The Fed released a 21 page report on its stress test methodology. It was so vague that there'll be no help for investors who want to try to figure out which of the 19 banks undergoing the test will need to raise capital and which won't. The focus of these tests is to measure how much common equity banks have. But we already knew that.
The Fed is overseeing 150 banking regulators who asked banks to estimate their loan losses based on forecasts of broad economic statistics -- for example, unemployment rising to 10.3 percent by next year, home prices falling an additional 22 percent this year, and the economy contracting by 3.3 percent this year and staying flat in 2010.
The banks have to estimate their losses for each loan category and compare the loss ratios to the Fed's "high" and "low" estimates. If a bank expected fewer losses than the government did, the regulators asked the institution to explain why. The Fed also required banks to forecast two years' worth of earnings so regulators could estimate how much capital they'd have to absorb losses.
I'd guess that each of the 19 banks has its own models for estimating those loan losses and future earnings and that the Fed does not understand the differences between them. Thus it has no way of judging which bank is blowing smoke and which is being conservative. As I posted, that's a big reason why I think the Fed should can the stress tests and require banks to produce viability tests instead.
Peter Cohan is president of Peter S. Cohan & Associates. He also teaches management at Babson College. His eighth book is You Can't Order Change: Lessons from Jim McNerney's Turnaround at Boeing.