A Warning Regarding Broken Speculative Peaks
Frankly, I wonder whether any amount of arm-waving will incline investors to actually examine their risk exposures here, much less consider the prospect of a 40%+ decline in the S&P 500 Index that would be required simply to bring stocks to historically run-of-the-mill valuations. But at a time when our estimates of prospective risk are surging, I would be remiss not to observe that fact.
At present, we have what might best be characterized as a broken speculative peak, in that market internals (particularly interest-sensitive groups), breadth and leadership have broken down uniformly following an extreme overvalued, overbought, overbullish syndrome. If you recall, the market also recovered to new highs in October 2007, weeks after the initial, decisive break in market internals at that time. Presently, we’re looking at the same set of circumstances. On some event related to tapering or the Fed Chair nomination, we may even see another push higher. It isn't simply short-term risk, but deep cyclical risk that is of concern.
My main goal here is to encourage investors to look carefully at their investment positions, before they lose the chance to alter them advantageously. As I noted in the October 15, 2007 market commentWarning – Examine All Risk Exposures:
“Whatever market exposure investors accept today ought to be the same market exposure that investors are committed to maintain for the duration of a bear market, without abandoning their investment plan. Investors with no plan to own stocks through a market decline, holding them only in the hope of selling at market highs, may discover in hindsight that these were them.”
In fact, they were. I get it. Nobody cares. This time is different. The Fed will not allow – allow – stocks to go lower. There’s no question that a few binary events – mainly the likely “tapering” of quantitative easing, and the choice of a new Fed Chairman – creates significant uncertainty about the short term. My concerns are more extended, and are specifically related to the likelihood that the present market cycle will complete in a way – as market cycles have historically – that wipes out more than half of the gains of the preceding bull market advance. My impression is that the losses even in a not-so-terrible completion of the present cycle may come closer to three-quarters of those gains.
Collective belief can create its own reality, and at least for the past few years, collective belief is that Fed actions simply make stocks go up, and so they have. The problem is that this outcome is based almost wholly on perception and confidence bordering on superstition – not on any analytical or mechanistic link that closely relates the quantity of monetary base created by the Fed to the equity prices (despite the correlation-presumed-to-be-causation between the two when one measures precisely from the 2009 market low). Some of the deepest market losses in history have occurred in environments of aggressive Fed easing (seeFollowing the Fed to 50% Flops). Markets move in cycles, bear markets wipe out more than half of the preceding bull market gains, and this one is likely to be no different.
There is certainly a strong link between the monetary base and short-term interest rates, because zero-interest bank reserves (and bank deposits that stand as evidence of them) are near-perfect portfolio substitutes for short-term Treasury securities. There is a weaker link with respect to long-term interest rates, as we're seeing at present, not only because of the much longer duration involved, but also because of liquidity features (for example, much of the weakness in the U.S. Treasury market here appears to be a reflection of foreign liquidation, particularly by Chinese holders facing a lending crunch). Unfortunately, the link between the monetary base and equity prices is virtually non-existent. The implied “duration” of equities – what mathematically works out to be closely related to the price/dividend ratio – is much like that of a 50-year zero-coupon bond. Equities also feature uncertainty about the payment stream and significant cyclical fluctuations in risk premiums. Investors will find all of this out later. My hope is those who think carefully about data, history, economics, risk, valuation, and investment discipline will consider it now.
It's worth emphasizing that with short-term interest rates pegged at zero, and with trillions of idle reserves in the banking system, further quantitative easing does not relieve any constraint in the economy that is actually binding at present, and decades of economic data (as well as theory that has won Nobel Prizes) demonstrate that there is virtually no impact from volatile assets (like stock prices) to consumption. Granted, there has been just enough of a “pure confidence” effect in recent years to promote a few bumps to consumption and economic activity for a quarter or two at a time, at increasing risk to longer-term financial stability, but even this economic effect has been weak and transitory. Indeed, even aresearch notefrom the San Francisco Fed last week estimated that QE2 added just 0.13% to real GDP – an effect that faded after two years. The note concluded that “asset purchase programs like QE2 appear to have, at best, moderate effects on economic growth and inflation. Moreover, the magnitude of LSAP [large-scale asset purchase] effects depends greatly on expectations for interest rate policy, but those effects are weaker and more uncertain than conventional interest rate policy.”
While quantitative easing is constantly discussed in reference to the unemployment rate, there is also no material or demonstrable link between the two. Indeed, a significant part of the decline in the unemployment rate in recent years is a reflection of stagnant growth in the labor force. Normally the labor force grows by about 150,000 workers a month, requiring the same amount of job creation just to hold the unemployment rate steady. Since 2009 the labor force has grown by only about 20,000 workers a month, allowing the unemployment rate to drop even with tepid job creation. Moreover, a significant portion of recent job growth represents a major loss in full-time employment that is being replaced with the creation of part-time jobs. Finally, even the relatively good numbers on new unemployment claims are lagging indicators, and are, as a percentage of the labor force, at exactly the same level as they were in October 2000 and January 2008, as new recessions were unfolding. This doesn’t mean that a recession must unfold in the present instance, but the depressed level of new claims certainly provides no basis to exclude that outcome.
Here's what I propose. Anytime you hear pronouncements about QE from FOMC members, just insert some complete non-sequitur. It's likely that QE has no greater cause-and-effect relationship with the economy, and the headline will not be any worse for it. For example:
Fed Governor Bullard Cautions Against Slowing the Pace of Spitting Watermelon Seeds Into a Can, Citing Continued Economic Risks
Fed Vice-Chair Janet Yellen Pledges to Keep Knitting Enormous Wool Scarf in Order to Provide Support for U.S. Job Creation
Fed Chairman Ben Bernanke Pledges to Keep Playing “Highland Laddie” on the Bagpipes Until Unemployment Rate Drops to 6.5%.
A Broken Speculative Peak
While market internals may recover and thereby lessen our more extreme concerns, at least for some amount of time, we presently observe what might best be called a “broken speculative peak” – a strenuously overvalued, overbought, overbullish, rising yield syndrome followed by a breakdown in market internals. Our estimates of potential market losses are surging, and our present return/risk estimates easily fall into the worst 1% of historical data. We associate these instances with average losses in the S&P 500 approaching 50% at an annual rate, though a scattered handful of similarly weak estimates since April 2012 have been devoid of any negative outcomes at all.
It's certainly been frustrating to forego recent gains as a result of our reluctance to speculate in overvalued, overbought, overbullish conditions. These conditions have persisted far longer than normal, without the typical losses that have historically followed, but the lesson that one draws from that matters. I don’t believe that all of history has become irrelevant; only that speculators are enjoying the temporary benefit of deferred consequences. In my view, the recent advance has rested on excessive confidence in monetary linkages that have no theoretical or historical basis, and I believe that there’s an anvil floating on this bubble.
Prior to 2012, the last time we observed return/risk estimates similar to the present was the week of July 27, 2007 (seeMarket Internals Go Negative). The instances prior to that fell between September and November of 2000 (see in particular theSeptember 17, 2000entry of our archived weekly comments). Back then, I used the term “Crash Warning” to describe such conditions. I think “broken speculative peak” is sufficient to describe my concern. As one might guess from other charts I’ve presented in recent months, the prior instance was in August 1987 (where part of the internal breakdown was in interest-sensitive securities as it is now). The instance before that was in January 1973, though I should observe that it was preceded by one less troublesome instance in April 1972, with several months of further market gains before the market peaked and then lost half of its value.
Numerous prior comments present a variety of criteria that help to identify overvalued, overbought, overbullish syndromes (seeCapitulation Everywhere). Meanwhile, with breadth weakening, leadership reversing to the downside, sector internals failing to confirm the major averages, and interest-sensitive sectors grinding lower, market action is signaling a shift to greater risk aversion. This is a pointed threat once risk premiums have been driven toward zero. A market plunge is usually really nothing more than a spike in risk premiums from previously inadequate levels.
As a side note, while we don’t use Hindenburg signals in our own work, it’s notable that when they occur with all the “confirming” bells on (deteriorating breadth, market above 10-week, multiple signals clustering), they’re really just another way of identifying an increase in the dispersion of market internals in the context of a recently rising market. Those “confirmed” signals (which also occurred just before the 1987 crash) are much less frequent and more troublesome than the numerous raw signals that people like to trot out to dispense with this concern (see2009 vs. 2013).
It seems to be an element of common knowledge that the Fed has made all downside risk obsolete and simply “will not allow” the market to decline. I understand this argument, despite my strong disagreement with its premise, based on decades of prior evidence indicating no material relationship between the monetary base and equity values. Even including the period since 2009, we estimate the correlation between the two to be negative at horizons shorter than 6 years, regardless of whether one examines levels or percentage changes. Quantitative easing does not operate on an economic mechanism supported by empirical relationships or sound theory (indeed, both data and theory refute it). Instead, QE is a superstition that has been effective largely because of the novelty of the concept and the misattribution of the market’s recovery from the 2009 low to monetary policy. On a careful analysis of that period, the main catalyst for the market’s recovery was the elimination of “mark-to-market” rules by the Financial Accounting Standards Board – removing the concern about insolvency of major banks (even if they were technically insolvent) at a point where equities were briefly undervalued.
Investors have put their faith not in a fact but in a concept. As my friend John Mauldin puts it, “Faith in central banks today is equivalent to faith in the word dot-com in 1999 or faith in the eternal rise of housing prices in 2006.” No doubt, Warren Buffett was right that “a pack of lemmings looks like a group of individualists compared to Wall Street once it gets a concept in its teeth.” But QE is also a concept that is rather fully played out, especially with a likely tapering ahead. Notably, the shift toward tapering is not driven by substantial economic improvement or victory of the policy, but rather by an increasing recognition within the Fed itself that its actions are creating dangerous financial distortions.
In any event, I would be remiss in not pointing out where our return/risk estimates stand at present. Keep in mind that the most relevant horizon for us is the completion of the present market cycle, not just the next few months. There are certainly factors such as the Fed’s decision on tapering and the decision regarding a new Fed Chairman that may significantly affect near-term market outcomes. So these return/risk estimates here should not be taken as a forecast, only an indication of the balance between prospective return and risk here. I believe it is critical to examine all risk exposures, with a focus on setting those exposures at levels that can be reasonably maintained through the course of potentially significant market losses.
Past performance does not ensure future results, and there is no assurance that the Hussman Funds will achieve their investment objectives. An investor's shares, when redeemed, may be worth more or less than their original cost. Current performance may be higher or lower than the performance quoted above. More current performance data through the most recent month-end is available at www.hussmanfunds.com. Investors should consider the investment objectives, risks, and charges and expenses of the Funds carefully before investing. For this and other information, please obtain a Prospectus and read it carefully.
© Hussman Funds