A.I.G.: Why it is appropriate to delay any liquidation/bankruptcy.


The American Government is poised to inject additional capital for the third time into the former world N° 1 insurer, bringing its total support to USD 130 billion!


The main question is not so much whether or not the company should be nationalised (the State owns already 80% of the common equity, is acquiring direct stakes in profitable subsidiaries as part of the new package and has a pledge over the entire assets of the company), but rather to decide whether it is worth while continuing to throw good money after bad into what seems to be a bottomless pit.


The American authorities responsible for managing the case (Treasury Department – Federal Reserve – FDIC) are unanimous in believing that an AIG bankruptcy would pose the threat of a significant systemic crisis. Their analysis is based essentially on the existence of “counterparty risks” through which major world financial institutions are exposed to AIG and vice versa. The astronomical amounts concerned are mainly attributable to “Credit Default Swap” (CDS) contracts, by which AIG “insured”  “policy subscribers” against default risk of their own debtors.


As has been widely reported in the media, the CDS market has exploded over recent years, outside of any regulatory framework, reaching a size, according to some estimates, in the area of 60 trillion USD. At its inception, the economic justification for this type of instrument was perfectly sound and provided a useful tool for prudential risk management. It fostered the development of a secondary market in bank loans, providing a broader dissemination of risk as well as a welcome significant increase in the market’s loan capacity.


However, several developments polluted progressively this unregulated market:


-          In the first place, CDS became increasingly used as a purely “speculative” vehicle rather than as a legitimate “hedging” instrument. This means that the purchaser of a CDS did not have an underlying economic interest, but was simply placing a bet on the failure of a debtor with whom he had no direct relationship. This type of transaction became preponderant over time.

-          Secondly, the dissemination of information and commentary by the media concerning “spreads” paid by a borrower relative to a benchmark (US Treasury securities or LIBOR) became an “indicator” of the default risk of an issuer (rather than questionable and less responsive Ratings). This played a role, not so much at the time of issuance (at which time the spread reflects the real cost of borrowing for the company) but more significantly in secondary market transactions over time (when the spread reflects the degree of risk supported by the creditor). Thus, a vicious circle emerged through which “imputed spreads” reflected in secondary market transactions were influencing the “insurance premiums” charged by CDS issuers. This would, in turn, impact refinancing costs for borrowers, the banks relying more and more on the indications provided by the CDS market rather than on their own internal evaluation to fix loan conditions. In this way the financial equilibrium of borrowers could be but in jeopardy. This single minded focus on the relationship between CDS premiums and credit risk removed totally from the equation the other factors that influence market spreads, among which the most important relates to the liquidity of the market itself. Hence, the freezing of the interbank market, after the emergence of the subprime crisis in the summer 2007, was the direct cause of the widening of lending spreads, long before the financial crisis spread to the economy at large or affected the solvency of borrowers.

-          A third factor was the combination of CDS with other financial products: “innovators” developed a range of new structured products (so called “guaranteed instruments”) each more complex and opaque in which “counterparty risk” linking the various participants in the structure were considerably underestimated (if at all considered). This phenomenon contributed significantly to the broadening of uses of CDS beyond their original purpose.  The AAA rating of AIG was put to use to underpin the credibility of the “guarantees” often incorporated in structured products reassuring investors who were incapable of evaluating independently the risks imbedded in the structure.


An AIG bankruptcy in the current environment would have a different impact on different types of policy holders:


-          Concerning policyholders in the classical insurance sectors, there should be no adverse consequences, each policy being lodged in a ring fenced solvent subsidiary, subject to effective regulatory control.

-          For a counterparty to a CDS issued by the “AIG Financial Products” division, which escaped entirely regulatory supervision (not by negligence of Regulators, but rather because of the lack of the necessary legal authority) let us distinguish between 3 scenarios:

o   The counterparty purchased a “naked” CDS covering a debtor with whom he has no other economic interest. An AIG failure penalises him to the level of the premiums paid and the lost profit opportunity in case of default of the debtor subject of the CDS. This case is least likely to have systemic consequences.

o   The counterparty purchased a CDS covering the default of a debtor of which he is a creditor. In this case, receivership of AIG restores to the subscriber the full risk of default of the debtor, precisely at a time when the financial crisis increases the likelihood of such a default. For instance if the subscriber is a bank who decided that it was prudent to “insure” a portfolio of credits (possibly subprime) bought from other institutions, it would find itself unwittingly exposed to the subprime market, while it believed it was protected by an AAA guarantee. The removal of AIG as counterparty could put into question the financial equilibrium of the bank and, at the level of the banking sector as a whole, create a systemic risk.

o    The counterparty integrates the CDS into the structure of another financial product which can already be itself the subject of a previous multi level structuring (such as the securitisation of a pool of diversified asset backed securities).The package is sold on to investors or other intermediaries. This technique was broadly used by both commercial and investment banks in the development of the now emblematic “subprime” market. In case of an AIG failure there is a dual risk: first the one described in the second case here above; in addition, banks or other financial intermediaries active in the structuring or distribution of these products would be subject to legal action by disgruntled investors. This risk has become a stark reality after the Lehman failure, which led - in Belgium alone - to complaints against both Citibank and Kredietbank for having placed “Lehman products” in their customer’s portfolios. An AIG bankruptcy would therefore create the possibility of failures of these intermediaries who, in light of their size (Citibank – J.P. Morgan – Goldman, Sachs), would create additional systemic risk. 


In such an environment, it is not surprising that authorities refuse to endorse the risk of the failure of AIG, whose consequences it is nigh impossible to evaluate. This reticence is all the more understandable that many observers blame the Lehman demise for the deepening of the financial crisis after September 15th 2008.


The 80% equity stake of the American Government in AIG coupled with the pledge of all the company’s assets protects – as far as possible – the tax payer and reduces to practically zero any hope for shareholders to recover anything whatsoever which is totally appropriate. Keeping the company alive, on the other hand, allows time to unwind, in an as orderly fashion as possible, the complex structures imagined by the high flying sorcerer’s apprentices of finance. Acting in this way, one attempts to avoid an immediate cascade of failures and a total meltdown of the financial system with its incalculable consequences.


In parallel, it is urgent to accelerate the implementation of a regulatory framework for the CDS market and in particular to institute a central clearing and settlement entity which could assist in shrinking the amount of outstanding contracts through the cancelling of reciprocal equivalent obligations (netting).


In this paper, I limited the discussion to the CDS market which, because of its size, opacity and lack of regulatory supervision is clearly the Achilles heel in the AIG case. This does not mean that there are not other financial instruments outstanding capable, either individually or jointly, of putting the structure of the whole financial sector in jeopardy.


It is therefore absolutely correct to maintain that, by avoiding the failure of AIG, the American Government is acting with the prime objective of saving the financial system in the country’s interests and not attempting to protect any particular institution or privileged group of individuals. That having been said, it will nevertheless prove very difficult to restore the necessary confidence needed to kick start the economy on a sound footing, as long as the amount of write-offs to be recognised in the books of financial institutions has not been established with a reasonable degree of accuracy. The purposefulness and determination shown by President Obama’s Administration are probably the best hope we have of surmounting the crisis. There can, however, be no guarantee that the colossal amount of public funds committed will prove sufficient to remedy a situation, the parameters of which still remain largely undefined.


Brussels, March 3, 2009


Paul N. Goldschmidt

Director, European Commission (ret); Member of the Advisory Board of the Thomas More Institute.