naked capitalism

The stress tests conducted on 19 large American banks by the US Treasury in 2009 were an amazingly effective exercise in salesmanship and sleight of hand. Banking industry experts, including Bill Black, Chris Whalen, and Josh Rosner, dismissed the process as mere theatrics: too little staffing and not enough “stress” in the economic forecasts and loss assumptions (particularly on second mortgage). My pet peeve was that the banks ran the tests on their trading books using their own risk models, the very ones that had performed so well in preparing them for them in the runup to the crisis.

But the Treasury’s Tinkerbell strategy worked. If they could create enough confidence, if they could get enough people to applaud, the banks would live – at least for a while. The spectacle of daily coverage in the business press of the tests, including the howls-on-cue from the banksters, outraged by supposedly-unreasonable demands the Administration, created the impression that Something Was Being Done. And the Treasury did get one critical bit right: it had a credible process for making sure the banks would be able to plug any capital shortfall it identified, and that was by having able to have the government pony up the money to fill any shortfall they identified that the banks couldn’t fill on their own.

Imitation is the most sincere form of flattery. The ECB and European bank regulators are copying the US playbook for the stress tests, with results for 100 banks expected to be released around July 23. But the European authorities seem to have failed to understand why the US effort worked. The first was that Team Obama is particularly good at PR, and it used those skills to full advantage. Despite considerable evidence otherwise, it got the press to convey the message that the tests were tough, and the banks really were sound. Second, Geithner & Co. had a kitty they could draw on.

By contrast, the Europeans have been simply dreadful at the optics of their various rescue operations, with disarray and disagreements covered extensively by the media. Admittedly, this exercise is being conducted by bank regulators, so it is likely to be more cohesive, but “more cohesive”, with a process involving agencies in different countries, may not be cohesive enough. And “show me the money” is a major problem. The reason for this exercise is concern over possible sovereign debt losses. Who is going to back up the banks at risk? Um, sovereign states, admittedly ones not considered at risk of default (France and Germany), but whose ability to bail out their own banks is limited for practical and political reasons.

A story in today’s Financial Times provides confirmation of the skeptics’ concerns:

After another fast-moving news week, as it emerged that about 100 European institutions will be included in the tests – four times the size of the original group – some bankers are confident that the expanded programme will reveal that much of the banking sector is healthier than investors think…

But big questions remain about how rigorous the expanded tests will be, particularly with respect to the sector’s exposure to Greek, Spanish and other eurozone sovereign debt

Institutions will be asked to disclose their total sovereign debt holdings, and the tests will now include a loss rate or so-called haircut of about 3 per cent on all eurozone sovereign debt investments, according to several sources.

Let’s see how this all stands up. Eurobank exposure to Club Med sovereign debt is roughly $900 billion (note this excludes debt to eastern Europe, another possible source of tsuris). A 3% loss on that is $27 billion.

Ahem, let’s take a more skeptical view. Eurobank holdings of Greece’s government debt is $190 billion. Williem Buiter, now chief economist at Citigroup and a bit of an expert on sovereign default, estimated the haircut on a Greek restructuring at 20% to 25%. But S&P later downgraded Greece, and remarked:

At the same time, we assigned a recovery rating of ’4′ to Greece’s debt issues, indicating our expectation of “average” (30%-50%) recovery for debtholders in the event of a debt restructuring or payment default. The ‘AAA’ transfer and convertibility assessment is unchanged.

Yves here. Do the math. Even if you assume the low end of Buiter’s now-charitable estimate, banks will take losses on Greece alone of $38 billion, 41% greater than the level provided for in the stress tests. If S&P is nearer to the mark, the losses will be $95 to $133 billion.

And the more Greece takes new loans before its debt is restructured (a restructuring or default looks inevitable; no country in the modern era has ever had this high a percentage of debt to GDP in a currency it does not control paid its creditors in full), the worse off the banks that hold debt now will be. The new loans will be senior to the current debt, which means the writedowns on the now-outstanding sovereign debt is likely to be high.

Analysts are discussing which banks will need to raise capital:

Friday’s edition of the Financial Times reported the expectation among bankers that the likes of Spain’s Banco Popular, and Monte dei Paschi and Banca Popolare di Milano in Italy were likely candidates for capital raisings.

All three said they had no capital-raising plans. Banco Popular said it was one of the best capitalised banks in Europe, with a core tier one capital ratio of 8.8 per cent.

BPM said its core tier one ratio was 7.9 per cent. Like tier one, core tier one ratio is a measure of capital strength.

At the same time, there were suggestions in Spain that policymakers were considering going further than counterparts elsewhere in Europe, increasing the required ratio for passing the tests in an effort to boost the confidence value of the exercise.

The FT took note of investor doubts:

News of the tests’ increased rigour has not fully eased concerns about transparency.

“The stress test idea is a shambles,” said one senior analyst in London.

“The whole thing is a complete joke.”

He said that the market’s expectations of securing meaningful disclosures through the tests were so low that any useful information would be a welcome surprise. “Ironically you might just get a boost if there are any decent disclosures at all,” he said.

Maybe the Europeans will pull a rabbit out of the hat, but the odds do not appear to favor them. John Gapper, in a comment on the stress tests, is doubtful:

Even some of those who in principle support the idea of banks being honest with investors are worried about the forthcoming European bank “stress tests” – successors to tests on US banks last year. “I have a horrible feeling that this will turn out to be an exercise in damaging confidence,” says one bank analyst….

The worst case is that southern European banks, loaded with bonds denominated in euros, will turn to governments for relief and trigger another sovereign debt crisis. Spain, which is trying to solve a crisis among its cajas – regional savings banks – is a potential victim….

he tests were certainly a turning point in confidence in Mr Geithner himself, who had suffered a rough few months in the job. Whether it turned the tide for banks is less clear; the US stimulus and other measures to restore consumer and business confidence were large factors.

In addition, European banks’ problems are more intractable and complex, and probably less amenable to a quick fix. For one thing, there was little question last year that the US could afford to rescue banks if it had to, whereas European governments are now heavily in debt.

Furthermore, European banks have inherent funding problems that their US and Asian counterparts lack – they are far more reliant on wholesale markets. US and Asian banks cover their loans with retail deposits, while Barclays Capital estimates the ratio of loans to deposits at European banks to be 120 per cent.

This leaves these banks vulnerable to a liquidity crisis…

Europeans do things in one way and Americans in another; Europe believes in discretion while the US likes openness even at the risk of embarrassment. We will soon find out if European banks can be salvaged by American methods and we had better hope so.

Precisely.