Detailed Disclosure Is the Key to Stress-Test Success

Op-ed in Eurointelligence

July 15, 2010

By agreeing to publish stress-test results for the European Union's most prominent banks, European leaders have made a bold step that is both indispensable and ridden with execution risks. Even as the announced date of publication, July 23, is approaching fast, several key aspects remain under discussion. For this unprecedented effort to be successful, a high level of transparency will be needed in order to force divergent national players into a consistent and credible process.

In the past, the key to resolving most systemic financial crises has been a system-wide "triage" of the most important financial firms, steered by public authorities and allowing market participants to identify which banks are strong enough to be relied upon-as in the United States in 1989-90, Sweden in 1992-93, or Japan in 2002-03. The same arguably occurred in the United States last year when the publication of detailed stress-test results for 19 large financial firms helped the gradual return of trust.

The European Union has long stayed deadlocked in refusal to address this need. Stress tests were conducted in September 2009, but with no quantitative data published, the market effect was minimal. Only last month were EU leaders eventually prompted to action by a combination of market pressure, encouragement from the United States and other outside parties, and Spanish leadership. The number of banks tested, initially set at 25, has been expanded to 91 banks from 20 countries, a welcome change given the large role banks play in financing the economy in Europe, which means that more must be assessed than was the case in America.

But many doubts linger, primarily because of the large number of players involved. In the United States, only the Federal Reserve and two other federal regulators were at work under the mandate given by the Treasury Department. In Europe, each national authority has powerful incentives to protect "its" banks and sugarcoat the results, including reluctance to acknowledge past supervisory failures, capture by local interests, and economic nationalism.

[The authorities] should disclose each bank's exposure to each country risk, with due measures to ensure this does not result in instability in the period leading to publication.

This obstacle could have been overcome by centralizing the triage process and entrusting it to a supranational body with a temporary mandate, as Adam Posen, now of the Bank of England's Monetary Policy Committee, and I had suggested in a paper published a year ago. However, this is not going to happen, at least not in this round. Thus, a disciplining device is needed to match national authorities' impulses not to cooperate.

The best way to reach this aim would be a high level of quantitative detail in the disclosure of results. This would compel regulators to base their conclusions on testable facts, and would go a long way toward preventing them from manipulating the numbers. After a string of supervisory failures, the credibility of many of the authorities that will carry out the stress tests is at a low point. Transparency is the key if they are to convince investors and the public that their pronouncements can be relied upon.

Specifically, the results must allow external observers to understand the banks' exposures to several different and relevant risk factors. In the US disclosures of May 2009, 17 quantitative parameters were provided for each of the stress-tested banks, including individual exposures along eight different asset categories under a single, uniform template. At least as many line items should be provided in Europe, also under a single template that must apply uniformly in all participating countries.

As regards eurozone sovereign risk, which is central to current market concerns, it is understandable that authorities are uncomfortable with including national defaults in their stress scenario while they otherwise insist that none is going to happen. But rather than factoring in implausibly low discounts on troubled countries' sovereign debt, as recent press reports suggest is their intention, they should disclose each bank's exposure to each country risk, with due measures to ensure this does not result in instability in the period leading to publication. On this basis, market participants could use the data to test their own scenarios, and identify which banks are strong enough to withstand them.

The European decision to publish stress-test results is momentous. If its outcome is to be credible, it will necessarily reveal significant capital shortfalls in a number of banks. Otherwise, the gap with market perceptions, anecdotal evidence, and past top-down assessments by the International Monetary Fund and European Central Bank will be impossible to reconcile and will increase market distrust and volatility. Not all the subsequent funding needs to come from the public purse. Last year, those US banks whose stress tests indicated insufficient capital were able to convince market investors to plug the gap. Many European banks may succeed likewise. But if some fail, government intervention and restructuring will be indispensible, which is sure to prove politically controversial and financially painful.

These risks can only be managed if the stress-testing process is seen throughout the European Union as comprehensive and fair. To ensure such fairness, a high level of detail in the disclosure of results, forcing national regulators to justify their eventual assessment of capital positions on the basis of facts not opinion, is a necessary condition.