Hard-Negative

With the exception of extreme market conditions (see Warning- Examine All Risk Exposures , and Extreme Conditions and Typical Outcomes ), I try not to wave my arms around about near-term market risks, but I think it's important to cut straight to the chase here. The present market environment warrants unusual concern, in my view. Based on a wide variety of evidence and its typical market implications over an ensemble of dozens of subsets of historical data, the expected return/risk profile of the stock market has shifted to hard-negative. This places us in a tightly defensive position. This isn't really a forecast in the sense that shifts in the evidence even over a period of a few weeks could move us to adjust our investment stance, but here and now we observe conditions that have often produced abrupt crash-like plunges. This combination of evidence includes elevated valuations, overbullish sentiment, market internals best characterized as a "whipsaw trap" on the basis of typical follow-through, heightened credit strains, and clear evidence (on reliable forward-looking indicators) of oncoming recession, among other factors.

As always, we try to align our investment positions with the evidence we observe. If the evidence softens, our hedges will soften. While the quickest route to a modest exposure to market fluctuations (perhaps 20-30%) would be a clear improvement in market internals - which could justify a less defensive stance even in the face of recession risks and rich valuations - the most likely route to a significant investment exposure would be a decline to much lower prices and correspondingly higher prospective returns. Presently, avoidance of major market losses takes precedence in our analysis.

On a valuation front, we estimate that the S&P 500 is likely to achieve an average total return over the coming decade of about 4.8% annually. This is certainly better than the projected returns that we have observed over much of the past decade, but then, the past decade has produced virtually no total return for equity investors at all. An expected total return of 4.8% is also clearly better than is presently available on Treasury bills, which are priced to return a single basis point of interest annually, and is also better than the sub-2% yield available on 10-year Treasury debt.

The problem is that the duration of a 10-year Treasury bond is only about 7 years, which is not only the weighted average time it takes to receive the future stream of payments, but also conveniently measures the expected percentage change in the bond price for a 1% change in long-term return. For stocks, the "duration" mathematically works out to be roughly the price/dividend ratio, which is about 45 for the S&P 500. Put simply, in order to achieve a given increase in long-term expected return, stocks would have to suffer about 6 times the price decline that bonds would experience. Stocks may very well outperform Treasury bonds over the coming decade, but for investors who have any sensitivity to price volatility, that is likely to be a small comfort in the next few years. We estimate that the S&P 500 would have to trade at about the 800 level in order to achieve 10-year prospective returns of 10% annually. Importantly, even a magical "fix" out of Europe would do nothing to change that algebra.


Click here to enlarge

On the sentiment front, Investors Intelligence reports that the percentage of advisory bears dropped below 30% last week, which has historically resulted in unrewarding market outcomes when valuations have been elevated even to a lesser extent than they are today. Thomson Reuters reports that negative earnings pre-announcements are exceeding positive ones by the largest ratio since mid-2001. Investors have eagerly accepted forward operating earnings as a basis for valuation assessments, without accounting for the fact that those earnings expectations assume profit margins about 50% above their historical norms. Unfortunately, profit margins are highly vulnerable to economic weakness, and we are beginning to observe that regularity here.

As noted last week, we continue to estimate a very high probability of oncoming recession. While the economic outlook seems fairly benign based on a "flow of anecdotes" approach (judging economic prospects on the basis of positive or negative surprises in individual reports as they arrive), the outlook is actually very unfavorable based on a more reliable ensemble of leading indicators of economic activity (see Have We Avoided a Recession? ).

That view is clearly shared by the Economic Cycle Research Institute, where Lakshman Achuthan noted on Bloomberg last week that "forward looking data since two months ago has remained weak, it's getting weaker, it's not turning up. So, to my fellow forecasters out there, I'd say they're roughly in two camps. There are those who say that the economy is firming and will continue to firm into next year. We reject that. There's nothing there that suggests that at all. I think there's a larger camp that says we're going to muddle through; we're going to get this kind of slow growth, 'I'm not terribly optimistic, but we're going to muddle through.' I would point out that that's never happened. We never muddle through. A market economy does not want to have a static state. It either accelerates or it decelerates, and these forward looking indicators say decelerate."

Achuthan also noted that "the other half of the GDP report," gross domestic income or GDI (which tends to be the more accurate measure of GDP) was up just 0.3% in the most recent quarter. The Federal Reserve has observed that when GDP and GDI differ, the GDP figure tends to be revised toward GDI, not the other way around. Achuthan warned that the GDI figures are "a big red recession signal." In response to the question "You had a recession call, what happened?," Achuthan simply answered "It's happening."

It's important to be clear that the hard-negative condition of our ensembles here is based on observable data, and our expectation for returns is based on market outcomes that have accompanied past observations that fall into the same classification "bucket" or "cluster" as we see today. The negative average return/risk profile associated with present conditions is, of course, an average, and this specific instance might turn out differently. The problem is that the average is dominated by poor outcomes, some very steeply negative, with a much smaller set of positive outcomes. In any case, our market expectations here are driven by observable data, not by our views about what may or may not happen in Europe. Of course, our concern about high recession risk here is also driven by observable data.

Continue Reading

randomness