AIG - CDS - Credit Default Swaps

As you can see from this slide, AIGFP has very large notional amounts of exposure related to its Super Senior credit derivative portfolio.

  • But because this business is carefully underwritten and structured with very high attachment points to the multiples of expected losses, we believe the probability that it will sustain an economic loss is close to zero. (p4)

2007 1205 - AIG Investor Conference Call Transcript_1.pdf - 66p

  • Senator COLLINS. Actually, I would argue that the credit default swaps were more like an insurance product and yet they were not regulated by State insurance agencies either.
  • Mr. GENSLER. They had many insurance attributes. 
    • There were many lessons, unfortunately, out of this crisis.
    • You were earlier asking Chair Schapiro, but I think one of the great lessons of AIG was that there was unregulated institutions.
    • That's why I am for regulating all derivative dealers, whether they're affiliated with banks or not.
    • But then these products, as you say, credit default swaps, have attributes of insurance, like monoline insurance.
    • They have attributes of securities.
  • Senator COLLINS. Exactly.
  • Mr. GENSLER. They have attributes of derivatives that the CFTC is the expert on.

2010 0930 - GOV (Senate) - Financial Services and General Government Appropriations for Fiscal Year 2010 - [PDF-239p-GooglePlay, VIDEO-?]

  • 2014 - SOA - Model Validation for Insurance Enterprise Risk and Capital Models - 47p
  • (p26) - A Case Study of AIG’s Model of Credit Default Swaps
    • {AIGFP, Gorton Model,  Gary Gorton
    • An October 31, 2008, article in the Wall Street Journal [16] included this observation:
      • AIG didn’t anticipate how market forces and contract terms not weighed by the models would turn the swaps, over the short term, into huge financial liabilities.
      • AIG didn’t assign Mr. Gorton to assess those threats, and knew that his models didn’t consider them.
    • Another observation from the same article:
      • Mr. Gorton’s models harnessed mounds of historical data to focus on the likelihood of default, and his work may indeed prove accurate on that front.
  • But as AIG was aware, his models didn’t attempt to measure the risk of future collateral calls or write-downs, which have devastated AIG’s finances.

These statements make it clear that any validation of AIG’s CDS model was done with reference only to potential claims, and not as a model of the full financial operation of the CDS business.

  • The CDS contracts required AIG to post collateral if the value of the insured loans fell.
  • The model did not include this collateral requirement, nor did it reflect the way that fluctuations in loan value could dramatically increase the short-term cost of meeting that requirement.
  • The model used to run the business was incomplete because it did not simulate all of the material risks of the business.

As a result, management did not maintain either the liquidity or the capital required to post collateral when the need arose. 

[16] Carrick Mollenkamp, Serena Ng, Liam Pleven, and Randall Smith, Behind AIG’s Fall, Risk Models Failed to Pass Real-World Test, Wall Street Journal, October 31, 2008. <and/or 1103 2008> - <WishList>