There is an alternative, more behavioral hypothesis for the fragility of the securitization market that does not rely on a predominance of short-term debt financing. This alternative hypothesis begins with the observation that a large proportion of ABS tranches–both in the traditional and subprime sectors–were rated AAA. The AAA rating may have encouraged investors such as pension funds or insurance companies to think of these securities as essentially riskless, and therefore to treat them as being equivalent to Treasury bonds when constructing their portfolios. When the problems in the subprime area became apparent, this premise was utterly destroyed, and investors who were determined to allocate a fraction of their portfolios to safe assets realized that they had to dump their holdings of AAA-rated ABS, and buy actual Treasuries instead. Thus instead of a short-term-debt-driven bank run, we have what might be called a widespread buyer’s strike. In this account, the mechanism of contagion from the subprime market to the traditional consumer ABS market is that the failures of the rating agencies with respect to subprime called into question their credibility more generally, so that any AAA-rated tranche of an ABS, be it linked to subprime or credit cards, was no longer considered to be a virtually riskless asset.
The full essay is here, interesting throughout, via David Warsh's very good column. Stein also makes the simple yet neglected point that higher capital requirements may simply shift more financial activity into the less regulated shadow banking sector.
Could be a nobody-got-fired-for-buying-IBM issue. I’ve heard a similar reasoning for why people really buy insurance when not required.
Are risk-free assets similar to fat-free foods? We know that they are healthier and will therefore think we can eat more of them.
I think the question is who was required by law or by CYA to buy AAA?
Good. Let them shift to the non-bank sector. Maybe take down a hedge fund or two for all I care.
The less regulated non-bank sector doesn’t take deposits, and their bond holders can take a hit ala CIT. Caveat emptor if you are a lender to such an entity or own their stock.
Andrew: We really need to dispel the sloppy thinking that AAA this = AAA that.
I’m tired of hearing stupid shit from Barney Frank and company how it is unfair that Arizona’s municipal debt is rated lower than corporation X. Dip shit is too damn stupid to look on the Bloomberg terminal to realize that the OAS spread on that Bbb Muni is lower than the Aa corporate.
I think where Frank and company run into trouble is when the munis fall below IG levels and then their retarded legislation kicks in and pension funds have to sell the asset in a fire sale even though they are safer than a Aa corporate.
It would be easy to regulate much if not all shadow banking activity.
Tom: The ratings agency compile tables of the % of each security by rating that gets downgraded/upgraded/goes into default. I’m not sure if they are public record, but the guys that are setting the prices pension funds/insurance companies/endowments/etc have no excuse not to look at the tables.
As for a limit on Aaa, you are way off base. The amount of Aaa in the system is meaningless. The Treasury could print $1 trillion Aaa rated notes tomorrow. If the FDIC came out again and guaranteed all bank debt that limit on the amount of Aaa in the system is meaningless.
There was no problem with the calculations that came up with Aaa on ABS. There was a problem with the assumptions on the price of the collateral in the future.
Tell me how I’m oversimplifying:
Rewrite the rule, so that you can invest where you please, but you no longer can “be” a bank (with its hallowed position in an economy, and corresponding FDIC backing). Want to be a bank? Then you can’t invest in voodoo…
Who cares what people are investing in, as long as taxpayer funds (the Fed, FDIC, etc) don’t have to clean up the mess?
When they legislate that “AAA this = AAA that” aren’t they also legislating in a version of Gresham’s law as applied to bonds?
Comments on this entry are closed.