“I cannot imagine any condition which would cause a ship to founder. I cannot conceive of any vital disaster happening to this vessel. Modern shipbuilding has gone beyond that.” – Captain Edward Smith, RMS Titanic
“One reason that risk premiums may be low is precisely because the environment is less risky… The Fed has long focused on ensuring that banks hold adequate capital and that they carefully monitor and manage risks. As a consequence, banks are well-positioned to weather the financial turmoil.” – Janet Yellen, July-September 2007
What was the difference between the Titanic that left Southampton with all of the advantages and technology of modern shipbuilding, and the Titanic that plunged to the bottom of the Atlantic – still carrying all of the advantages of modern shipbuilding? Just one condition, really: a hole in the side of the ship. But distinctions matter.
One of our central themes in recent months is that – departing from what we’ve observed for most of the advance in recent years – market internals and credit spreads have deteriorated materially (see The Ingredients of a Market Crash). Particularly since late-2011, extreme overvalued, overbought, overbullish syndromes have persisted without significant corrections, despite the historical tendency for those same syndromes to devolve rather quickly into severe market losses. What follows is the primary lesson we’ve drawn from our own struggle with this. I am going to repeat it again because even a few days of strong market action make it easy to forget. My hope is that those who value our work will find it useful in placing the recent cycle in perspective. Moreover, it provides a consistent framework to understand the bubbles and crashes that have manifested over the past 15 years, and the behavior of market cycles across history.
Market history, including the series of bubbles and crashes over the past 15 years, does not teach that valuation is irrelevant, but instead that a key distinction affects whether stability or instability is likely to prevail. When rich valuations are coupled with tame credit spreads and uniform strength across a broad range of market internals and security types, one can infer that investors remain tolerant toward risk. In that environment, risk premiums may be low, but there’s no particular pressure for them to normalize, even if the speculation is driven by mindless yield-seeking. Trend uniformity and well-behaved credit spreads are an indication of risk tolerance, which allows overvalued markets to remain overvalued without immediate consequence. In stark contrast, increasing dispersion across securities and sectors, deteriorating market internals, and widening credit spreads are all subtle but observable indications of growing risk aversion – icebergs that can easily rupture the Titanic of severe overvaluation. Monetary easing then no longer supports risky assets, because risk-free liquidity is no longer seen as an inferior asset. This risk-aversion creates upward pressure on low risk premiums, which normalize not smoothly but in spikes, resulting in air-pockets, free-falls and crashes.
As warned in October 2000, “when the market loses that uniformity, valuations often matter suddenly and with a vengeance. This is a lesson best learned before a crash rather than after one.” I offered the same warning just as Janet Yellen was making her dismissive remarks about low risk premiums in 2007 (see Market Internals Go Negative).
With regard to our own experience in recent years, the conditional tolerance for extreme overvaluation and overextended conditions was a feature embedded into our pre-2009 methods. Despite their admirable record, my 2009 insistence on stress-testing our approach against Depression-era data led to a difficult transition in recent years, as we eventually discovered the need to explicitly overlay this feature (albeit in a slightly different form) even in the most overvalued, overbought, overbullish conditions we define. Our difficulty resulted from neither our pre-2009 methods of classifying market return/risk profiles nor from our present methods. Both effectively navigate investment conditions in data from market cycles across history,as well as the half-cycle since 2009. Our difficulty was in the awkward transition between the two.
It’s notable that this framework applies even in an environment of monetary easing. To reiterate the key points from last week:
“If credit spreads are widening and internal uniformity is deteriorating, one can infer that investors are shifting toward risk aversion. In that environment, safe, low-interest liquidity is no longer an inferior asset but a desirable one, and creating more of the stuff is not reliably supportive to risky asset prices. This is borne out across a century of history, including the 2000-2002 and 2007-2009 declines, and can be observed even in several episodes since 2009.
continue reading: http://www.hussmanfunds.com/wmc/wmc141222.htm
