Related audiobook: You're It!: On Hiding, Seeking, and Being Found
“Science is the systematic classification of experience.” – George Henry Lewes
I’ve noted frequently in recent months that the lessons to be drawn from the recent market cycle are not that historically overvalued, overbought, overbullish extremes can be dismissed. Rather, the lessons to be drawn have to do with the criteria that distinguish when such extremes have little near-term impact from periods where they suddenly matter with a vengeance.
Although we agree, as John Templeton once observed, that the four most dangerous words in investing are “this time it's different,” the fact is that one very specific effect of quantitative easing made the half-cycle since 2009 different from history, and forced us to struggle quite a bit. Market cycles throughout history have demonstrated an important regularity: once a syndrome of overvalued, overbought, overbullish conditions was established (not one condition alone, but the full syndrome), the behavior of the stock market took on what I’ve often called an “unpleasant skew” – the market would typically follow with a few weeks of persistent small advances, followed by an abrupt and steep vertical plunge that wiped out weeks or months of gains in a handful of sessions.
In the face of quantitative easing, however, that pattern changed. As short-term interest rates have been held near zero, investors have been drawn into “carry trade” mentality, believing that they must take risk in stocks, regardless of valuation, because they have “no other choice.” Given that mentality – and make no mistake, this ispsychology at work, not financial calculation – overvalued, overbought, overbullish syndromes have persisted and extended in the half-cycle since 2009, often with no downside effects at all. Admittedly, I relied too heavily on the wicked historical record of these syndromes. But rather than discarding the lessons of history altogether, we did what we always do when faced with a challenge – which is to look for adaptations that are consistent both with historical fact and with new evidence.
The upshot is this. Quantitative easing only “works” to the extent that default-free, low interest liquidity is viewed as an inferior holding. When investor psychology shifts toward increasing risk aversion – which we can reasonably measure through the uniformity or dispersion of market internals, the variation of credit spreads between risky and safe debt, and investor sponsorship as reflected in price-volume behavior – default-free, low-interest liquidity is no longer considered inferior. It’s actually desirable, so creating more of the stuff is not supportive to stock prices. We observed exactly that during the 2000-2002 and 2007-2009 plunges, which took the S&P 500 down by half in each episode, even as the Fed was easing persistently and aggressively. A shift toward increasing internal dispersion and widening credit spreads leaves risky, overvalued, overbought, overbullish markets extremely vulnerable to air-pockets, free-falls, and crashes.
What’s rather beautiful about this distinction is that it applies equally well to bubble periods such as the late-1990’s, the housing bubble, and on imputed sentiment data, the advance to the 1929 peak, and these considerations help to identify the shifts that invited subsequent crashes. So unless one believes there’s something magical about quantitative easing that goes beyond any well-articulated or identifiable transmission mechanism, it’s quite a good idea to pay close attention to market internals and risk premiums here.