« Federal Reserve Takes the Subprime Meltdown Seriously | Main | Paul Krugman's Weblog »

September 18, 2007

Comments

Doug

Hi Brad,
Thanks for linking to my notes. I think you misunderstand my point #5, where the key word is "federal." Ned Gramlich pointed out that half of subprime mortgages in 2005 were issued by mortgage companies chartered and supervised by the states (compared with essentially no prime mortgages), and that the problem mortgages are heavily concentrated in that sector. He concluded in his book that all mortgage lenders should be covered by a *federal* supervisor. Others have argued that *federal* supervision is important for the Fed's ability to restore liquidity to these institutions. However, in Ned's Jackson Hole remarks, he said that the same regulatory end might be achieved by collaboration among federal and state supervisors, which was already underway. My notes cited this point and added some other reasons why I thought replacing state supervision with *federal* supervision was not needed (at least for these reasons--whether our overall supervisory structure makes sense is a harder issue). I was not arguing for no supervision of these institutions at all. I wonder, though, about your argument that any institution "borrowing short and lending long" is like a bank and needs to be regulated. I see your logic, but doesn't it apply to essentially all issuers of commercial paper?
On my point #6 about credit ratings agencies, I wish I had better ideas than I do. But regulators can, at the least, illuminate the potential conflict of interest between rating securities and soliciting business from securities issuers, and generate more separation between these functions within those organizations. However, I would not (and hope I do not in my notes) overstate the likely effectiveness of these steps.
Doug

mgrund

In regard to the oversight of the ratings agency-- They are already defacto regulators as they hold the key to the Investment grade criteria. If the regulators are to refer to these ratings when evaluating the health of an institution then there must be regulation.


How about establishing a random audit process to evalute the ratings on an on going baisis. Say 1% of the ratings would be redone by a federal watch dog-- if they concur no action is taken. If they do not then addtional sudits are conducted. If the watch dog finds a systemic problem then the agency can be placed on propation, publicly if necessary and then as a last resort they can lose the status as a National reconginsed rating agency.

I think you will find that they will be much more carefull in the future if this threat exists.

And I think we all agree that we need to have more compition in this field.

The comments to this entry are closed.

Pages

Search Brad DeLong's Website

  •