The Myth of Systemic Collapse
Contagion theorists argue that the global financial system is inherently fragile, meaning that if several significant institutions fail, the system will implode. But the evidence points to the opposite conclusion. The financial system is robust and able to deal with the failures of some of its largest institutions.
The justification put forth by regulators and politicians is that bailout money needs to be spent in order to ward off systemic collapse. In announcing additional aid to AIG on Monday, the Federal Reserve and the Treasury issued a joint statement pointing to the “systemic risk AIG continues to pose.” Given the large and growing price of keeping systemic risk at bay, it is important to explain exactly what it is.
The concept is straightforward. A large bank fails, and its counterparties, to whom it owes money, also fail and so on in a domino effect. Those who support this theory invariably point to vast sums of derivative contracts outstanding. If there are tens of trillions of dollars of derivatives contracts, it must be the case that the system is vulnerable to the failure of a large player.
Derivatives, those “financial weapons of mass destruction,” to quote Warren Buffet, are the transmission mechanism whereby the financial world ends with a bang. It has become an article of faith that the system will implode if some of the larger players fail, because the derivative claims presented by their counterparties would go unpaid, leaving their counterparties unable to meet their own obligations, and so on.
We need to question this article of faith, and it might help to start with the Lehman bankruptcy. Lehman’s failure was a perfect storm. Lehman was not only the 4th largest U.S. securities firm, it punched above its weight in fixed income, and specifically in credit derivatives. Lehman was both a huge dealer in Credit Default Swaps (CDS) and one of the most common reference names to be traded in individual swaps and CDS indexes.
Worse, the settlement of Lehman CDS was about as bad as could be imagined. The settlement price determined by dealer auction was a measly 8.625 cents on the dollar. This means that of the $400 billion in Lehman CDS outstanding, Protection Sellers (those long Lehman) had to pay $365 billion to the Protection Buyers (those short Lehman).
Before the settlement of those swaps last October, concerns surrounding the payments were widespread. The New York Times’ Floyd Norris worried that “we do not know how many such swaps are even outstanding, let alone who is on the hook to pay.”
But settlement was a non-event. Given the widespread trading in Lehman CDS, and the near zero recovery rate, it is remarkable that the CDS market was able to function as designed. More impressive, there does not appear to have been a significant knock-on effect where Lehman’s own counterparties suffered sizable losses on under-collateralized trades.
While it is fashionable to decry the unregulated nature of CDS, the market did its job fairly well because the collateral system among banks and hedge funds insured that most of the money protection sellers would need to pay protection buyers was already in place.
Then came AIG. And in order to avoid systemic collapse the government invested $150 billion and counting. Now that AIG’s stock price has fallen below $1 and its market capitalization is about $1 billion, it is worth remembering why this investment was made in the first place.
AIG was a special case because banks would trade with it on an uncollateralized basis. AIG Financial Products was a Protection Seller against mortgage CDOs, meaning it insured the top tranches of these now toxic securities. Because of its size and credit quality, AIG would often not have to post collateral for this insurance, provided it maintained its AAA rating. In retrospect, the decision to buy protection from AIG without adequate collateral mechanics was just another foolish credit decision by the banks.
Facing imminent downgrade, and realizing it would not be able to post sufficient collateral to satisfy its contractual obligations, AIG found an equity partner in the U.S. taxpayer. After this, the story gets murky, but we know a large part of the taxpayer’s investment went to satisfy the “margin calls” of banks and brokers to whom AIG had sold protection. It is something of a national scandal that we do not know exactly how much and to whom.
What we do know is that last September a number of banks and brokers were able to convince Treasury Secretary Paulson that if AIG failed, the stress to the system would be too great and systemic collapse would follow. I believe Paulson was gulled, and I think the Lehman bankruptcy bolsters my case.
AIG was unusual in that very few institutions get away with not posting collateral on derivative trades. The general trade for a bank with a bank or a bank with a hedge fund is two-way margining. That is, as the price of the underlying contract fluctuates, collateral goes back and forth on a daily basis. These arrangements are governed by Master ISDA Agreements and while there are variations among them, it is unusual between large institutions for the contract to be one-sided, meaning one side posts margin and the other side does not.
Where Paulson appears to have been hoodwinked was in believing that system-wide collateral arrangements were hugely inadequate. This myth persists today. AIG’s unusual arrangement became a metaphor for the whole system. But AIG’s collateral arrangements were the exception, not the rule.
Clearly AIG’s failure to post collateral upon a ratings downgrade would have severely damaged a number of its immediate counterparties who traded with it on the basis of “ratings triggers” rather than actual collateral. But in order for the contagion theory to be true, it would need to be true that these direct counterparties of AIG had also not posted sufficient collateral to their counterparties. This is very unlikely.
In the world of derivative trading, Lehman, not AIG, was the norm. What this mean is that in general, banks have adequate collateral against counterparty claims. Those who traded derivatives with Lehman seem to have had sufficient collateral to cover the unwinding of their trades following Lehman’s bankruptcy filing. The system worked as designed, even while unsecured creditors of Lehman can expect to be virtually wiped out in the final bankruptcy settlement.
Thus the Lehman fact pattern indicates that whatever contagion would have followed an AIG failure would have been limited to AIG’s largest direct counterparties and not created wave after wave of failure. Those who convinced the government to save AIG to save the system were really saving themselves.
The hysteria whipped up by those who insist that we must keep moribund institutions alive to prevent financial Armageddon is now bordering on recklessness. Monday’s joint Treasury-Fed announcement mentioned AIG’s “30,000,000 policyholders in the US” and “retirement insurance for teachers.” Surely Bernanke and Geithner know that the insurance subsidiaries of AIG are separately capitalized and insulated, by design, from the problems of AIG Financial Products and the holding company. What possible reason could there be for frightening policyholders and retired teachers?
Nobody expects that we can have a series of large bank failures in America without suffering some pain. But the ensuing dislocation must be weighed against the cost of the bailouts. As the situation continues to worsen, we need leaders who can explain these tradeoffs with credibility. At the moment, we have a perverse situation where our leaders seem to have a vested interest in making the system appear weaker than it is.