Five Unasked Questions About the Stress Tests
1. Why will these (weak) stress tests lead to more realistic evaluations than the (far tougher) stress tests that Congress mandated for Fannie Mae and Freddie Mac?
Congress mandated a purportedly “stringent” stress test for Fannie Mae and Freddie Mac over a decade ago. It required them to have adequate capital to withstand the simultaneous onslaught of severe credit, interest rate and operational risks that continued for 10 years. The current Treasury test concentrates solely on credit risk and assumes it ends after two years. How well did the far more stringent Fannie and Freddie stress tests work? In August 2008, Freddie reported that “even [our] most severe stress tests [show] losses … less than $5 billion.” It failed in September. Actual losses: 20 to 40 times greater.
2. Where else were stress tests used?
Stress tests were used for the Rating Agencies, IndyMac, and AIG. The Rating Agencies’ stress tests gave AAA ratings to toxic waste. Actual losses: more than an order of magnitude greater than those predicted by the stress tests. IndyMac sold over $200 billion of “liar’s loans.” Actual losses: 160 times greater than its tests. AIG (2007): “It is hard for us, without being flippant, to even see a scenario within any kind of realm of reason that would see us losing one dollar in any of those [CDS] transactions.” AIG (2008): “Using a severe stress test … losses could go as high as $900 million.” Actual losses: 200 times greater.
3. When did Geithner begin to claim that stress tests were the keys to safe operation?
As president of the Federal Reserve Bank of New York, in a speech in 2004, he first praised stress tests. He was the principal regulator of many of the largest bank holding companies in the U.S. Every large bank has long used stress tests – and Geithner’s Federal Reserve examiners reviewed their stress tests. The big banks’ stress tests on nonprime loans and derivatives failed, and the Federal Reserve examiners consistently failed to understand the failures.
4. How can you conduct a stress test without reviewing the bad mortgage assets’ (missing) underlying loan files?
A Standard & Poor’s (S&P) memorandum recently unearthed reveals the sad truth about how non-prime collateralized debt obligations (CDOs) were purchased, pooled, rated and sold: ”Any request for loan level tapes is TOTALLY UNREASONABLE!!!. … Most investors don’t have it and can’t provide it. … we MUST produce a credit estimate. … It is your responsibility to provide those credit estimates and your responsibility to devise some method for doing so.” The email message is from a senior S&P manager to the professional rater. The word “investors” means the entity that created the CDOs. One cannot evaluate loan quality or losses accurately without reviewing a significant sample of the underlying loan files. The banks and the regulators virtually never do this. They did not do this during the stress tests. They do not even have access to the files that they need to review.
5. How can you ignore fraud losses during an “epidemic” of mortgage fraud?
The FBI began testifying publicly in September 2004 about the “epidemic” of mortgage fraud. It has also stated that lending insiders participated in 80 percent of mortgage fraud losses. The presence of massive fraud losses means primarily produced by lenders makes it absurd to rely (as Treasury did in its stress tests) on the lenders’ loss evaluations. Stress tests produce fictional results that massively understate real losses and produce complacency.