How Market Cycles Are Completed
The market rebound of recent weeks has essentially been grounded in exuberance that the global economy is deteriorating so quickly that central banks will insist on accelerating their monetary interventions. While both corporate earnings and revenues are now in retreat, we also see enthusiasm about the remaining economic activity being captured by a handful of winner-take-all companies. Those two dynamics largely summarize the tone of the market here. The following chart updates our standard economic review of regional and national Fed and purchasing managers’ surveys. The October Philadelphia Fed report was particularly weak on the new orders front, which is complicated by the fact that it’s also one of the more reliable surveys as an indication of broad economic activity. The chart below reflects available data through Friday.
The underlying thesis, of course, is that monetary easing — regardless of its fruitless effects on the real economy — is a reliable signal that the financial markets are open for speculation. But that’s not really the lesson that one should draw even from recent years, because it directly contradicts the evidence from severe market collapses (e.g. 2000-2002, 2007-2009) that were also accompanied by aggressive monetary easing. The proper lesson is that monetary easing is only reliably beneficial for the market when investors are risk-seeking, but may provide no support at all for the market when investors are risk-averse (see When An Easy Fed Doesn’t Help Stocks). Historically, the most reliable way to infer those preferences is from the uniformity or divergence of market action across a broad range of individual stocks, industries, sectors, and security types, including debt securities of varying creditworthiness. That, not the actions of central banks per se, is what investors should remain focused on here.
Valuations are the central driver of long-term investment returns, while market returns over shorter portions of the market cycle are primarily driven by the preference of investors to seek or avoid risk. At present, valuations remain obscene and market internals remain unfavorable. Credit spreads remain wide, and despite last week’s market advance, the percentage of individual stocks above their own respective 200-day moving average hardly budged, increasing only from 37% to just over 38%. Our broader measures of market internals remain unfavorable here as well.
In the event that broad market internals improve materially from here — enough to infer that investors had shifted back toward risk-seeking preferences — we wouldn’t suddenly become bullish, but the immediacy of our downside concerns would be deferred. Emphatically, our expectation of poor long-term market returns on a 10-12 year horizon would not change, nor would our expectation that the S&P 500 will likely retreat by more than half by the completion of the current market cycle.
Absent a material retreat in valuations, we don’t see any likely combination of outcomes where our outlook would not be considered defensive in some way, but we’re certainly open to relaxing the degree of our defensiveness — which is presently close to full-throttle — to a more neutral or subdued near-term outlook where a severe and potentially abrupt market collapse isn’t the most relevant outcome to consider. Certain conditions could potentially support an outlook that we might describe as "constructive with a safety net," but that would require a reversal in a broader range of factors, including credit spreads, for example.
Of course, this message isn’t new. It’s the same one we’ve emphasized almost weekly since mid-2014, when we addressed the challenges that resulted from my 2009 insistence on stress-testing our investment methods against Depression-era data (see A Better Lesson Than “This Time Is Different” if you’re unfamiliar with that narrative).
In short, a sufficient further improvement in market internals would shift us at least to a more neutral outlook, with any additional constructive potential dependent on other observable evidence that emerges (e.g. economic activity, corporate revenues and earnings, investor sentiment, monetary conditions, and so forth). We don’t see any likelihood of advocating an outright bullish or unprotected investment stance, but we know quite well what an overvalued market looks like when investors have the speculative bit in their teeth. It doesn’t matter whether we agree with those risk-seeking inclinations, and we certainly don’t need to speculate if other conditions aren’t supportive. But we’ve learned to temper the inclination to build strongly defensive positions when speculation seems reckless but market internals are uniformly favorable. For now, the market remains highly vulnerable, but we’ll shift our outlook if the evidence shifts.
How market cycles are completed
It’s essential to keep in mind that the equity market moves in cycles between extreme optimism, rich valuations, and poor prospective returns at market peaks, to extreme pessimism, favorable valuations, and high prospective returns at market lows. In our view, the primary factors that identify the likely market return/risk profile as it evolves over the market cycle are: 1) valuations, 2) market internals (which convey information about risk-seeking and risk-aversion), and 3) broader features of market action (such as overbought/oversold status, which can help to identify emerging risks or opportunities). Wide market cycles have been a central feature of the stock market across history. Even a run-of-the-mill bear market decline typically wipes out more than half the gain of the preceding bull market.
A hard-negative market outlook should be supported by observable conditions that have actually been followed, on average, by substantial market losses historically. Based on our current methods of classifying market return/risk profiles, the most severe market losses across history are captured by observable conditions that have emerged only about 9% of the time — a subset that that includes the present. Another 31% conditions are associated with market returns that fall short of Treasury bill yields, on average, but with high volatility, making a generally neutral stance more appropriate. About 60% of market conditions we identify are associated with a prospective market/return risk profile sufficient to justify a fully unhedged or even leveraged market outlook. We expect to observe the full range of market conditions and opportunities as this cycle unfolds. Avoid being lulled into the belief that the recent bull market, or the sideways top formation of recent quarters, are representative of what lies ahead. More likely, the completion of the current market cycle will have no resemblance to either.
The dogmatists running global central banks have encouraged investors to believe that volatility and downside risk have been, and can sustainably be, removed from the financial markets. No — by encouraging the illusion that normal cyclical fluctuations have been eliminated from the dynamics of the markets and the economy, the result has been far more risk taking, far heavier issuance of low-quality and covenant-lite debt, far more yield-seeking misallocation of capital, and far more extreme equity valuation in this cycle than would have been possible otherwise. The consequence will be far more violent market behavior over the completion of the cycle than investors would have faced otherwise.
We’re open to a flatter near-term outlook if conditions emerge that would support that outlook. This would at least require a broader improvement in market internals than we currently observe. At the same time, we would see it as a dangerous failure of historical perspective and imagination to assume that future market returns will be limited to either 1) a continuation of an aged and hypervalued bull market, or 2) a range-bound consolidation where every 5-10% dip is a buying opportunity. The declines and recoveries we’ve seen over the past year are nothing that we did not also see during the extended 2000 and 2007 top formations. Once persistently overvalued, overbought, overbullish conditions were followed by deteriorating earnings growth and a breakdown in market internals, the goal wasn’t to speculate on rebounds, but only to limit the amount of frustration one had to endure during the top formation — in anticipation of the more severe market losses that followed.
Even in periods where interest rates have been quite depressed, not a single market cycle in history failed to end with estimated prospective 10-year S&P 500 returns close to 10% annually, if not dramatically higher. Based on the most historically reliable valuation measures, the S&P 500 would have to lose literally half of its value for prospective returns to rise to that level. A 50% market loss isn’t a worst-case scenario. Given current valuations, it’s the standard, run-of-the-mill outcome that investors should expect over the completion of this cycle.
The measures we find most strongly correlated with actual subsequent S&P 500 total returns now project zero total returns for the S&P 500 on a 10-year horizon, and about 1% annual nominal total returns for the index on a 12-year horizon. The chart below shows one of these measures, MarketCap/GVA, on an inverted log scale (left) compared with actual subsequent S&P 500 nominal annual total returns over the following 12-year period.
Investors are now facing the second most extreme episode of equity market overvaluation in U.S. history (current valuations on similar measures already exceed those of 1929). The belief that zero interest rates offer no alternative but to accept risk in stocks is valid only if one believes that stocks cannot experience profoundly negative returns. We know precisely how similar valuation extremes have worked out for investors over the completion of the market cycle, and those outcomes have never been deferred indefinitely. The only question at present is how many grains are left in the hourglass. On that front, the outlook remains unfavorable at present. Our attention remains fixed on measures of investor risk preferences, as conveyed through the uniformity or divergence of market internals, and the immediacy of our concerns will ease if the evidence changes.