Can’t Get Enough: Goldman’s Profit is Citi’s Pain?

By Thomas Adams, at Paykin Krieg and Adams, LLP, and a former managing director at Ambac and FGIC.

Many thanks for the thoughtful comments on my earlier post. If you can take a little more on the subject, I thought I would add some clarification to some of the issues raised.

First, on the merits of a monoline vs. AIG bailout, consider a case of counterparty compare and contrast, using Goldman Sachs and Citigroup as examples.

The popular wisdom in the aftermath of the crisis is that Citigroup is the sick man of the banking world: they lacked sufficient controls to protect against unwise exposures to CDOs and other structures, they are still alive only due to massive government loans, etc. Goldman, in contrast, has been praised for their superior risk management and their shrewdness in avoiding the worst aspects of the crisis.

In light of the benefits Goldman received through the bailout of AIG, are these interpretations still accurate?

Citi had massive exposure to the monolines due, in part to their unique history together. Way back at the time of their first bailout around 1990, Citi owned two bond insurers: Ambac and Capmac. At the strong suggestion of the government, Citi unloaded Ambac and Capmac through initial public offerings in the companies. Capmac was subsequently acquired by MBIA. Though Citi sold off their interests in Ambac and Capmac, the companies have long been in bed with Citi. The senior management of both companies consisted of many former Citi executives. Both Ambac and MBIA were very active insurers for Citi’s asset backed commercial paper programs and for a variety of other structured finance assets.

Over the course of late 2007 and early 2008, as the bond insurers faced pressure and eventually downgrades due to their CDO exposures, the stock market would swing wildly around based on news of possible bond insurer bailouts. At times, these previously obscure companies seemed to hold the fate of the market in their hands. Despite widespread concerns about the impact monoline insolvency might have on counterparties, municipalities and investors, various commentators and bankers suggested that a bailout for the monolines would be inappropriate or might create the risk of moral hazard. Goldman Sachs, back in January of 2008, suggested that a bailout of the monolines would be ineffective and that putting the companies into run-off made more sense. As hedge fund manager Whitney Tilson (who was gleefully shorting the insurers) helpfully points out in the same article, a bailout of the monolines would have helped banks like Citi and Merrill who had been foolishly writing dozens of CDOs, but would not be in Goldman’s interest, since Goldman had avoided such risks. The monoline bailout never arrived.

It is strange then, after months of debate about the market-wide risk presented by monolines and the apparent government decision not to rescue them, that the Treasury and Fed suddenly discovered systemic risk when AIG faced a potential failure. This discovery coincided with AIG’s massive new collateral posting requirements after the company’s ratings were downgraded in September 2008. As it turned out, this posted collateral eventually flowed to the benefit of Goldman, among others. As it also turned out, the transactions for which Goldman received the collateral revealed that Goldman had not entirely been avoiding CDOs after all. They just had a different counterparty for their trades than Citi or Merrill.

Had the government bailed out the monolines, rather than AIG, Citi might have been in much better shape than they are now. The status of Citi’s huge off-balance sheet risk, some of which received enhancement from Ambac and MBIA, might be materially different. Today, Citi might be the bank being hailed for navigating the crisis successfully and paying impressive 2009 bonuses. In addition, had the Fed bailed out the monolines, no collateral posting would have been required and cash would not have flowed directly into Citi’s pockets.

Goldman, which had argued against a monoline bailout, was a major beneficiary of the bailout of AIG. In the absence of an AIG bailout, Goldman might now be the bank propped up by huge government loans. For this, Goldman can be credited for being a better political operator in a time of crisis than Citi. Or maybe they just managed to hide their bailout better.

Next, I would like to add a few clarifications.

AIG’s “regulatory capital trades” have been the subject of some confusion in the comments to my prior post and elsewhere. I believe that this phrase is just another name for “negative basis trades” which is just another way of describing collateralized debt obligations (CLOs) which were insured by AIG and the monolines via CDS. AIG had a massive portfolio of insured CLOs but, in aggregate, the monolines did as well. Some of the monolines who avoided real-estate related CDOS (such as FSA and Assured) had very large insured exposure to CLOs. In fact, all of the insurers had much bigger exposures to CLOs than to those CDOs.

As the crisis was emerging, the insurers sought to distinguish CDOs backed by subprime and mortgage collateral from CLOs backed by high yield corporate debt. However, these two asset classes were just offshoots of the same business. Like CDOs, the insurers would insure AAA rated bonds on a “secondary” basis by delivering credit default swaps to the protection buyer. Because both CDOs and CLOs were insured via CDS, they were both subject to mark to market accounting, whereas traditional insurance was not. In the wake of the financial crisis, the insurers experienced mark to market losses on both CDOs and CLOs.

However, by mid-2008, while no CDO had resulted in an actual claim yet, the insurers could no longer credibly argue that they would not incur real losses on the CDOs (though they tried). In contrast, CLOs, while suffering from impaired pricing, were still rated AAA and looked like they might survive without any losses. CLOs were still considered “safe” and therefore not a significant source of future insured losses. Thus, the impact from AIG’s and the monoline’s downgrades was much less for holders of insured CLOs. The regulatory arbitrage that the European banks had played with the CDS might be unwound, but the underlying bonds were still AAA, whereas the insured bonds in the many CDOs were worth pennies on the dollar.

With respect to questions regarding Goldman’s security in the collateral posted by AIG: this was no sure thing for Goldman. Had AIG gone bankrupt, the bankruptcy trustee could have tied the money up for years, as has happened in the UK to hedge funds exposed to Lehman. If Goldman had a pressing need for this cash, such a tie-up might have been a real problem. In addition, as the Skeptical CPA pointed out in a series of excellent posts some time ago (,, and, the posted collateral could be subject to claw back, or worse, by a bankruptcy judge, depending on whether AIG was determined to be “insolvent” at the time of the posting. Had the deals that lead to AIG’s dire straits been examined in detail by a bankruptcy judge, Goldman might have faced questions about their own culpability in their AIG’s demise.

Goldman has made it clear that they are upset and concerned about all of the media attention on their role in the bailout. In my view, they should really turn their gaze inward and be angry at the people who put them in a position of massive exposure to AIG. Instead, they (and their friends such as John Paulson) boasted about how much smarter they were than everyone else by profiting from the market downturn and having superior risk management. As we know now, some of these claims turned out to be an exaggeration.