John Hussman: Market Peaks Are A Process

|
 |  Includes: DIA, IWM, QQQ, SPY
by: John Hussman

Excerpt from the Hussman Funds' Weekly Market Comment (6/2/14):

Click to enlarge

I should emphasize that the circled areas on the chart above aren’t chosen arbitrarily but reflect points where similar overvalued, overbought, overbullish extremes were observed. As I’ve noted in recent weeks (see The Journeys of Sisyphus and Exit Strategy), depending on how tightly we define this syndrome the 1972 and 1987 peaks can also be captured among the set of extremes that include 1929, 2000, 2007 and today. Remember that at a fine resolution, the full syndrome sometimes doesn’t precisely align with the final market high. Still, we remain convinced that any near term continuation we observe in this advance is likely to appear quite insignificant in the context of what the market loses over the completion of the cycle.

It’s fascinating how investors come to forget that markets move in cycles and not perpetual diagonal lines. As value investor Howard Marks wrote in The Most Important Thing, "Rule number one: most things will prove to be cyclical. Rule number two: some of the greatest opportunities for gain and loss come when other people forget rule number one." A normal, run-of-the mill cyclical bear market wipes out more than half of the preceding bull market advance. We should not be surprised at all to see the S&P 500 back at 2010 levels or below over the completion of the present cycle. From a valuation standpoint, we estimate that the S&P 500 Index would have to fall to the 1000 level to bring prospective 10-year nominal total returns toward their historical norm of about 10% annually. With the exception of the 2000-2002 bear market, valuations have typically been lower, and prospective returns higher, at cyclical troughs throughout recorded history (even in data prior to the 1960’s when interest rates were similarly depressed). Not that we need to forecast such an outcome, and certainly not that we would require anything near historical valuation norms to encourage a constructive stance, provided support from other factors. As always, the strongest prospective market return/risk profile is associated with a material retreat in valuations followed by an early improvement in broad measures of market internals.

A few side notes, in the continued belief that knowledge is power. It's quite accurate to observe that the recent half-cycle has been challenging for us, first because of our direct miss in the interim of our 2009-2010 stress-testing, and then because the resulting ensemble methods - though robust to Depression-era outcomes and having stronger performance in complete market cycles than any approach we've evaluated - did not fully capture certain bubble-sensitive features that were part of our pre-2008 methods. Those features relate mostly to what we called “trend uniformity” during the late-1990’s bubble. Repeated bouts of QE forced us to reintroduce similar bubble-sensitive features as an overlay. As I've often noted, nothing in the historical data encourages the attempt to speculate in a bubble once extremely overvalued, overbought, overbullish features emerge - and nothing in the historical record, even considering prior bubble periods, indicates we should speculate here. In any event, there’s no question or argument that my insistence on stress-testing against Depression-era data was a painful shot to the foot in this cycle, and is more painful because, at least to-date, it turned out to be unnecessary in hindsight.

...

In my view, investors should be thinking very seriously about the extent of potential market losses over the completion of the present market cycle. It is the wrong question to ask “where else am I going to put my money with short-term interest rates near zero?” The problem with that question is that it carries the implicit assumption that the expected return on stocks is even positive or adequate given the prospective risks. At present, the better question is “do I prefer a zero loss to the prospect of a 40-60% interim loss in a market that is strenuously overbought and overbullish, and has returned to valuations that are more than double reliable historical valuation norms?”

On Friday, our estimate of prospective 10-year S&P nominal total returns set a new low for this cycle, falling below 2.2% annually. This is worse than the level observed at the 2007 market peak, or at any point in history outside of the late-1990's market bubble. It's possible that investors could drive prospective returns even lower than they are now, and valuations even higher than they are now, as investors did during that bubble. Still, even that advance started to be punctuated by abrupt vertical declines or "air pockets" once overvalued, overbought, overbullish features emerged (recall for example the increasingly frequent and distinct peak-trough corrections of 10% or more in Oct 1997, Jul-Oct 1998, Jul-Oct 1999, Jan-Feb 2000, Mar-Apr 2000, and Sep-Oct 2000 even before more severe losses got underway). See Setting the Record Straight for a general review of current valuations and prospective returns.

If your answer is still to take your chances in stocks, my only hope is that you do it consciously, fully aware that you’ve decided to ignore or discount the lessons from a century of historical evidence. The simple fact is that somebody has to hold stocks at present levels, so there’s really no point in trying to convince others to embrace our concerns, or to answer every second-guess that presumes that “this time is different.” I’ll be quite happy if at least stocks are being held by those who understand the full narrative of the recent half-cycle, and have seen, considered, and discarded our work.