McClatchy, the only major US news organization to question the Iraq war until is was obvious to all that it was a misguided exercise in neocon hubris, has started a series on Goldman’s famed “short subprime” exercise. While the timing and overall outline are not new (as to when and allegedly why the investment bank went short), it delves into some details that have heretofore not been examined, as to how much subprime paper it dumped onto investors during this period, whether this duplicity was permissible, and what sort of damage was visited on foolhardy borrowers.
Unfortunately, for my taste, the series does not appear to be getting enough into the nitty gritty (and it indicates clearly that Goldman has successfully kept mum about the details of how it executed its short). I am keenly interested, because my understanding is that any simple subprime index short would have blown out spreads and thus been very costly to execute.
Goldman used another route….and the road, not surprisingly, was through AIG. From an e-mail over the summer:
This also points out a *VERY* good nugget re: banks who used CDOs/AIG offensively as opposed to as a hedge. This is likely what bothered me most about the AIG debacle. The trades GS had on with AIG were generally *not* super senior CDOs GS was long simply because they had
underwritten CDOs and were “stuck” with the AAA risk as a result. Rather, GS had a whole program of issuance — something they called “Abacus” — which were deals they put together with the sole purpose
of getting short subprime/CDO risk. Their sole purpose in doing the deals was to get long protection/short risk on the underlying collateral. AIG was simply the vehicle they chose to moneitze that PnL. Call me crazy, but I put the AIG counterparties in two different camps: guys like SocGen, who bought bonds in good faith and then hedged the credit risk by buying CDS from AIG, and guys like GS, who used AIG as their lottery ticket for offensively constructed trades to capitalize on mispriced subprime risk. The former, to me, seem much more deserving of a bailout than the latter…
DeutscheBank had a broadly similar program called Start.
This of course makes complete sense. There simply was not enough insurance capacity (the monolines plus the volume on the Markit indexes) to account for the big names that went short (Paulson, Goldman, one other large but secretive player we are aware of). That road had to go through AIG as well.
And bear in mind another fact: asset backed securities CDOs (and the subprime kind were that type) were managed rather than passive. That mean when the collateral paid down, the manager would go and find new collateral. Again from an e-mail:
AIG got out of subprime in 2005/2006 – whenever – but it didn’t matter. Why?? Because the same crappy borrowers that made it into 2005/2006 subprime RMBS refinanced and ended up in the 2007 vintage. Guess who had to buy the 2007 subprime RMBS paper when the 2005/2006 paper repaid? You got it – the 2005/2006 CDOs. CDOs have reinvestment periods (4 yrs for SF CDO) whereby they have to continue to be fully invested rather than letting their liabilities get repaid. The liability buyers don’t want their valuable paper to be repaid early – or, do they????
Readers who know the terrain, and Abacus and Start in particularly, are very much encouraged to comment or ping me at yves@nakedcapitalism.com.
Now to McClatchy:
McClatchy’s inquiry found that Goldman Sachs:
Bought and converted into high-yield bonds tens of thousands of mortgages from subprime lenders that became the subjects of FBI investigations into whether they’d misled borrowers or exaggerated applicants’ incomes to justify making hefty loans.
Used offshore tax havens to shuffle its mortgage-backed securities to institutions worldwide, including European and Asian banks, often in secret deals run through the Cayman Islands, a British territory in the Caribbean that companies use to bypass U.S. disclosure requirements.
Has dispatched lawyers across the country to repossess homes from bankrupt or financially struggling individuals, many of whom lacked sufficient credit or income but got subprime mortgages anyway because Wall Street made it easy for them to qualify.
Was buoyed last fall by key federal bailout decisions, at least two of which involved then-Treasury Secretary Henry Paulson, a former Goldman chief executive whose staff at Treasury included several other Goldman alumni.
The article continues here.






I wonder how the crisis would have played out if AIG had split into AIG Financial Products and AIG Insurance-and everything-else in, say, mid-2006. At that time I believe the FP side would have been valued higher than the rest of the company, based on earnings.
It might have been a bit harder politically to throw the requisite money at a relatively small Financial Products company to make their counterparties whole. Especially if AIGFP had domiciled in London . . .