The stress test charade

May 11, 2009

Petrino DiLeo examines the Obama administration's "stress tests" of major banks and shows how they were rigged to provide a positive result.

WITH GREAT fanfare, the federal government late last week announced the results of its "stress tests" of 19 major banks.

The tests--designed to see how much capital banks might need if certain "worst case scenarios" played out--concluded that the banks collectively would need $74.6 billion in fresh capital to survive $599.2 billion in losses. That's a lot of money. But every single bank, even the worst off, was judged to be solid, provided it raised the extra money.

The announcement came with pronouncements of success from the Obama administration, and was roundly cheered by the mainstream media. "The results released today should provide considerable comfort to investors and the public," Fed Chair Ben Bernanke said in a statement. Edward Yingling, president of the American Bankers Association declared, "The results of the stress tests should put to rest the harmful speculation we have seen over the past few months."

Bank of America was judged to be in need of the most money--$33.9 billion in additional capital. Wells Fargo and Citigroup were next, at $13.7 billion and $5.5 billion respectively. Seven other banks need between $600 million and $2.5 billion while nine institutions were deemed to not need any additional funds.

Stress tests revealed that if the crisis persists two more years, Bank of America would need $33.9 billion to stay afloat
Stress tests revealed that if the crisis persists two more years, Bank of America would need $33.9 billion to stay afloat (Gerald Carter)

But there are a few of problems with the triumphant scenario projected by the Treasury Department and its cheerleaders in the media.

First, the stress tests weren't all that stressful--they didn't allow for the real scale of possible losses to come, say analysts. Second, the discussion of the results neatly skirts the issue of how exactly banks are supposed to come up $74.6 billion--the money is supposed to come from private markets, but no one is yet buying banks' bad assets.

Third, the banks won big concessions from the government on what figures were finally reported. And last, the government is imposing a lenient condition on banks by allowing a Tier 1 capital ratio of 4 percent--which means banks could still have 25 times more money committed than assets to cover potential losses, which hardly provides much of a cushion for future crises.

THE GAPING holes in the tests led even the Wall Street Journal to give the whole process a "B-minus." The paper pointed out, for example, that the stress tests were far from rigorous in their assumptions of losses from commercial real estate and too generous on earnings projections.

William Black, a former senior bank regulator and now an academic, went even further, calling the stress tests "a complete sham" before the results were announced.

The scenarios used by the Treasury Department made a number of assumptions about stabilization or even improvement in the economy. In other words, the purpose of the stress tests seemed to be to prove the banks were solvent--rather than provide an accurate picture of how banks would fare if further losses and writedowns materialize.

For example, the Treasury Department's baseline scenario--i.e., what it expects to happen--assumes an 8.4 percent unemployment rate for 2009 and 8.8 percent for 2010. Its "negative" scenario assumed a year-round average in unemployment of 8.9 percent for 2009 and 10.3 percent in 2010.

But the latest numbers from the Bureau of Labor Statistics show that unemployment had already reached 8.9 percent in April. While the pace of job losses slowed slightly in April from previous months, the economy is still losing 500,000 jobs a month--a pace that would push unemployment past 10 percent before the end of the year. In other words, the Treasury Department's assumptions are far from a worst-case scenario.

As left-wing economist Doug Henwood explained in an interview with

They're trying to reassure people, and they have to give the appearance of credibility in their reassurances. But for the assumptions they've used, the worst-case scenario is unemployment going a little over 10 percent. That seems to be the likeliest scenario, and the worst-case scenario would be considerably worse than that--more like 15 percent unemployment.

But if they used that, they would have gaping holes in the balance sheets of the biggest banks. So they have to cook the thing. I think the whole thing was pre-cooked from the start.

The stress tests also assumed that house prices would fall 14 percent in 2009; the negative scenario was a decline of 22 percent. However, housing prices have been falling faster than that in recent months--at around 24 percent when projected over a whole year. Again, the stress tests assumed improvement, rather than attempting to see how banks would fare in current conditions or in an economy that continues to deteriorate.

Assuming greater job losses and steeper housing declines would reveal that banks actually need quite a bit more capital. Center for Economic Policy Research Co-Director Dean Baker estimates the banks might actually need $195 billion--well over twice what the Treasury Department concluded.

In addition, the banks got the government to lower its estimates from initial findings of how much new capital is needed. The Wall Street Journal reported that Bank of America's number was lowered from $50 billion to $33.9 billion. Wells Fargo's capital hole shrank from $17.3 billion to $13.7 billion--and Citigroup's shortfall was reduced from $35 billion all the way down to $5.5 billion after executives persuaded the Fed to include estimates of the future effects of pending transactions.

The stress tests reveal the same pattern that has characterized the Obama administration's handling of the financial crisis since taking office--the big banks dictate the terms of the "rescue," and the federal government, led by veteran Wall Street insiders like Lawrence Summers and Tim Geithner, makes sure corporate interests come first.

WHAT HAPPENS now? On June 8, the banks will have to present their plans for getting fresh capital from new stock, asset sales or more government involvement.

However, they may not have to raise anything close to $74.6 billion. Instead, regulators will allow banks to bolster their balance sheets by converting preferred shares to common shares. With this shell game, the banks won't actually have to raise money, but just move some around--an accounting trick.

In addition, the Financial Times reported that banks have been given government assurances they will be allowed to raise less than the $74.6 billion if earnings outstrip forecasts over the next six months. That agreement was left out of Treasury Secretary Timothy Geithner's statement on the results. It means banks have an incentive to fiddle with their books some more to make profits look better in the next six months.

In the end, the stress tests were nothing more than an elaborate public relations stunt, meant to increase confidence in a financial system that remains mired in a deep crisis. The stress tests do nothing to change the reality--that the banks should be nationalized.

Further Reading

From the archives