John P. Hussman, Ph.D.
As of last week, the market continued to reflect a set of conditions that have characterized a wicked subset of historical instances, comprising a Who’s Who of Awful Times to Invest . Over the weekend, Randall Forsyth of Barron’s ran a nice piece that reviewed our case (the chart in Barrons has a problem with the date axis, but the original chart is in last week’s comment Warning: A New Who’s Who of Awful Times to Invest). It’s interesting to me that among the predictable objections to that piece by bullish readers (mostly related to our flat post-2009 performance, but overlooking the 2000-2009 record), none addressed the simple fact that the prior instances of this syndrome invariably turned out badly. It seems to me that before entirely disregarding evidence that is as rare as it is ominous, you have to ask yourself one question. Do I feel lucky?
From our perspective, accepting stock market risk is not presently a venture that is priced to achieve reasonable investment returns (we estimate a likely 4.3% annual total return for the S&P 500 over the coming decade, and a great deal of volatility in achieving that return). Nor is market risk attractive on a speculative basis, given present overbought conditions, overbullish sentiment, and growing set of hostile syndromes (what we call Aunt Minnies) that have historically been associated with negative return/risk tradeoffs. Then again, what keeps slot machines spinning all around the world is the hope – despite the predictably and reliably negative average return/risk tradeoff – that this time will be different, and this spin will work out. So you have to ask yourself one question. Do I feel lucky?
Investors Intelligence notes that corporate insiders are now selling shares at levels associated with « near panic action. » Since corporate insiders typically receive stock as part of their compensation, it is normal for insiders to sell about 2 shares on the open market for every share they purchase outright. Recently, however, insider sales have been running at a pace of more than 8-to-1. The dollar amounts are even more lopsided, as Trim Tabs reports a recent pace of $13 of insider sales for every $1 of purchases. Indeed, some of the weekly spikes have been to levels that are associated almost exclusively with intermediate market peaks, the most recent being the run-up to the 2007 market peak, the early 2010 peak, and the 2011 peak, all of which resulted in significant intermediate corrections or worse. Of course, it’s sometimes the case that insiders are early, and therefore miss part of the tail of a market advance. So it might be worth ignoring the heavy pace of insider selling for a little while. But you have to ask yourself one question. Do I feel lucky?
As disciplined investors who align ourselves with the average return/risk profile that is associated with prevailing market conditions, we don’t believe that this is a good time to take significant market risk in hopes of getting lucky. On an objective basis, we identify present conditions among the lowest 1.5% of historical periods in terms of overall return/risk profile. Maybe investors will get lucky, but the odds are still unfavorable.
It’s likely that for some investors, our defensiveness since 2009 bleeds into a general inclination to take our concerns about risk with a grain of salt. On that subject, it’s important to recognize that our defensiveness in 2009 did not result from unfavorable valuations or hostile indicator syndromes, but from the inability to distinguish prevailing conditions at the time from much of what was observed during the Depression-era. In response to the credit crisis, and what I continue to view as a misguided « kick-the-can » policy response, I insisted that our methods should perform well with reasonable drawdowns not only in post-war data, but also in Depression-era data (when for example, stocks lost two thirds of their value even after they were priced to achieve 10-year total returns in excess of 10% annually).
The resulting ensemble methods allow us to make distinctions that we were not able to make in 2009, but that period of stress-testing also left us with a « miss » (2009-early 2010) when the same indicators and methods that are so hostile today would have been much more favorable toward investment risk. One could ignore that fact, and use our miss in 2009 as a reason to ignore demonstrably hostile evidence today. But one would also have to overlook the fact that the narrow syndrome of conditions we observe today mirrors what we observed at the 2000 peak and the 2007 peak, and very few times in-between (including the 2010 peak and the runup to the 2011 peak – see last week’s comment for a chart). Notably, whatever market returns we missed by being defensive too early in those instances were wiped out in short order anyway during the subsequent declines. Yes, stocks might move even higher before the present bull-bear cycle moves to completion. But you have to ask yourself one question. Do I feel lucky?
A note on extracting economic signals
While investors and the economic consensus has largely abandoned any concern about a fresh economic downturn, we remain uncomfortable with the divergence between reliable leading measures – which are still actually deteriorating – and more upbeat coincident/lagging measures on which public optimism appears to be based.
Much of our research effort in recent weeks has been focused on developing a deeper understanding of this divergence. The historical evidence clearly indicates that such divergences are settled in favor of the leading indicators (see last week’s economic discussion in Warning: A New Who’s Who of Awful Times to Invest ), but since our views are so out of sync with the broad consensus, the issue demands as much additional investigation and research as we can amass. It bears repeating we neither desire a recession, nor have any interest in « pounding the table » about it. We would rather have the data convincingly shift to a condition that eases our recession concerns. But if we’re likely to have a recession, we don’t want to be surprised or lulled into complacency by improvement in what are largely lagging indicators.
Probably the best, if slightly technical, way to understand our reluctance to discard recession concerns is to think about observed economic data as being driven by a series of unobserved « true » states of the economy. For example, suppose that the true underlying state of the economy can be summarized by a single positive or negative number each month, call it X(t), and that each economic indicator we can actually observe is driven by a series of those current and past monthly economic states. So for example, some variable that we can observe might be written as a series of unobserved economic states:
Y(t) = a0*X(t) + a1*X(t-1) + a2*X(t-2) + …. + a « shock » from truly new developments and random noise.
[Geek’s note – this is a version of an unobserved components model, of the kind used in nearly every modern signal-processing application. For example, when you go to the hospital to get a CT scan, the picture you see is not actually « taken » directly. Instead, the machine shoots X-rays at you from every possible angle, and then uses all of that sensor data to compute an image that could never have been obtained directly. That’s why it’s called computed tomography (CT). If signal processing methods weren’t applied, every image of an internal organ would be obstructed by artifacts from bone, cartilage and other organs. Similar signal processing methods are used to combine data from multiple sensors in order to navigate anti-ballistic missiles and other objects that can’t be directly observed (and of course, to identify meaningful genetic signals in autism data)].
Leading indicators essentially place weight on the unobserved true state a few months into the future (allowing that state to be estimated today based on those observed indicators), while coincident and lagging indicators load on previous components. To see what this looks like, the following chart presents the load factors we estimate for dozens of widely followed economic variables, including one-month, 6-month and year-over-year employment gains, new unemployment claims, real consumption growth, ISM data, consumer confidence, quarterly and year-over-year GDP growth, stock returns, credit spreads, OECD leading indices, and a score of other measures. Notice that most of the variables load not on the first (most leading) economic state variable, but instead on the fourth or fifth component. That’s another way of saying that most observed economic variables actually lag the best leading indicators by several months. A good example is year-over-year growth in payroll employment, which trails year-over-year growth in real consumption with a consistent lag of about 5 months.
So what do the unobserved components look like today? In the chart below, the green line shows the average standardized value (mean zero, unit variance) of dozens of economic variables, and provides a very good summary of what can be observed directly. Note that this observable data has enjoyed a clear bounce in recent months. The blue line presents estimates of the unobserved economic states that drive the observable data. Importantly, the extracted signals lead the observed economic composite by several months. This is really simply a reflection of the underlying structure of the data – leading indicators lead, and lagging indicators lag (the full analysis uses data since 1950, but the lead of a few months is hard to distinguish visually on a long-term chart).
The most recent estimates we obtain for the extracted economic signal (most recent first) are as follows:
What strikes me about these estimates is short-lived spike in the implied economic signal between September and November. My thinking on the recent improvement in economic data was that it was primarily driven by the large intervention by the ECB near the end of last year. But even when we estimate the parameters of the model using half the data set, and then run true out-of-sample estimates of the economic signal through the present, we still get that burst of improvement in the September through November period. What’s interesting about that improvement was that it was not driven by any obvious shift in the observable data. Rather, the spike was driven by the failure of the data to deteriorate during that period to an extent that would have been expected, given the trajectory of the economic state (this is similar to shifting your expectation for a bird’s flight path not because it turned, but because it failed to turn as much as expected).
In that context, we can see that the improvement in the observable data in recent months has faithfully followed the improvement in that underlying state, which actually happened months ago. Since then, however, the estimated state has deteriorated to a point that is now worse than it was last July.
This is important, because given the deterioration in the inferred economic condition between October and December, it follows that we would expect to see a clear deterioration in observable economic variables over the next 8-12 weeks. If the trajectory holds, the weakness is likely to emerge slowly and then accelerate. For example, the preliminary expectation would be for continued positive payroll growth in March (roughly 50,000-70,000 jobs) and a shift to net job losses in April.
Equally important, to the extent that we observe economic variables coming in better than expected, the inferred underlying state of the economy is likely to improve sharply, as it did last September. This will take a few months of data, but it’s not going to require quarters and quarters of it. At present, we have to view the economic situation negatively, but on the optimistic side, we should also be able to abandon our own recession concerns if the observable data move off of the trajectory that we’ve imputed to-date. On this, leading measures such as consumption growth and OECD leading indices will be most important in driving our estimates of future prospects, while the employment data over the next few months will be useful in confirming the downturn we’ve seen in the inferred state estimates since November.
As noted above, the Market Climate for stocks remains among the most negative 1.5% of historical instances. This time may be different. We see no reason to expect so other than the hope of being lucky. Still, one thing is certain, and that is that present conditions will be impermanent. With certainty, market conditions will shift in a way that removes the present syndrome of overvalued, overbought, overbullish conditions. We don’t know whether that shift will involve a moderate retreat that removes the overbought and overbullish aspects, or a major decline that removes the overvaluation, or just maybe with a further advance that then corrects enough to clear this syndrome at a higher level than the market is today. Conditions might improve without a major breakdown in market internals, or they may improve by a firming of market internals following an extended period of weakness. But with certainty, market conditions will shift in a way that provides us an opportunity to accept market risk at a positive and favorable expected return/risk tradeoff.
Also, with certainty, the present divergence between leading economic measures and coincident/lagging measures will also be resolved. While the historical likelihood is that the coincident and lagging measures will follow the leading measures, we also know that our estimates of the underlying state of the economy could shift quickly in the coming months simply in response to an economy that achieves moderate positive growth, and thereby deviates from what is now an unfavorable projected trajectory.
For now, Strategic Growth and Strategic International remain tightly hedged. Strategic Dividend Value is hedged in an amount equal to 50% of its equity holdings (which is the largest hedge the Fund can establish, but which also leaves the Fund more exposed to general market fluctuations in both directions). Strategic Total Return also remains relatively conservative here, with a duration of about 3.5 years in Treasury securities, and a very small percentage of assets in precious metals shares, utility shares and foreign currencies.