≡ Menu

Brad Delong’s analysis of the housing crisis

In this post, Delong comments on debate between Joe Nocera and Peter Wallison and then sums up his view of Fannie and Freddie and the crisis:

When the housing bubble took off–and when mortgage loans and MBSs became really risky–that’s when Fannie Mae and Freddie Mac pulled back. The GSEs were dominant players in the market up through the end of 2002: most of the loans made through the end of 2002 are still good. The GSEs were small players by the start of 2004–which is when the mortgages that have turned out to be toxic assets started to come through.

This is the counter-argument to the view that Fannie and Freddie caused everything. Well, they didn’t. But that doesn’t mean they didn’t contribute to the crisis in a significant way.

First, Delong makes it sound like the bubble started in 2002 or maybe 2004. Here is the Case-Shiller housing price indices for the nation and for the ten largest cities:

The boom in housing prices began in the mid to late 1990s. It does accelerate sometime in the early 2000s.

Next, let’s ask the question, when did loans start getting really risky. In 2001, Josh Rosner chronicles the deterioration in underwriting standards in this paper. Rosner notes (writing in 2001) that Fannie Mae had begun a push toward low downpayment loans, away from 20% down and was beginning to buy loans with 5% down and 3% down. Here are the data:

The chart, from my paper, shows Fannie and Freddie’s purchases of owner-occupied loans where the loans have less than 5% down. Having Fannie and Freddie purchase these loans means that a lot more people are going to be able to buy houses than before. It also means they are willing to pay more for a given house. This helped create the housing bubble. In 2004, Fannie and Freddie did “pull back” as Delong mentions and as Krugman has claimed as well, in the sense that they did buy fewer low downpayment mortgages in 2004 relative to 2003. But it was still about triple what they bought in 1999. And the number rises steadily between 2004 and 2007. Not really a pull back.

Delong says Fannie and Freddie were “small players by 2004.” No. They weren’t. Here is the picture of Mark Thoma’s that Krugman cites to show that “Fannie and Freddie pulled back sharply after 2003.”

It’s a busy picture but it shows that Fannie and Freddie’s market share of all mortgage originations grew pretty steadily from 1973 to 2000 or so, peaking at just about 50%. Then it fell after their accounting scandals and as Wall Street began securitizing subprime mortgages. Fannie and Freddie bought significant amounts of Wall Street MBS, helping to create the demand for those assets. But they weren’t the whole story. Not even close. Wall Street investment banks played a crucial role in the subprime crisis. Fannie and Freddie’s role was helping to create the bubble that Wall Street tried to exploit. Still, defenders of self-regulation have to explain Wall Street’s frenzy that was independent of Fannie and Freddie.

My explanation is that past bailouts of creditors allowed Wall Street to use excessive leverage to buy toxic assets. Previous bailouts reduced prudence. I tell that story in detail here. But my point here is that Fannie and Freddie were not small players in the mortgage market as Delong claims. Even with their loss in market share, they were the dominant player. They were not the dominant player in subprime. But they did contribute to the housing bubble and allowed lots of dangerous prime loans to be made by buying loans with little money down.

Delong does have a chart to support his case. It is a mystery chart with data from March ’98 to March 2008:

I’m sorry it’s so hard to read. But it purports to show that Fannie and Freddie, the GSE’s shown as the green line were unimportant after 2004 and their activity only surged in 2007. I don’t know what to make of this chart. The title is “Who makes the mortgage loans? (the short view)” Now the GSE’s don’t make loans, they buy them or guarantee them. So I suppose it is supposed to really mean who financed the mortgage loans. The lines in the chart are supposed to represent “Share of net home mortgage lending (percent)” yet the lines go from -100 up to 200. How does one’s share of lending become negative? What does the word “net” mean in the legend? Is it the share of new home lending. I don’t know.

It took me a while to find where Delong got the chart. I think he got it from this blog post (from July of 2008) at Time magazine. There you’ll see the exact same chart along with another one called “Who makes the mortgage loans? (the long view)” with a longer period of time for the data. What I think the chart actually shows is the change (not the share) in lending activity by the various groups or maybe the change in market share.So when the green line goes down to zero in the 2003-2004 period, it’s because Fannie and Freddie are reducing their activity. But they don’t have zero market share as Delong is implying.

Now maybe the chart shows something else entirely that I’m missing. I look forward to hearing from Delong. I’m sure he has a good explanation. But I’d stick with the Thoma chart I showed above that Krugman cites. It shows that yes, Fannie and Freddie lost market share in the 2003-2004 period but they were still the dominant financer of mortgages even after they “pulled back.” And yes, their share fell, but they were still aggressively expanding into risky low downpayment loans during that period and that helped inflate the bubble.


Next post:

Previous post: