The 2008 crisis demonstrated unequivocally that when the borrower is a large financial institution, collateralized lending does not protect the lender from losses. Without the intervention of the Federal Reserve as a lender who was willing to accept collateral that nobody else was taking, many of the largest repo market participants would have been forced to sell this collateral in order to meet their obligations. These forced sales would have driven asset prices far below those observed in 2008 – and all the major players in the repo market would have posted much larger losses than they did.
The case of AIG provides further evidence that collateralizing derivatives fails to protect the “in the money” counterparty from losses. Collateral calls following a rating agency downgrade of the firm drove it towards bankruptcy and precipitated a bailout in the form of a loan from the Federal Reserve. Notably $22 billion was passed from the Fed through AIG to counterparties in 2008. It is abundantly clear that this collateral would not have been posted in the absence of government intervention and thus, if standard bankruptcy procedure had been followed, AIG’s counterparties would have been short $22 billion on their derivative contracts.
The 2008 crisis demonstrates that the only protection a bank has against the failure of a large counterparty is the intervention of the central bank. Because of the fear of fire sales, collateral fell in value just when its protection was most important to lenders. Similarly, large collateral calls themselves precipitated bankruptcies – with the result that without the help of the Federal Reserve the collateral would never have been posted. In short, collateral was worse than useless throughout the crisis, because it served to destabilize financial institutions rather than to stabilize them.
The 2008 crisis raises this question: Is collateralized interbank lending an inherently destabilizing force in a financial system? Three quarters of a century ago J.M. Keynes expressed the problem perfectly:
Of the maxims of orthodox finance none, surely, is more anti-social than the fetish of liquidity, the doctrine that it is a positive virtue on the part of investment institutions to concentrate their resources upon the holding of “liquid” securities. It forgets that there is no such thing as liquidity of investment for the community as a whole.[1]
The exemptions to the bankruptcy code for derivatives and the collateralized interbank lending regime that grew out of these exemptions are built on the fallacy that there is such a thing “as liquidity of investment for the community as a whole.” Keynes also remarked on the most important cost created by ignoring the fallacy of liquidity:
The fact that each individual investor flatters himself that his commitment is “liquid” (though this cannot be true of all investors collectively) calms his nerves and makes him much more willing to run a risk.[2]
The collateralized interbank lending regime encourages banks to believe that their collateral is “liquid”, protecting them from losses in the event that a counterparty defaults. Adherence to this fallacy has two consequences: (i) banks do not set aside reserves or hold capital to protect themselves against losses that they cannot imagine, and (ii) banks do not monitor counterparties carefully, because they believe that they are fully protected by collateral. Both of these consequences are extremely detrimental to financial stability: the financial system as a whole is undercapitalized and in the absence of screening for credit risk the weakest financial institutions end up interconnected with every other firm. In such an environment, when one firm starts to wobble the whole financial structure can easily come tumbling down.
As last year’s Federal Reserve intervention demonstrated, collateralized interbank lending only protects lenders if the central bank is willing to intervene to prevent a fire sale of collateral. But then, what is the role of collateral? After all, the lender of last resort has a long tradition of protecting financial systems where interbank lending is unsecured. Collateral serves only to create the illusion of a security that does not exist. This illusion causes banks to reduce the capital they set aside to protect against unexpected losses and to cut back on monitoring the credit risk of their counterparties. In short, the existing collateralized derivatives regime is inherently destabilizing: It is not designed to function in an environment where a large financial institution can fail, it tends to reduce capital levels and increase lending to weak firms, and finally, because of the safe harbor exemptions, it all but guarantees that a run on a large financial institution will take place.
[1] Keynes, 1935, General Theory, p. 155.
[2] Keynes, 1935, General Theory, p. 160.