Friday, October 24, 2008

 

Fed Credit Charts as Fine Art

For those who have long suspected that economics is not a science, I am ashamed to report that several of us PhDs can't even agree on whether there was a "credit crunch" during the first half of the year.

Exhibit A is the Mark Thoma/Alex Tabarrok feud. They look at the exact same charts and conclude opposite things. (Here is Alex responding to the Thoma link, so you can see what I mean.)

For my part, I side with Alex. Below I reproduce just two of the charts Thoma shows, but the others all show a similar pattern. During the dark days of the credit crunch, when we were warned the entire financial system would collapse if we didn't cough up $700 billion, the supply of "industrial and commercial loans at all commercial banks" was at an all-time high, and in fact its year/year rate of growth at its lowest point was "only" 12% or so. If you exclude the last few years--which everyone agrees was an insane period with perhaps criminally lax lending standards--then the year/year rate of growth in these loans was higher than at any point going back to the early 1980s. Really, this is amazing, given all the fear-mongering we've been hearing. Just study the picture below for a few moments and let it sink it. (Remember, this is is year/year growth; look at the units on the left axis.)



But here's my favorite. Even if we look at "real estate loans at all commercial banks," we see that the year/year growth rate was always positive, and in fact stayed above 4% the entire time (I'm just eyeballing the chart, it's probably close to 5%).

That's right folks, during the "credit crunch" fueled by the subprime crash etc., the volume of real estate loans made by commercial banks only experienced a significant slowdown in its rate of growth. Sounds like something worthy of bank nationalization to me.




Good job on all this, Alex. Fight the power.

One closing caveat: If you dabble with these series at the Fed site (and it's very convenient to do so--you might as well get something for the depreciating dollar), you will see that in some cases, there was a drop in the level starting in 2008. But still, the charts above show that there was still year/year growth. I.e. even though some credit items fell from, say, December 2007 to March 2008, those items were still higher than they had been in March 2007. If one wants to argue that the economy was slipping off of a cliff, and Paulson needed to save us, OK that's consistent with the charts. But it is NOT true that we were already in a crisis (vis-a-vis the entire scope of the various credit markets) when Paulson and Bernanke had no choice but to go for the Fidel Castro option.



Comments:
These charts should go to MoMA. An exhibit called the "dismal science" would have perfect timing.
 
Post a Comment

Subscribe to Post Comments [Atom]





<< Home

This page is powered by Blogger. Isn't yours?

Subscribe to Posts [Atom]