These recent posts on the subject of active management have sought to distil the work of those who have found pattern in asset prices and to consider what relevance their work has with respect to today's markets. In this post, I look at asset price bubbles and the work of theoretical physicist turned bubble-spotter, Didier Sornette.
"Can one spot asset bubbles and predict when they'll burst?"
Former Federal Reserve governors Alan Greenspan and Ben Bernanke both said that it was impossible to spot bubbles in asset prices until after they had burst. It is possible they believed this. More likely, they realised that an admission to the contrary would impose on them a responsibility to pop bubbles, perhaps complicating or, at worst, endangering their key objectives with respect to inflation and employment.
Sornette founded the Financial Crisis Observatory (FCO) in the aftermath of the 2008/9 financial crisis. He wanted to understand how a few hundred billion of losses in one small corner area of the financial world — US subprime lending — triggered a US$5 trillion contraction in world GDP and almost US$30 trillion of losses in global stock market capitalisation. Furthermore, was the carnage directly related to the 30 or so years of stability — known as the Great Moderation —that preceded it?
The FCO was established as "a scientific platform aimed at testing and quantifying rigorously, in a systematic way and on a large scale the hypothesis that financial markets exhibit a degree of inefficiency and a potential for predictability, especially during regimes when bubbles develop." In other words, can one spot asset bubbles and predict when they'll burst?
Use of the word bubble with respect to financial markets dates back to 1720 and the passing in June of that year by the British parliament of The Bubble Act, a response to the 87% collapse in the stock price of the South Sea Company and consequent bankruptcy of numerous and important investors. Since then, bubbles — or, as they are also known, asset bubbles, financial bubbles, speculative bubbles — have been defined in numerous different ways but all seem to relate in some way to prices that are far above intrinsic value.
Yale's Robert Shiller defined a speculative bubble as "a social epidemic whose contagion is mediated by price movements. News of price increase enriches the early investors, creating word-of-mouth stories about their successes, which stir envy and interest. The excitement then lures more and more people into the market, which causes prices to increase further, attracting yet more people and fuelling "new era" stories, and so on, in successive feedback loops as the bubble grows. After the bubble bursts, the same contagion fuels a precipitous collapse, as falling prices cause more and more people to exit the market, and to magnify negative stories about the economy."
Sornette's own, and simpler, definition of bubbles is that they are "significant persistent deviations from fundamental value". It is the study of complex systems that links the financial world with that of theoretical physics and thus what attracted Sornette to the former. The weather and the stock market may not at first appear related but both are driven by a vast multitude of positive and negative feedback loops that can be described using a common framework.
Sornette and his colleagues at the FCO developed a theory called "dragon-kings" as a framework for understanding asset bubbles. Unlike so-called black swans — events that come completely out of the blue — dragon kings are catastrophic occurrences but ones
whose origins are very much traceable.
In financial markets, dragon kings are evidenced by price movements that fall far outside a normal expected probability. The term dragon king is a reference both to the mythical creature falling far outside the normal classification scheme —Sornette himself describes dragons as "extraordinary animals of extraordinary properties" — and to the King Effect, in which a simple linear relationship describes the distribution of wealth of all members of a society other than those at the very top. According to Sornette, "The root mechanism of a dragon king is a slow maturation towards instability, which is the bubble, and the climax of the bubble is often the crash."
Getting a little more technical, one particular signal that a bubble is developing is super-exponential growth with positive feedback. Rather than being followed by a price reversal as is often the case, a 1% price rise instead triggers a 2% price rise which in turn triggers a 4% price rise, etc.
In the real world, of course, things are not so simple, but Sornette's theoretical models not only appear to mimic closely the growth phase of financial asset bubbles but also to define a finite-time singularity, the point at which a bubble bursts. This bursting may be quick, a crash, or slow, a plateau. Either way, according to Sornette, "the information about the critical time is contained in the early development of this super-exponential growth."
It appears that Sornette has had some success in recent years in identifying ex ante a number of asset bubbles. In September 2007 he predicted that the bubble in Hong Kong and Chinese shares would "change regime" by the end on the year and that there "might be a crash". More recently, on 17 May last year, he noted that the US stock market was on an unsustainable trend and that there would be a correction — as indeed there was — but that this was only part of a "massive bubble in the making".
Perhaps in response to criticism, and to add rigour to his approach, Sornette now encrypts his predictions, posts them on an international archive, then releases a public key six months later. Sornette now regularly posts on his website an FCO Cockpit — an assessment of bubble tendencies of 435 systemic assets or indices. The latest, dated 1 April, noted that his proprietary bubble risk indices were turning red for Spanish, Italian and Irish bonds, as well as for European financial services subordinated bonds, reflecting a belief that "the European sovereign debt crisis is over".
In equity land, Portuguese and Egyptian shares were flashing red. Notably, price action in the US in February and March had taken some of the "bubble pressure" off the likes of Amazon, Netflix, and Testa, though Life Sciences Tools and Services, Pharmaceuticals, and Healthcare Providers and Services sectors were still exhibiting strong bubble signs. All of these to varying degrees have seen some sort of correction. Perhaps the main argument against the notion that it is possible to identify bubbles and predict when they'll burst is that if that were indeed possible, investors would anticipate them such that they'd never have a chance to develop in the first place. However, that assumes investors always behave rationally whereas the reality is that at times, and en masse, they don't. Maybe that is what Sornette's models have going for them.
Published in Investment Letter, April 2014
The views expressed in this communication are those of Peter Elston at the time of writing and are subject to change without notice. They do not constitute investment advice and whilst all reasonable efforts have been used to ensure the accuracy of the information contained in this communication, the reliability, completeness or accuracy of the content cannot be guaranteed. This communication provides information for professional use only and should not be relied upon by retail investors as the sole basis for investment.