Test may burst myth around financial bubbles

Popular economic thinking among academics is that bubbles cannot be identified until it's too late Reuters

Swiss researchers claim to have found a way of forecasting financial bubbles, as yet unpredictable forces in the boom and bust nature of economics.

The findings could have implications in the aftermath of the recent global financial crisis, says the project leader, Didier Sornette, head of Financial Crisis Observatory at the Swiss Federal Institute of Technology in Zurich.

Sornette, who has long tried to understand predictability in crises ranging from earthquakes to stock exchange crashes, says the team’s research could trigger a paradigm shift.

“The financial crisis was regarded as unpredictable and consequently no one was blamed. This suited many down to the ground. If we can prove we’re right however, the textbooks will have to be rewritten,” he said.

Results of the team’s so-called Financial Bubble Experiment were announced to the media six months after the project began.

According to popular belief financial bubbles – loosely defined as trade in products or assets with inflated values – cannot be identified while inflating. But the researchers went on the premise that bubbles could be diagnosed before they burst and significantly, that their end could also be predicted.

More predictions to follow

Sornette says hallmark signs of a bubble are faster than exponential growth, followed by “positive feedback” from optimistic investors, which inflates prices.

For his experiment four assets were selected in which they expected a financial bubble to form over the next six months: Brazil’s Povespa Index, the Merrill Lynch bond index, the gold spot price and cotton futures. They then announced forecasts for when a “regime shift” of strong growth followed by moderate growth or declines – or vice versa – would happen to the assets.

The researchers found all four showed a clear regime shift within the forecast time window or shortly before it, which Sornette says proves there is an identifiable structure to the dynamics of financial markets and so confirms that financial bubbles can be diagnosed while they are in development.

Sornette says it is also possible to forecast the end of a bubble, although to varying degrees of success, but adds: “If you can diagnose a bubble during its run, predicting the ultimate end point might be secondary especially if you are going to act on it.”

The team will now be carrying out further tests to refine the diagnosis methods, with forecasts of seven new financial bubbles due to be announced next week.

Different schools of economic thought continue to debate the impact of economic bubbles, their harm and whether they can and should be deflated. Former Federal Reserve chairman Alan Greenspan said back in 2002 that economic bubbles have to burst to confirm their existence.

Exaggerated claims?

Silvio Borner, from Basel University’s economics faculty, told swissinfo.ch that Sornette may have gone too far in the interpretation of the results.

“I think my academic colleague has some very good points, especially with regard to the shortcomings of the traditional view on this, that is, that everything is really random. I think the [recent financial] crisis and the breakdown of corresponding models proves that these models or these views are really not up to the task.”

“On the other hand I think his claim is quite daring and not really marked by much modesty. It’s a little bit unusual for someone to come out and say the textbooks have to be rewritten.”

Crucial would be to determine when the so-called regime shifts take place, he said. “If you know it is a bubble and you go short and get out of it too early then you lose everything also. If you can really predict bubbles I would advise the guy not to be a professor but to become the richest man in the world.”

Uphill battle

This is not the first time Sornette has put his theory to the test. In 1999 he correctly forecast that the Nikkei index would recover by the end of the year, but in 2002 wrongly predicted the mood of the US stock market in 2003-2004.

In another test, he predicted bubbles and market crashes in two indexes in the Chinese stock markets between 2005 and 2009.

The author of a book on the financial sector and various studies into predictability, Sornette has been focusing on fluctuations in the financial markets since 1992 and has thrown his weight behind developing empirical evidence on the issue. The Financial Crisis Observatory in Zurich was designed to scientifically test predictability in financial markets.

“There is so much vested interest in ignoring what we are trying to say. I see it as an uphill battle,” he told swissinfo.ch.

“What is lacking in this field is that the approach is much more nominative than empirically based, that is, how it should be rather than how it is. We are trying to unearth the underlying mechanisms.”

“I am not saying we should throw away the economic approach. It is very interesting but it is too limited.”

Jessica Dacey, swissinfo.ch

Big bubbles in history

Tulip mania (1633-1637)

British railway mania (1840s)

Stock market bubble (mid to late 1920s)

Japanese asset price bubble (1980s)

Dot.com bubble (1995-2000)

(Sources: Encyclopedia Britannica, World Bank)

Didier Sornette

PhD in statistical physics approach to models of biological cell membranes and interfaces (1979-1985).

Military duty in France on acoustics, chaos and hydrodynamic turbulence (as a scientific researcher).

Focus on earthquakes in 1987, later became professor of seismology and geophysics at UCLA from 1996 to 2006.

Research director at the French National Centre for Scientific Research working on quantum chaos, self-organised criticality and self-organisation.

1994: co-authored landmark paper on a functional integration approach to financial option pricing and hedging.

Current: Swiss Federal Institute of Technology: Professor of Finance (Chair of Entrepreneurial Risks); co-founder of the Competence Center for Coping with Crises in Socio-Economic Systems; professor of physics associated with the Department of Physics; professor of geophysics associated with the Department of Earth Sciences.

Adjunct Professor of Geophysics at IGPP and ESS at UCLA

Economic stance: In his blog he “challenges the community to dare thinking of new concepts and develop new methods in order to diagnose in advance the future crises. This is opposed to the general view that crises are intrinsically unpredictable and that our predicaments come from bad luck, the way it is via natural business cycles.”