Note: this post is by Thomas Adams, at Paykin Krieg and Adams, LLP, and a former managing director at Ambac and FGIC, with some minor additions by yours truly. This is a significant piece of some puzzles he, some other experts who prefer to remain anonymous, and I have been pushing on for several months.
As we have been reading the latest coverage on the AIG bailout from the SIGTARP report and the Treasury Secretary Geithner’s Congressional testimony, a nagging question remains unresolved: why did AIG get bailed out but the monoline bond insurers did not?
The business that caused AIG to blow up was the same that caused the bond insurers to blow up – collateralized debt obligations backed by sub-prime mortgage bonds (ABS CDOs). This was actually one of the few business that AIG Financial Products had in common with the monolines. AIG didn’t participate in municipal insurance, MBS or other ABS deals, which were all important for the monolines.
Certainly, AIG was larger than any of the bond insurers, but in aggregate, the bond insurers had a tremendous amount of ABS CDO exposure, which at the peak was probably over $300 billion. Despite AIG’s claims to have withdrawn from subprime at the end of 2005, we have identified particular 2006 deals with substantial subprime content that AIG most assuredly did guarantee.
In addition, the monolines had exposure to many other assets classes that AIG did not which created chaos for the holders of those bonds when the monolines were downgraded. The chain reaction risk of the bond insurers was arguably greater, when you throw in the damage to the aucton rate securities market, which was rooted in the muni market. In 2007, MBIA had over $650 billion of par insured, Ambac had about $500 billion, FSA had about $380 billion and FGIC had about $300 billion. Throwing in CIFG and XLCA, the total insured par of the monolines was about $2 trillion – this amount certainly would qualify as large enough to be “systemic risk” if the insurers were allowed to fail.
In contrast, while AIG’s aggregate insured par was greater, the only portion that really presented a systemic risk exposure was the CDS and structured finance exposures, which had an aggregate par exposure of about $400-500 billion. a persuasive argument could be made that the monolines were just as intertwined in the financial system as AIG and, thanks to their municipal exposure, presented as great or greater a systemic risk to the financial markets and the economy.
Yet AIG was bailed out and the monolines were not.
So what happened? How did the monolines get dropped and AIG get rescued? The popular reason given has been that AIG was so big that they affected all segments of the economy, whereas the monolines were only midsized and not critical to the economy. i believe that SIGTARP repeated this version of events last week. I understand that Treasury Secretary Geithner last week repeated this notion and added new information – that he was concerned about the cascading risk of AIG’s non CDS exposure.
While this produces a bigger par exposure for AIG, these other areas did not have the huge risks of loss, have largely remained functional, and did not have the issue of collateral posting. The risk were at the parent level, at AIG FP; the bulk of AIG’s business was written by regulated subsidiaries whose claims-paying ability would not be impaired by an AIG FP failure. So, in my view, this is a fairly weak, after the fact argument. A more plausible case might be made that AIG also had a securities lending business that had sprung a $20 billion leak, but that wee problem hasn’t gotten much mention in the official defenses.
I have a different interpretation. I should note that I am a former employee of a bond insurer, so I admit to a bias. However, I my general perspective had been, until recently, that neither AIG or the bond insurers should have been rescued.
When I was at FGIC, Deutsche Bank, Lehman, Bear and UBS were all over my company with sales coverage for CDO deals. But we never heard much from Goldman. I was actually surprised to see that they were so big with John Paulson’s CDO adventures (as recently disclosed in “The Best Trade Ever”), because I never thought they were that big in the CDO market.
One big reason I didn’t know Goldman was so big in CDOs – they didn’t work with the monolines.
Goldman wanted their counterparties to post collateral so they would have protection against corporate downgrades. The monolines refused to have collateral posting requirements in their CDS contracts. The rating agencies supported them in this position on the argument that maintaining their AAA rating was “fundamental to their business”.
AIG, on the other hand, agreed to collateral posting requirements. in fact, they used this as a competitive advantage – they got more business because of it and marketed their flexibility on this issue to the banks. There were two the key distinctions between the monolines and AIG – first, AIG had other businesses, whose losses could threaten AIG’s financial guarantee business while monolines promised to pay claims first, to protect investors. Second, AIG had a history of negotiating before they paid claims (there is an interesting history with a ABS film receivables deal where AIG refused to pay, while the monolines covered similar deals and did not have the same “out” in their policies. this deal did serious damage to AIG’s reputation in the ABS market and shut them out of many deals). So despite their AAA rating, AIG was not as trusted by the structured finance and CDS market – there was a fear that AIG would wiggle out of their obligations in a way that the monolines would not.
All of the other banks got comfortable with the monolines not having to post collateral for CDS trades because of their AAA ratings. Goldman never did.
Of course, Goldman was one of the few banks that clearly set out to profit from shorting CDOs. They obviously realized that if their CDS counterparty was on the hook for a lot of ABS CDOs that were going to blow up, the insurance provider would likely get downgraded. If the downgrade of the insurer was very likely, the only way the short-CDO strategy worked was if the insurer would post collateral.
So Goldman only used AIG, who would provide protection against their downgrade, which Goldman knew would happen because they were stuffing AIG with toxic ABS CDOs.
The banks that used the monolines for their ABS CDOs were making a major error by taking on the monoline downgrade risk without protection, especially if they knew that the ABS CDOs were toxic. So I suspect that most of the banks did not really know that the ABS CDOs would be as toxic as they turned out to be.
This is, of course, what happened. The ABS CDOs blew up, the bond insurers got downgraded, the banks that used them got crushed because their hedges against their CDO risks were now in jeopardy. A death spiral between the monolines and the banks ensued (the ARS meltdown added to the troubles). Goldman didn’t care, because they had collateral posted by AIG once AIG got downgraded..
All of the banks who faced the monolines had to start considering commutation deals with the monolines because it was obvious the monolines did not have enough capital to cover all of the CDO losses. in these commutations, the banks accepted payments as low as 40 cents on the dollar.
Most of the monoline ratings roubles had unfolded earlier in 2008 – many of them had been downgraded, several commutations had already occurred by the time of the AIG bailout. AIG managed to put off the threat of serious downgrade for a long time, despite the junk in their portfolio (as 2008 progressed, it was a mystery to me and many others why the onolines were being downgraded but AIG was not). While AIG had been downgraded to AA some time earlier, this hadn’t caused much of a disruption because the real trigger for collateral posting was if they went below AA. For a variety of reasons, this wasn’t a threat until September of 2008.
I hate to get sucked into the vampire squid line of thinking about Goldman, but the only explanation i can think of for why AIG got rescued and the monolines did not is because Goldman had significant exposure to AIG and did not have exposure to the monolines.
When it became clear that AIG could face bankruptcy, Goldman’s plan to profit by shorting ABS CDOs was threatened. While they had the collateral posted, thanks to the downgrades, this collateral could be tied up or lost if AIG went bankrupt. This was a real crisis for Goldman – they thought they had outsmarted the subprime market with their ABS CDOs and outsmarted all of the other banks by getting collateral posting from AIG when they got downgraded. But if AIG went away, this strategy would have blown up and cost Goldman billions.
All of this is essentially factual and based, for the most part, on public information.
As a matter of speculation, i believe that Goldman and their helpers deliberately pumped up the media with the threats that the subprime market posed in order to hasten the collapse of the subprime market. this allowed them to realize their gains sooner from shorting ABS CDOs – they had become impatient waiting for it to blow up.
In addition, I believe that Goldman and their helpers – including their many connections with the White House and the Fed – pumped up concerns about the systemic risk that the market was facing from a Lehman and AIG failure, so that they could force the government to step in and bail out AIG. This would also explain why Lehman was not bailed out. Lehman didn’t really matter to Goldman. But the fear created by Lehman’s failure served as a good excuse for why they should rescue AIG.
I have been wondering why the sub-prime market blow up led to such a massive crisis when subprime and structured finance had experienced big problems before without the issue of systemic risk and financial market collapse.
Certainly, the ABS CDOs were toxic and caused big holes, but not so big that it couldn’t be addressed by an RTC type of clearing system. Various analyst reports of the bad subprime deals (and ABS CDOs) makes it pretty clear that the 2006-2007 vintages were the worst and will probably only create about $500-700 million of aggregate losses. Terrible, but not insurmountable.
This leads me to conclude that the bailout was prompted by fear mongering and deliberate strategies and manipulation on the part of Goldman and a few select others, to make sure that AIG would be bailed out to protect their trades in shorting ABS CDOs.
i believe that John Paulson benefited from this bailout, on his $5 billon or so of ABS CDOs with AIG. But not as much as Goldman benefited themselves, via Abacus and, perhaps, other deals.
AIG, Goldman and ABS CDOs were tied together at the center of the crisis. From Goldman’s perspective, all of the other participants were secondary – they had no exposure to the monolines and they were probably hedged against the other banks. The only loose end was the collateral posted by AIG.
The final question that this raises for me: would it have been cheaper for the government and the taxpayer to have bailed out the bond insurers instead of AIG? The total amount of CDOs and credit default swaps that would have needed to be guaranteed would have been smaller. In the number of investors across the market that would have benefited would probably have been larger. The auction rate securities market, the muni market, the investors that held bond insurer exposure to MBS and ABS would have all benefited. None of these markets were aided by AIG’s bailout.
But a bond insurer bailout would not have helped Goldman much and the AIG bailout did.
Yves here. Note I differ with Tom on how much Goldman could or did pump up subprime fears. A lot of people focused on Jan Hatzius’ bearish calls on financial system losses, but quite a few people in a position to know claim that Hatzius is not the sort to take commercially expedient views. But once the asset-backed commercial paper started imploding in August 2007, the officialdom was very much engaged. So if one can connect the dots between Goldman and the fear and loathing that hit the ABCP market (recall all paper was repudiated as in being possibly tainted by subprime), the story becomes very tidy.
Update 1:50 PM Another possible gap in this line of thinking are the now-infamous AIG regulatory capital swaps, which allowed European banks to carry much less equity (or put it another way, pump up their balance sheets much bigger than they would have otherwise been). But there has been a remarkable lack of coverage of this issue. That would be one reason to save AIG and not the monolines.
There isn’t any evidence that this issue factored into official thinking. That could mean that the officialdom has scrupulously avoided mentioning it (as in why alert the peasants that their tax dollars are supporting profligate Eurobanks), but Sorkin’s Too Big to Fail makes clear that AIG itself was not on top of how badly the ship was leaking, so if AIG didn’t bring this issue up as a need to be rescued, no one would have factored that into their decisions. One way to be certain would be to compel disclosure of the phone logs during the AIG scramble. A absence of calls to European banking regulators would be indirect confirmation that these swaps were not one of the proximate causes of the AIG rescue.