Everyone Agrees!

That I'm wrong.

Over the last couple of days I've been harping on the fact that the popular explanation for why the financial world imploded -- that mortgage lenders started lending to anybody with a pulse -- is probably wrong. But in this post I'd like to highlight some people who disagree with me. 

I'll go over my argument again here very briefly for context, but if you're familiar with it then skip down. For something more in-depth, here are posts 1, 2, and 3.

I based the posts on two papers. The first is from economists affiliated with St. Louis Fed who argued that previous research which found that standards had loosened was incomplete. The problem, the Fed paper said, was that a lending decision is multi-dimensional, meaning just because the number of no-doc subprime loans increased doesn't by itself mean that standards declined. Lenders have more than one way to evaluate a borrower and without looking at all of the dimensions -- or at least coming up with good explanations for why you don't need to -- you can't say much conclusively about the direction of standards.

The researchers then show that, in fact, FICO scores increased in recent years for subprime borrowers, suggesting that lenders were compensating for the added risk of greater no-doc loans.

The second paper was by Yale's Gary Gorton who gives a number of anecdotal, but reasonably convincing, reasons for why lender incentives to scrutinize borrowers wasn't broken because of subprime securitization. I go into these here. That's not to say there wasn't something wrong down the line with subprime securitization process however.

Now, here's the feedback. First up is an email from Barry Ritholtz who believes that poor lending practices led to the meltdown:

I've been using this line of discussion for quite some time now -- and I simply don't buy Gorton's meticulously made, but erroneous arguments.

To be blunt, they struck me as ridiculous -- house prices fell, and the securities and derivatives built on top of them also fell.  
(In related news, the dinosaurs are now extinct).

I've had first hand experience with the mortgage business, via clients  -- and the no income, no credit check loans alone make the professors arguments fail.  The sole reason to make a loan where you willfully avoid learning ANYTHING about the creditworthiness of the borrower is to resell it.  Period.

Then there is res ipsa loquitor -- the net results speak for themselves
-300 mortgage originators went bust.
-foreclosure rates have risen dramatically
-losses to the buyers of mortgage based paper are nearing trillions.

Next up is Felix who argues that it's the small lenders fault:

Essentially what happened is that many businessmen (and for some reason they were nearly always men) founded companies which would make them millions of dollars a year for some unknown period of time. When the good times ended, they ended -- but they were fun and profitable while they lasted.

Now consider the position of a much bigger company like Countrywide, faced with all this new competition. In order to maintain market share and profitability, it has to adopt the same business model of the fly-by-night operators. But the difference is that Countrywide did have assets and did have equity, and therefore was at risk of suffering the substantial losses which eventually transpired.

Eventually, big banks started wanting to get in on the game as well -- hence such ill-fated deals as Wachovia's acquisition of Golden West, which will go down in history as a classic case of suicide-by-acquisition. But all of these deals can be considered a special case of Gresham's Law, which states that bad money drives out good. The banks were using good money, the fly-by-nights were using bad money, and the bad money, as it always does, won.

And the bad money was precisely the money which relied for its existence on the originate-to-distribute model.

So if you're looking for misaligned incentives, don't look at Wachovia or Countrywide. Look instead at Merit, Ownit, and their ilk. If you still think that incentives are aligned, then I'll be much more convinced.

Finally, here is a comment left by BusinessWeek's Aaron Pressman on a previous post:

Tried to get at this in our FICO story in February. The increasing reliance on FICO, as a way to compensate for other risk factors going up, was fatally flawed.

Key bits: FICO score became increasingly easy to game. FICO score became less reliable measure of future payment likelihood.

Here is Pressman's story on the declining utility of FICO.

I'll have more on this later, but I'll leave you with a few questions. This 2006 paper by Souphala Chomsisengphet of the Office of the Comptroller of the Currency and Anthony Pennington-Cross of the St. Louis Fed found that as

the subprime market has evolved over the past decade, it has experienced two distinct periods. The first period, from the mid-1990s through 1998-99, is characterized by rapid growth, with much of the growth in the most-risky segments of the market...In the second period, 2000 through 2004, volume again grew rapidly as the market became increasingly dominated by the least-risky loan classification.

What changed after the end-of-2004/beginning-of-2005 which could lead originators to lower their practices in such a way that would cause the carnage we've seen since August '07?

And why is it wrong to think that the thing that did change materially was that housing prices fell nationwide for the first-time in decades? Subprime borrowers needed home prices to go up in order to be able to refinance (either into another subprime or prime loan), otherwise they would have to default -- there was no way they would be able to pay their mortgages once the ARM's reset. And since subprime securities were so opaquely designed and wrongly rated, once the defaults started few were prepared, and nobody knew who was going to get hit.

---