Thoughts on Stress Test Leaks

May. 7.09 | About: SPDR S&P (KBE)

From our vantage point it seem like amateur hour in analyzing the stress tests and the various leaks.

People who have little experience with government and policy-making are rushing to judgment about the process.

Let us consider a prominent example. David Wiedner at the WSJ writes as follows:

On the eve of the official announcement, the U.S. Treasury Department and the Federal Reserve are leaking stress-test results like pipes in a 100-year-old house.

Out of the tens of thousands of readers of his column, we are probably the only ones who clicked through to all of the sources and tried to determine what the journalists said. None of them are citing a "high government official" or a "well-placed government source." Quite the opposite, and no support for the allegation.

We had a similar experience today on RealMoney,'s site where we write. (Regular readers know that we subscribed for many years before becoming a contributor, and believe that the debate there has good value.)

Our colleagues at RealMoney were all pounding on the Administration for leaking, despite a complete lack of evidence on the source of the leaks. Here is what we wrote:

Jeff Miller
Source of Stress Test Leaks
5/6/2009 2:10 PM EDT
Click to enlarge

I am curious. Why is everyone so confident that the government is the source of leaks about stress test results? Where is the evidence for this?

In fact, banks were instructed not to discuss results during earnings. Part of the reasoning was that some would claim they were "OK" and then others would be suspect. Most of the media reports cite "unnamed people familiar with the situation." Let's face it, once a wider circle of people connected with a given bank have some facts, the probability of a leak increases. The nature of the process involved a discussion of results with the institution, and the development of a plan. This was not something that could be embargoed like CPI or employment numbers.

Some of the earlier alleged leaks were bogus stories, easily recognized by anyone familiar with government reports. Some of the others were speculations from private parties based upon the publicly known methodology and their own calculations.

Public officials have commented on the test only in general terms and emphasizing that the process is designed to find problems and fix them. That seems to be working. I suspect that the leaks are all from private sources, including some that were intentionally misleading.

Click to enlarge

and later.....

I am going to borrow a little from the fine work of Linda Lord of UBS who has demonstrated consistent excellence in covering this topic. She goes back to the original purpose of the test, inspired by Summers and Geithner, to provide some independent verification that the already existing conclusions of official regulators were correct. She draws from Bernanke to emphasize that these were not "pass/fail" exercises. She points out that the conversion of government preferred to common could happen if and when the possible losses actually occur

This is all very reasonable, but it does not mean the stress test idea was perfect. The name was bad, especially given the assumptions. They originally expected to keep the results private, as is the norm in bank regulation. Some bank accountants felt that these results were "material" information and therefore reportable. So the Administration chose not to fight it.

The tests were not expected to show that banks were insolvent. Regulators had already reviewed that using measures like tier 1 capital. But bank bears have emphasized Tangible Common Equity, so the test took a look at that. The standard was to be "well capitalized" even if the economy got worse. The methods were publicized.

This is an open and healthy process, and I invite anyone to suggest one that would have been better. At some point, the government had to discuss the results with the banks. The idea was to discover any errors before announcing results and also to consider possible plans.

The real source of the problem is the media -- all looking for a big scoop -- and the thousands of bloggers willing to run with any information. As you read these stories, look to see how many times the journalist cites "a well-placed government official" or the like I haven't seen any.

I do not mean this in a partisan way, but I am surprised by the anti-government sentiment displayed by so many. The President, whether Bush or Obama, cannot control what various strategists say, or what one of the 535 members of Congress might say. None of these people had data anyway.

To summarize, most of the media criticisms stem from false ideas about what the process was supposed to do, possibly falsified documents, speculation derived from guesses about the methodology, and leads from private "well-placed sources." There was no reason for a deliberate Obama attempt to parcel out information, and no sound reason to suspect that the Administration is the source.

Our position on RealMoney had zero support from colleagues, but no real refutation. A continuing and important theme here is separating one's political opinions from investment decisions.

We see no basis for the widespread criticism of the Administration with respect to these leaks. The only comment yesterday was from Sheila Bair, in a direct response from a Congressman in a hearing. The response was very general in nature -- nothing specific about individual banks.

We endorse this interpretation from John Carney:

Will anyone investigate these leaks? We'd guess they're coming from the banks themselves. There's at least the potential for serious harm to investors occurring as news spreads unevenly, first through proprietary news services such as Bloomberg and then across the wider media.

This is entirely consistent with our concerns about potential manipulation from blogs and (unwittingly) MSM sources racing for a scoop.

Our Take

We expect the stress test results to provide valuable information and a path toward restoring confidence. The Administration has made a bow to the bearish critics by including a Tangible Common Equity test, even though that is not strictly relevant for normal solvency questions.

We further expect that the tests will look forward a bit, considering the strong earnings prospects given current interest rate differentials.

Disclosure-- as we noted in comments, we are long GS in personal and client accounts.