Arnold Kling  

The Crisis Response Worked Wonderfully

The Behavioral Effects of "Cat... Would Posner and Weyl Ban Gamb...

So says the Treasury. Below is the chart that proclaims that lending markets were unclogged.

Pointer from the indispensable Mark Thoma. A few curmudgeonly remarks:

1. The indicator is the percentage of banks reporting easing or tightening of lending standards. I do not believe that this is the typical indicator. I understand that using actual lending as an indicator has the disadvantage that it can be affected by demand for loans as well as whether conditions are "clogged." But, actually, that is relevant. If lending is down because of demand, then "unclogging" the financial system was not the answer, was it?

2. Even using this indicator, you have to look at the shape of the chart, not its content, to see that the crisis response worked. I mean, the shape is a nice "U" with a bottom at the point where the interventions began, so isn't that wonderful? Except that when you look at the Y-axis, you see that it shows that the indicator was below zero for close to 18 months after the intervention, which means that more banks were tightening than loosening standards during that time. The chart is designed to make you focus on the second derivative. If instead it were to focus attention on the first derivative by plotting the cumulative percent of banks that had tightened credit standards since, say, the end of 2006, it would show that the bottom of the crisis was reached in the latter part of 2010, not when the massive interventions took place.

By the way, I am not saying that this is the most deceptive chart in the package (look at the others). But it is the only chart that touches on the central goal of TARP, which was to "unclog" the financial system. If this is the best they can do, then I think the question of whether TARP and the bailouts did any good is still an open one.

Comments and Sharing

COMMENTS (5 to date)
Jeff writes:

Easening/tightening lending standards strikes me as a really awful measure of whether the Fed/Treasury interventions "worked." The crisis was triggered by lending standards being too loose! They needed to tighten! Maybe they tightened too much, but who's to say? If you were a highly leveraged financial institution back in 2009 and most of your capital cushion had been wiped out in the previous two years, how aggressive should you have been in the mortgage or commercial lending market? I don't know, and I'm betting the Treasury doesn't have a much more informed opinion than I do.

But this is all beside the point. At the end of the day, what this amounts to is the Treasury saying "we decided to evaluate our own performance during the financial crisis, and what do you know? In our opinion, we did a great job!" How 'bout that?

Trespassers W writes:

When I first read your post, I thought you must be interpreting the graph wrong.

Then I looked at the original document, and realized that you have it exactly right.

What if the line had jumped right back up to zero after the interventions? This graph would make the interventions look like a miracle. But really, it would mean that lenders had frozen their lending standards at some fairly restrictive level. Maybe it will start coming back down now.

The reality was much worse -- lending standards continued to tighten for years, as you say. This is clear if you look at something like the average FICO of new mortgage originations (something closer to a level of tightness of underwriting, rather than the first derivative, which is what the Fed surveys and the Treasury plots here). Average FICO was relatively low before the crisis, increased through the crisis, and has for the moment plateaued at a high level.

This is shameless deception by the Treasury.

Trespassers W writes:

Oops -- this line

Maybe it will start coming back down now.

should have come after this one

Average FICO was relatively low before the crisis, increased through the crisis, and has for the moment plateaued at a high level.
Jack writes:

Post hoc ergo propter hoc, as usual. The claim that the crisis response ''worked'' is nonsensical without some kind of benchmark to compare it with (even if not fully experimental).

E. J. Miller writes:

[Comment removed pending confirmation of email address and for multiple policy violations. Email the to request restoring your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Comments for this entry have been closed
Return to top