On Wed, 16 Nov 2011 14:39:31 -0500, Jeffrey Turner
Why exactly do you think the retirement fund managers would be
interested? The underlying assets in Abacus waere rated Baa2, not AAA
and all of the investors knew that. A bit over 20% of the synthetic
instruments created from that collateral were rated AAA. The fact that
99% of the underlying assets were downgraded further is irrelevant;
IIRC, Moody's Aaa rating is set at the point where the underlying
collateral pool experiences 30% or more defaults. That is, if the
percentage of defaults in the underlying collateral pool is less than
30%, the Aaa-rated assets won't default. The last time I checked,
defaults in the underlying assets were at about that level, so it
appeared that the available capital would be barely enough to avoid a
significant AAA default.
That situation may have changed, but it is useful to note that as of
last year, despite the heavily-reported decline in underwriting
standards and the worst housing downturn since the Great Depression,
less than 4% of all of the AAA rated tranches created over 2000-2009
were either impaired or expected to become impaired.
Most of the actual losses in the meltdown were in plain vanilla
securities, like mortgages and corporate bonds, not in the exotic
structured products themselves. Again, in the structured products
sector, most of the losses were in the below-investment grade
mezzanine and equity tranches.
From a policy perspective, what you should be taking away from this is
the impression that these vehicles are in fact very robust. In case
you didn't notice, the Obama administration structure the bailouts
using the same, rational method. On some level it makes little
difference whether you like them or not. You must have some method of
providing liquidity to the housing and credit markets, or you will
have neither; new issuance of those vehicles is back to about the same
level as in 2004, and liquidity is improving in both the commercial
and residential real estate sectors. Whether you should permit
undercapitalized betting on those vehicles in the derivative markets
is a separate issue.
Abacus is the lazy man's example, but I'm glad you picked it. I can
kill two birds with one stone. Back when the SEC announced its
charges, I was instantly on the warpath against Goldman's alleged
fraud. The SEC's position implied that the product was a standard
collateralized waterfall; it would be unconscionable to permit a short
seller to choose the assets in a first loss vehicle, particularly
while implying that he would be one of the equity holders. Fraud was a
slam dunk.
In the interim, much more information has been disclosed. A trader
wanted to bet against the US real estate market. Another trader wanted
to bet the other way. They did their trade in the *derivatives*
market. I'm not aware of any evidence that either the short or long
trader committed any fraud as they were negotiating what assets would
go into the underlying asset pool.
In this case, the short position was held by Paulson, but who the
short is is irrelevant. ACA, backed by ABN AMRO (Dutch), mainly
because Goldman didn't trust ACA's capital position, wanted to be
long. IKB Deutsche Industriebank AG was long in exchange for a higher
yield. Goldman was long because they weren't able to sell off their
entire exposure. The longs, including Goldman, lost. The shorts won.
Where's the fraud?
Anyway, mea culpa. I convicted Goldman without examining the details
of the deal. I seriously doubt that you could convene an expert jury
that would convict Goldman of the fraud alleged by the SEC.