Skip to main content

Will Paulson's Proposal Revitalize Subprime Lending?

  • Author:
  • Updated:

It's often said that regulators are always solving the last problem while the next one creeps up on the markets unaware. That notion was certainly tempting yesterday as everyone who's anyone in market regulation gathered at that conference near Washington, DC. But we couldn't help thinking that despite speeches from the Fed's Ben Bernanke, Treasury's Hank Paulson and JP Morgan's Jamie Dimon, the folks there didn't even get as far as addressing what led us to the current financial crisis.
Let's take Paulson's proposal that covered bonds could restart the mortgage market. Ensuring that mortgage originators and those securitizing the problems have ongoing skin in the game may in fact improve the quality of those bonds. But Paulson made his proposal in the context of talking about reviving loans for subprime market. Will covered bonds restart interest there?
Not very likely.

Over in Europe, they've had a lot more experience with covered bonds. As it turns out, they aren't a very effective way of securitizing subprime. Why not? Because subprime loans are inherently risky. You know a certain amount of them are going to default. In fact, the securitized subprime was designed to accommodate the fact that there would be relatively high default levels. (They went wrong by underestimating the defaults levels.)
Here's how Sam Jones at FT Alphaville describes the disconnect between trying to use covered bonds to revitalize subprime home lending:

Subprime securitisation works because you overcollateralise to accommodate statistically modelled losses: in a subprime RMBS or CDO, you expect losses to occur, and the tranching - the structuring - is what compensates for this and allows for ratings all the way to triple A.
The tranches are 'attached' to the collateral in such away so as to reflect the statistical unlikelihood of - say - 50% collateral default with only 50% recovery. In that very bearish scenario, 25% of the collateral's value is lost. If the AAA attachment point was 70% - it would still not see those losses. Thus it's AAA.
In a covered bond though the security - the AAA rating - is achieved differently and not through tranching. The triple A (at least going by form) comes from the quality of the collateral on offer and the fact that it is secured in an on-balance sheet pool, immediate recourse to which is protected by specific statute.
Further, the market value of that pool is set and the issuing bank has to make substitutions in the event of collateral deterioration. And Further in many cases the bondholder has full recourse to the issuers balance sheet in the event of a shortfall.
Thus the backing for covered bonds in Europe is unexciting: prime mortgages, public sector debt and ship loans.