Stress tests under stress
David Wessel has an interesting idea today: while the bank stress tests were a good idea at the time they were announced, in February, a lot has changed since then, and none of it in a good way. Most obviously, the tests’ worst-case scenario is now looking more like a base-case scenario, making the tests less credible. What’s more, the very announcement of the stress tests’ existence caused all manner of confusion and second-guessing in the markets, none of which was helpful. And most profoundly, Congress passed executive-compensation rules which mean that no banks have any interest in accepting government funds should they be found to have insufficient capital.
For me, the fact that all these things managed to happen so quickly after the stress tests were announced is an indication that the stress tests probably weren’t such a good idea in February after all. But never mind that: as Wessel says, “Treasury has to deal with the world as it is, not as it hoped it would be”. And that means being extremely transparent about both the tests themselves and their results.
Near the beginning of the crisis, in the early months of 2008, it was still possible for Treasury to attempt a “trust us, we know what we’re doing” approach to bank regulation. That won’t fly any more. Treasury has to internalize the show-don’t-tell rule which is commonly hammered into journalists. Because no, we don’t trust them to know what they’re doing. Especially when the official org chart still looks like this.