Why didn’t the WSJ stress-test the big banks?

By Felix Salmon
May 19, 2009

The WSJ has a very pretty interactive graphic, showing what’s likely to happen to the capital of 940 small and medium-sized banks if the “adverse” scenario in the government’s stress test comes to pass. Essentially, using public information, the Journal tried to replicate the stress tests across the industry as a whole.

I’m disappointed by this, however:

The 19 big banks that underwent a Fed stress test weren’t included in the Journal’s calculations….

The calculations don’t reflect any efforts made by individual banks since the start of this year to shore up their capital, such as shedding assets or cutting costs.

Obviously an exercise in crunching the numbers of 940 banks using public information is never going to be as detailed or granular as the stress tests were. But the stress tests put great store in efforts to shore up capital, and it’s entirely possible that those efforts might have made a significant difference. On the other hand, it’s possible they wouldn’t.

So I would really have liked it had the WSJ decided to increase the number of banks they did this exercise on from 940 to 959. It wouldn’t have been all that much in the way of extra work, and then we could see very clearly how the WSJ’s approximation of the stress tests compares to the real thing. Does the WSJ methodology produce similar numbers to the actual stress tests or not? If they printed the results from the 19 big banks, we could see for ourselves. Instead, they made a conscious decision not to run the numbers on the 19 big banks, thereby making it much harder to tell how reliable the numbers are. Weird.

Update: DollarEd puts it more sharply:

Obviously, the WSJ had sufficient data and a sound enough methodology that they were confident enough to apply it against 940 banks and publish the results. So, it would be very valuable, both to validate the WSJ’s work and to learn more about the Big Banks, to apply it to those 19 Big Banks. After all, they comprise 58% or something like that of the banking sector, so to not extend the coverage of the article really shifts the article from National News to some sort of stock picking (or dumping) exercise.

DollarEd reckons this is a function of the WSJ being essentially captured by the big banks. Is there another explanation?

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
Comments
5 comments so far

(assuming a warm, chiding tone)

Felix, you are trying to say something, but your dry and indirect style only serves to prove the point.

Here’s the point: there is a reality distortion zone around the big banks. They spend so much money on PR and lobbying that they distort their own regulatory regime, they distort political calculations, and they intimidate journalistic organizations into not reporting accurately on them.

Obviously, the WSJ had sufficient data and a sound enough methodology that they were confident enough to apply it against 940 banks and publish the results. So, it would be very valuable, both to validate the WSJ’s work and to learn more about the Big Banks, to apply it to those 19 Big Banks. After all, they comprise 58% or something like that of the banking sector, so to not extend the coverage of the article really shifts the article from National News to some sort of stock picking (or dumping) exercise.

And why not cover the Big Banks? Reality distortion zone – endless calls from PR flacks, threats to cut off access, calls from CEOs to Rupert Murdoch.

And you, Felix, what do you do? A dry, indirect questioning tone, and then on to the next subject. And that’s not in your best interests. You are a superb writer – you could have summed up the cowardice of the WSJ and the clout of the Big Banks in two sentences, and Eschaton and Yglesias and Calculated Risk would have linked to your sharp, accurate quip, and traffic would have flowed to you.

So why not? Are you in the Zone?

Posted by DollarEd | Report as abusive

Felix, i think the WSJ methodology produces similar numbers to the actual stress tests. It is difficult to tell whether the number are reliable or not but i think these are reliable. But I agree with you that they should have printed the results so we could see for ourselves.

And then we could see very clearly how the WSJ’s approximation of the stress tests compares to the real thing.

That would be the point, would it not? The stress tests rely on so many variables and assumptions, that it’s almost inconceivable that the two tests would produce similar results. In the minds of most of the readers, that would invalidate the results of the stress test that the WSJ spent so much time and money to produce. (Most will assume the government got it ‘right’.)

Of course, what it *should* probably do is invalidate the assumption that you *can* do modeling of a system with as many uncontrolled and unknown variables and expect it to produce highly meaningful results, whether it be the government or anyone else performing the modeling.

Posted by Tom West | Report as abusive

Consumers in America need to realize they have the ultimate power— The power of CHOICE.

We have a CHOICE as to where we will put our hard earned money. Are you going to put that money with a big bank, who is sucking American tax dollars DRY??

If you are, you are a FOOL.

Boycott the big, corrupt, lying institutions America.

CHOOSE a better way.

Posted by JanB | Report as abusive

Felix:

Regulators had 157 full time examiners going over the 19 largest banks for weeks. The WSJ probably had two guys with an excel spreadsheet try to put together their methodology in about 2 days.

Maybe there is a better reason than the grassy knoll for the wsj to punt on the largest 19. Cost vs benefit for a story.

Regards

Posted by Jeffrey Levin | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/