Ignoring the obvious

IMAGINE a patient clutching at his heart, complaining of sharp chest pains. A doctor arrives, examines him carefully and pronounces him healthy—provided he is not having a cardiac arrest. The same air of unreality infects today’s stress-test results for European banks: most institutions are fine unless there is a sovereign-debt crisis.

The tests, which were conducted by the European Banking Authority (EBA), found that eight banks out of 90 tested had failed to pass the threshold of a core Tier-1 capital ratio of 5% under a stressed scenario. Five were Spanish, two were Greek and one was Austrian. Another sixteen banks posted ratios of between 5% and 6%, dangerously close to failure.

Despite some claims to the contrary, the tests are not a waste of time. All of the banks involved in the tests have to disclose lots of information about their sovereign-debt holdings, which will help analysts to identify the banks that are most exposed to the unfurling euro-area crisis. They may have helped encourage some banks to raise capital ahead of time: European institutions added around €50 billion ($71 billion) of capital between January and April of this year.

The eight banks that failed to meet the EBA’s pass threshold and those that floated just above it will come under pressure from the markets to raise equity. The task facing the Spanish banking system has been made clearer, in particular: five of the eight banks that failed were Spanish, and without the capital-raising in the first four months of the year, nine out of the 20 banks that would have failed were Spanish.

Counting the flawsIn at least three respects, however, the tests are badly flawed. The most obvious is that they gloss over the severity of the sovereign-debt crisis. They assume that no euro-zone default occurs; to do otherwise would have been to trample over the official European line on the crisis in Greece and elsewhere.

Haircuts on sovereign-debt holdings are applied only to those in the trading book; whereas most banks hold government bonds in the banking book, where they are carried at their original value. Those haircuts are in any case too slight: they assume a 25% drop in the value of Greek government debt, for example, when the market is pricing them at closer to a 50% write-down. The EBA did make banks provision for their banking-book exposures, but they have not exactly cracked the whip: out of a grand total of €377 billion in provisions across the 90 banks tested, just €11 billion relate to sovereign debt.

The second flaw relates to the way in which the crisis has evolved in just the past week, with Italy joining Spain in the firing-line. The tests are a useful indicator of the direct vulnerability of banking systems to shocks. But they are much less useful, as the EBA itself acknowledges, when it comes to working out second-order effects of sovereign-debt distress on things like investor confidence and availability of funding. And both the direct and the second-order effects of the euro-area crisis get worse when bigger economies are in the frame. The stress tests can guide policymakers if they want to beef up the defences of the banking system in case a minnow like Greece topples. They are much less help with a country like Italy or Spain because the contagion effects would be so much greater and less predictable.

The final flaw in the tests relates to what comes next. Europe’s banks need to add capital if they are to withstand the effects of a debt restructuring in Greece, Portugal or Ireland. But that has been the case for well over a year and yet they have still dragged their heels on recapitalisation. Between the end of 2009 and the end of the first quarter of this year, American banks added roughly four percentage points of core Tier-1 equity (see chart). No European banking system came close to doing the same, and their absolute levels of capital remain much lower.

These tests may be the catalyst for a wave of capital-raising but there are clear signs that the EBA will face lots of resistance from national regulators, many of whom dispute its purist definition of what counts as high-quality capital. One German bank, Helaba, this week refused the EBA permission to publish all of its data after the EBA disqualified some of its capital. Helaba must surely have had the backing of Bafin, the German banking supervisor, in so doing. The Spanish authorities are also on a clear collision course with the EBA, which wants banks that failed the tests to put together plans to fill the shortfall within three months. The Spanish central bank tonight issued a press release saying that  no Spanish bank would be required to add capital as a result of the tests.

The newly formed EBA has done a better job than its predecessor did with the 2010 stress tests. But the 2011 version still looks doomed to irrelevance: too soft to reassure the markets, too unenforceable to prompt the recapitalisation that is needed.