Learning from the shock

The terrorist attacks in the USA violently shook stockmarkets and financial markets. The Dow-Jones-Index, the most important of the US indexes, lost 15 percent of its value within a few days. Can one prepare for falling prices and what lessons can we learn from such shocks? An interview with Wolfgang Breymann, currently visiting Risklab of the ETH where he is investigating volatility in financial markets.

Interview: Norbert Staub

The attacks on the World Trade Center could speed up the decentralisation process of the markets, says PD Dr. Wolfgang Breymann, Senior Researcher at RiskLab of the ETH.

Wolfgang Breymann: Yes, I was expecting a strong reaction from the markets. According to a report in the «Financial Times» it was the greatest loss in a week since 1933. Nevertheless, the markets rose sharply on Monday after trading was resumed on Wall Street, a sure sign of a highly volatile market. The attack was a horrendous shock which no one was expecting. The stock markets were in a difficult situation even before the event. Fear of recession was rife in the USA. The attacks caught the markets off balance, so to speak.

Can you give us some examples?

Breymann: Certain branches were directly effected, the insurance and reinsurance industries, for example, airlines and plane industries — think about the aid programme for Swissair, for example — not forgetting the entire tourist industry. Naturally, these titles are all doing badly within the current stock market development. But these are also economic questions. Important direct effects were also felt by the financial markets.

Such as?

Breymann: Some major US investors sold big positions after the attacks. There were rumours that these investors were forced to sell stock to support other positions in the market. Without being overly pessimistic — this is a sure sign of trouble. The attacks hit the markets in their very infrastructure. When companies that play a great role in the functioning of the market are themselves wiped out, insecurity follows. Suddenly, no-one knows who sold what to whom; reconstructing transactions is tremendously complicated. One broker firm, for example, Fitzgerald Cantor, which had offices in the World Trade Center, handled more than a third of the total liquidity of US bond market. One consequence was that the volume of trading shrank considerably.

Your work involves the scientific analysis of financial market volatility. What does this mean?

Breymann: Volatility is a measure for the strength of fluctuations that can be expected. In my research I try to predict volatility in the long-term using very short-term data.

How can such volatility predictions be used?

Breymann: They are useful, for example, in assembling risk models — to measure expected risk. Volatility is also an important parameter when it comes to pricing options, which are important derivative financial instruments.

Do events such as those that happened in the USA influence your research?

Breymann: These events are unprecedented. At the same time they are a chance to learn. It will be interesting to see how this shock effects volatility. One thing is certain: external shocks, such as this, cannot be foreseen. One aspect that becomes central is the importance of volatility research for insurance companies. This kind of problem clearly falls within the area of work being carried out at RiskLab at the ETH. New York could become a test case for questions such as, how good are existing prediction models when tested against developments after the event? The Iraqi invasion of Kuwait ten years ago could be used as a basis for comparison; this was also an unpredictable event and the markets reacted very strongly at the time.

Can what the markets are going through at the moment be integrated into a typical fluctuation picture?

Breymann: Of course, fluctuations are greater at the moment than in «normal» times. In itself, however, volatility is not a bad thing. Financial markets can be volatile and still show constant — «lateral» — development. Volatility opens up the possibility of profit for dealers; big fluctuations, not small ones, make profit possible. But naturally, stronger volatiality brings higher risk. The catastrophe in New York is influencing short-term development. In the long run the markets will pick up.

Are there differences in the degree to which the markets were affected by the attacks?

Breymann: Yes. We have to differentiate between the stock exchanges, currency markets and bond markets. The US bond market was especially hard hit by the destruction of the company Fitzgerald Cantor, much harder than the stock market. With hindsight we can probably say that it would have been wiser had the infrastructure not been concentrated in one place, with one company even. On the other hand, the specialisation of Fitzgerald Cantor meant that the service was handled efficiently and at lower costs. In contrast to the LTCM-Crash three years ago, the attack in New York was not a disturbance created mainly by the market itself (such as a liquidity crisis) but an interruption in the infrastructure of the market — which can still cause serious problems. The Federal Board came to the help of investors with cheaper credit in order to avoid liquidity bottlenecks. And it worked.

Do you think it will be possible to set up a sort of «early warning system» to protect markets from such events?

Breymann: That is difficult because the more interwoven markets are, and this is global reality, the greater a chance a crisis has to spread. Should such a system ever be developed, however, volatility research could provide important data. One thing is certain, the voices claiming that the high physical concentration of markets on Wall Street is not necessary — and even dangerous — will increase in number. The decentralisation process already started in April with new technology at the NASDAQ powered by the development of communication technology.

Founded in 1994 at the ETH Zurich, RiskLab is a research institute in collaboration with other universities, which carries out applied research in the areas of risk management for financial and insurance institutions. RiskLab's partners in this undertaking, apart from the ETH are: Credit Suisse Group, Swiss Re and UBS. RiskLab also has informal ties with individual members of the Department of Mathematics of the ETH Zurich and the Swiss Banking Institute and the University of Zurich.

ETH Life, 25. September 2001 (Translation commissioned by RiskLab, links set by RiskLab)