WASHINGTON - The federal government's so-called stress tests to determine if 19 major banks could survive a further downturn in the economy may have relied on too rosy a scenario and should be repeated, independent investigators said in a report released yesterday.
The Congressional Oversight Panel for the government's $700-billion financial rescue effort found that the Federal Reserve used a "conservative and reasonable" approach to assess the health of the nation's 19 largest banks.
But, the panel added, the Fed's worst-case scenario did not go far enough. For example, the stress tests conducted by the Fed in March and April were based on a projected 2009 unemployment rate averaging of 8.9 percent. Unemployment in May climbed to 9.4 percent.
As a result of the tests, 10 of the banks were told to raise a total of $75 billion to withstand losses if the recession were to worsen. The other banks were found to be stable.
"While no one should gainsay the potentially positive results of the tests, it would be equally unwise to think that those results reflect a diagnosis of all of the potential weaknesses or create a necessarily sufficient buffer against future reverses for the banking system," the panel wrote in yesterday's report.
Separately yesterday, 10 of the large banks were given the green light by the Treasury Department to repay $68 billion in government bailout money, freeing them from restrictions on executive compensation they say are making it hard to keep their top-performing executives.
Industry experts said the move shows some stability has returned to the system, but they cautioned that the crisis wasn't over.
Asked about the findings of the oversight board, Treasury Secretary Timothy Geithner defended the stress tests, saying they were rigorous and used projected loss estimates for the banks that were worse than any two-year period during the Great Depression.
Elizabeth Warren, the Harvard University law professor who heads the panel, told lawmakers yesterday the Fed should release more details about how it conducted the tests.
"Without this information, it is not possible for anyone to replicate the tests to determine how robust they are or to vary the assumptions to see whether different projections might yield very different results," Warren told the Joint Economic Committee.