Monday, November 26, 2007

Three principles for avoiding future subprime messes

Changes in policy should accomplish three things:

(1) It should make sure that all parties in the lending chain have “skin in the game.” While reputational risk mitigates against bad behavior, there is no substitute for financial incentives.

(2) It should make sure that all parties in the lending chain are subject to federal supervision. This will reduce regulatory arbitrage.

(3) It should do what it can to improve disclosures throughout the lending chain. Borrowers must be better informed as to the consequences of their lending choices (although this will be difficult); ratings must be consistent, and securities must be more transparent.

4 comments:

Austin Kelly said...

Richard,

I noticed your first principle "Skin in the Game" and wondered if you had seen this paper of mine. Its main result is that borrowers are much less likely to default if they have "skin in the game" - that there is a large difference between borrowers with a very small down payment and borrowers with no down payment at all, way beyond the effect you'd get just from extrapolating an LTV effect.

"Zero Down Payment Mortgage Default" available at
http://mpra.ub.uni-muenchen.de/5370/

Bruce Wilder said...

For "transparency" to be meaningful, there really need to be crisp, clear recourse options that describe the path to working-out a default. Can the so-called "homeowner" walk away? Who handles the foreclosure? Can anyone re-negotiate terms?

I don't know how to phrase it, but I think you might want to add something about agency and recourse

Richard K. Green said...

Thanks, Austin. I haven't read the paper, but I will. The history of the HUD 235 program also gave us substantial evidence that people who buy without putting any money down default in large numbers.

Austin Kelly said...

I thought a down payment was required in 235. But borrowers got an interest subsidy that was subject to recapture, which probably pushed default rates through the roof, although I've never seen a statistical study of the 235 program. Wasn't 235 more a demonstration of the riskiness of Neg Am rather than a demonstration of the risk of Zero Down? An interest subsidy subject to recapture looks an awful lot like a Neg Am.