resembled a diagonal line, as did the advances to those extraordinary market peaks. Diagonal lines always have a nearly perfect correlation even if there is some amount of variation along the way.

Again, we would dismiss historical analogs like this if the recent market peak did not feature the “full catastrophe” of textbook speculative features – particularly the same syndrome of extreme overvalued, overbought, overbullish, rising-yield conditions observed (prior to the past year) only at major market peaks in 2007, 2000, 1987, 1972, and 1929. The main temptation to ignore this concern is that similarly extreme conditions emerged in both February and May 2013 without consequence. Less extreme variants of this syndrome have also emerged periodically in the past few years (these variants also capture 1937 and a few other bull market peaks, as well as the April 2011 peak after which the market briefly retreated by nearly 20%). Overall, my view continues to be that the consequences of the more recent instances have not been avoided, but merely deferred – and those consequences will be worse for it.

To offer some perspective of how major peaks have typically evolved, the following charts present the Dow Jones Industrial Average in the final advances toward, and the few weeks after, what turned out in hindsight to be major stock market peaks. For reference, let’s examine the recent market peak. Notice several features:

  1. A series of moderately spaced peaks forming a broad sideways “consolidation” over several months;

  2. A breakout from that consolidation, leading to a steep and only briefly corrected speculative blowoff into the market’s peak;

  3. A steep initial selloff from the market peak, and finally;

  4. A “reflex” rally (classically on low volume – indicative of a short-squeeze with sellers backing off) that retraces much of the initial selloff.


From the standpoint of investor psychology, it seems understandable that the speculative enthusiasm and short-covering that contributes to bull market tops is fueled when the market “breaks out” after a period of consolidation (and what Lindsay called a “separating decline”). The blowoffs that followed – both recently and in the examples above – were accompanied by clear overvaluation on reliable measures, and also featured clearly defined overbought conditions and lopsided bullish sentiment.

I’ll say this yet again, because it’s a crucial point – we would have zero interest in historical analogs like this if the recent market peak was not also accompanied by an extreme overvalued, overbought, overbullish, rising-yield syndrome that – prior to the past year - has emerged only at major market peaks. As with the recent log-periodic bubble, the pattern itself should be simply be treated like an interesting experiment. Lindsay’s and Sornette’s observations are largely consistent in that Lindsay’s pattern is basically a slightly irregular refinement of the very end of a Sornette bubble. The “three peaks” of Lindsay’s pattern correspond to log-periodic oscillations, and the left side of the “domed house” is essentially a final blowoff and short squeeze featuring much shallower fluctuations approaching the high.

Without tracing through over a century of price data for every instance (including "setups" that did not follow through), we can't say categorically that this pattern is always followed by market losses. What one can say is that when Lindsay’s pattern has been accompanied by rich valuations (say, a cyclically adjusted P/E over 18), the market has typically been at the cusp of a significant retreat. Those periods of rich valuations are easily identified, and the corresponding blowoff patterns emerge in 1901-02, 1906, 1929, 1937, 1961, 1966, 1969, 1972-73, 1987, 1998, 2000, and today. The 2007 peak had less fidelity to Lindsay’s pattern because the reflex rally in October slightly exceeded the July peak of that year. Of course, lopsided bullish sentiment – as we observed in December and January – further narrows the set and worsens the average outcome, as do razor-thin risk premiums on corporate debt, soaring margin debt, and severely overbought conditions following a largely-uncorrected multi-year market advance. It’s not any single pattern that concerns us here, but rather the entire syndrome of classic speculative features that accompanied the recent peak – in the words of Zorba the Greek, “the full catastrophe.”


To the extent that we are concerned and defensive about the prospect for steep equity market losses over the completion of this market cycle, we are also encouraged and optimistic about the prospect for strong investment opportunities that we expect to emerge as a result. Similar optimism for improved investment opportunities was certainly vindicated following periods that shared features like the present – 1929, 1972-73, 1987, 2000 and 2007, not to mention many less extreme instances of overvalued, overbought, overbullish conditions.

We currently estimate prospective nominal total returns for the S&P 500 of just 2.4% annually over the coming decade. With the 10-year Treasury yield at 2.8% and short-term yields expected to remain depressed for years, we expect the 10-year return from an equally-weighted portfolio of stocks, bonds, and cash to average only about 2% annually from current prices. Our optimism that better opportunities will emerge over the completion of the present market cycle is unbowed. That optimism demands that investors refuse to lock-in the prospect of dismal long-term returns, or surrender in the face of a deafening speculative refrain that has lured others into the abyss throughout history.