Newspaper article The Evening Standard (London, England)

Let's Not Get Too Uptight over Bank Stress Tests; CITY COMMENT

Newspaper article The Evening Standard (London, England)

Let's Not Get Too Uptight over Bank Stress Tests; CITY COMMENT

Article excerpt

Byline: Anthony Hilton

PEOPLE in the currency markets have been caught out by the strength of the euro in the past few weeks. Many were still stuck with the idea that its weakness was long term, given the problems of Greece and the potential political difficulties with the massive stabilisation fund created to buy the Greek government time to address its finances.

Instead, sentiment has been turned by the intended publication at the end of this week of the results of stress tests on most of the leading banks in Continental Europe.

Ever since the financial storm broke two years ago, buyers of the euro have worried about how bad things truly are in the banking sector, and whether there will have to be further government bailouts. The strength of the currency comes because they think they are about to find out.

There is some logic in this but not a lot. About a year ago, the American authorities insisted that all US banks conducted such an exercise. When they did, markets responded positively. The hope is that publication of the European results will trigger a similar return of confidence.

The problem is that most people do not understand what stress tests are, or how limited they are in what they tell us.

New York-based Michael Thompson of Valuation & Risk Strategies, an independent business within Standard & Poor's, says markets make a big mistake in thinking that stress tests are a crystal ball -- that they predict with certainty what will happen when economic conditions change.

The value at risk calculations that lie at the heart of stress-testing are based on historic correlations between asset classes -- for example, that when bonds go up equities are likely to go down.

However, as the crash showed two years ago, these correlations operate only in normal times. When a real crisis erupts, everything goes down and the value at risk models simply don't work. Thus, stress testing tells you nothing about what happens when the system explodes.

This is not to say the process is useless, but we should know its limitations.

General Eisenhower, the US Second World War commander, said that no plan survived five minutes' contact with the enemy, but nevertheless it was vital to plan. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.