European Union Flunks Test for Stress Test

It's easy to mock the legitimacy of government-run stress tests, even when they work. In just over a year, many observers have done it three times.

Like its predecessors, a European Union test of 22 banks in October and the May 2009 U.S. exam of 19 banks, Friday's 91-bank euro-zone revue drew a modest market reaction and a justifiable round of catcalls dismissing it as political theater.

But that alone didn't make it worthless. The exercise's real import, like its antecedents, was whether regulators would use the results as cover for a broad, coordinated campaign to delever European banks and shore up their capital. They didn't.

"As we learned in 2008, no one really understands large financial institutions, and that includes a lot of CEOs of large financial institutions," said James Bianco of Arbor Research. "What the [U.S] stress did successfully was force them to raise $85 billion in capital. We may not understand what these banks are doing or their loss potential, but we do understand that more capital is better than less for those who are not investors."

If regulators had used the tests to prod European banks to raise capital, American institutions might soon be trading, and competing, with stronger peers across the Atlantic, some observers said. But because the tests identified only seven banks with $4.5 billion in capital needs out of a system holding many trillions of dollars of assets, it seems unlikely that the tests will produce significant action.

"That's peanuts," Bob Eisenbeis, an economist at Cumberland Advisors, said of the seven banks' capital shortage.

There are plenty of reasons to dismiss the integrity of the European stress-test conclusions. The most obvious may be the predominantly clean bill of health appears to be based on the dubious precept that no European banks are carrying unrealized losses.

"The representation is that all the balance sheets were pristine at the start of the scenario," Eisenbeis said.

Another possible shortcoming is the examiners' use of a double-dip recession causing an aggregate drop of 1.1% in Tier 1 capital as the "adverse scenario" in their simulation. Had the U.S. done that last year — rather than applying expected loss assumptions throughout banks' various portfolios — few banks would have appeared undercapitalized. (Ten of the 19 U.S. banks were required to raise capital.)

"The European banks are "going to have to [raise new capital], and it's a matter of when and how much," Eisenbeis said.

As an objective measure of bank health, the EU test suffered from a checklist of crippling flaws. Opacity? Yes. Assumptions that appear to ignore tail risk widely assumed in credit default spreads and bond prices? Oui. Reliance on regulatory capital ratios discredited during the crisis? Ja. Partisan national regulators? Si.

But to say that the test lacked credibility is to miss the point. Many of its flaws were present in the U.S. stress test, which Eisenbeis derided as "a fig leaf" at the time. Though the U.S. test results did provide some new information on bank health — at least more than the European tests did — their main value was to diminish the stigma of diluting shareholders and assure investors that the government would stand behind the banks once they cleared a prescribed capital bar.

Bert Ely, a banking consultant in Virginia, said that any renewed confidence the U.S. stress tests produced didn't come from giving all the banks a clean bill of health, but from giving them a clean bill of health while compelling them to raise capital. "That was basically was what the whole exercise was about," Ely said.

Eisenbeis went further, suggesting that an essential part of the U.S. stress tests was that their accuracy appeared to have been premised on the government's continuing efforts to nurture markets and the industry with ultralow interest rates and support for vital asset classes like housing.

"You can't forget the structure of the subsidies, the risk-free arbitrage that was set up," he said. "It was essentially money in the bank."

Both Eisenbeis and Ely questioned whether European governments could uniformly instill that kind of confidence, given sovereign debt fears about Greece and other euro-zone nations.

Market reaction on Friday seemed to suggest that investors were largely ignoring the results. Credit default swap spreads fell slightly after the announcement, and the euro's value was flat.

Some U.S. bank executives went as far as writing off the tests' outcome as irrelevant to their companies. Richard L. Carrion, CEO of Popular Inc., said in an interview that he had little reason to believe the tests would affect his business.

But Eisenbeis suggested that the tests might yield a competitive advantage for globally focused U.S. institutions, assuming that the results mean European regulators don't have the stomach to force banks to recapitalize.

"This is a competitive advantage," he said. "On the margins, if you're a business in the U.S. or Europe needing facilities, you go to the stronger institution."

But even if stress testing alone is ineffectual, it isn't free. As in the U.S., European regulators have now designated, and blessed, a rather lengthy list of potentially systemically important banks. LECG Global Financial Services chairman Bill Isaac, former Federal Deposit Insurance Corp. chairman, was and remains opposed to last year's U.S. stress tests because of market distortion concerns. He said regulators both here and in Europe may live to regret trying to use their credibility to backstop banks.

"It's a horrible idea to play all this out in the political and public arena," Isaac said, "and I think Europe is going to have some of the same issues. It so frightened the markets that we had to declare 19 banks 'too big to fail,' and that has created a competitive inequity that exists to this day."

For reprint and licensing requests for this article, click here.
MORE FROM AMERICAN BANKER