Are the Fed's New Bank Stress Tests Stressful Enough?

By Daniel Indiviglio

The Federal Reserve has completed another round of stress tests to determine the strength of the 19 largest U.S. banks. You might remember back in 2009, the Fed stress tests helped restore confidence in the financial market. Since that time, these big banks have mostly paid back their government bailout tabs, so whether they have adequate capital without the government money is an important question. Although the Fed hasn't made the results public yet, it did explain its methodology. Are the tests tough enough?

Here's a chart from the Fed summarizing the scenario it considered:

stress tests 2011-03.png

Really, only the first column matters here. According to this scenario, there's little stress tested in 2012 or 2013. This exposes a potential flaw in the tests: they are short-term in nature. They don't, for example, test for a prolonged contraction that could result from a double dip, or further dramatic home price deterioration.

But let's go through the statistics in the first column. First, it's hard to imagine that the economy would shrink by more than 1.5% this year. But that's not a very drastic assumption. After all, GDP shrunk by 2.6% in 2009. So these stress scenario doesn't appear to be testing for a truly stressful period in general, but one hitting in the near-term.

As for unemployment, this could be tricky. For example, although it has been declining recently, the number of workers who want jobs but can't get one has gradually risen. That's due in large part to their not being considered technically unemployed. You could argue that this more inclusive measure matters more exiting a recession with prolonged unemployment where more people temporarily leave the workforce due to frustration with the job hunt. If they're counted, then unemployment is actually around 12.6% currently. Suddenly 11% doesn't seem so stressful, even at this time.

The decline in home prices also seems more realistic than a doomsday scenario. Prices have already begun declining again and many estimates predict that they'll decrease by between 5% and 10% over the next two years. That's almost exactly what the stress tests imagine.

Finally, will equities decline much this year? They aren't expected to, but if Japan's disaster, European sovereign debt troubles, unrest in the Middle East, and out-of-control oil prices ignite a global double dip, then equities would certainly take a big hit.

So overall this stress scenario doesn't seem like that unrealistic of a possibility. But isn't the entire point of a stress test to imagine something beyond what even the grimmest of market-watchers believe could be possible? It also doesn't appear to take into consideration any big losses that banks could face from pending litigation related to mortgages and foreclosures.

The Fed does say that it did an additional test for the six largest banks, however. Here, it imagined a 2008-like global market shock that caused trading to seize and asset prices to plummet. That sounds like a little bit more more conservative of a test. But now that the government has forbidden future bailouts, a credit crunch could be even more dangerous.

To be sure, any bank that fails these stress tests should be forced to adhere to strict capital restraints. But the tests may not be aggressive enough to ensure the survival of the banks that pass if a serious economic catastrophe hits.

This article available online at:

http://www.theatlantic.com/business/archive/2011/03/are-the-feds-new-bank-stress-tests-stressful-enough/72697/