Hedge funds have mostly been exonerated in the typical narrative of the financial crisis, which concentrates blame on some combination of mortgage lenders, investment banks and government agencies.
A new paper by Yale professors Gary Gorton and Guillermo Ordonez, however, may indicate that hedge funds and other well-informed, aggressive traders played a much more important role in triggering the crises than is widely understood.
The paper, titled “Collateral Crises,” examines the important role short-term collateralized debt plays in the financial system. In other papers Gorton has argued that short-term collateralized lending between banks and money market funds is a form of private money within the banking system. It is the medium of exchange within this 'private' system.
In order for collateralized debt to perform this function, the debt needs to be “information insensitive.” Which is to say, the institutions lending the money need to be able to implicitly trust quality of the collateral without investigating to make sure it is sound—largely because a money market fund doesn't have the resources to investigate the quality of every triple-A rated mortgage-backed security that backs up its short-term loan to a bank.
But how can a money market fund avoid being given the worst quality collateral? Here the complexity of the collateral comes in to play. In order to prevent borrowers from engaging in adverse selection—i.e. giving the money market fund junk collateral—the collateral must be complex enough that it isn't profitable for anyone to produce enough information about the debt to carry out any kind of predatory behavior.
In short, predatory trading must be too expensive to work.
“In other words, optimal collateral would resemble a complicated, structured, claim on housing or land, e.g., a mortgage-backed security,” the authors argue.
Mortgage-backed securities were complex and opaque enough that they made ideal collateral. A party on one side of a collateralized loan could count on the fact that the other side was just as ignorant about the collateral as he was.
Although Gorton and Ordonez do not go into the collateralized debt obligation market, it is easy to extend the argument to CDOs. The higher the level of complexity, the more expensive it would be to engage in predatory collateralization. A CDO would be even more “information insensitive” than a MBS. And a CDO-squared would be even more so.
The trouble is that the lack of information means that borrowing occurs against good and bad collateral. This leads to a credit boom, with “blissful ignorance” leading to increased consumption and more lending. Over time, ignorance about the quality of collateral and the financial health of the lenders and borrowers becomes more and more pronounced.
It would stand to reason that a type of Gresham’s law would develop in this kind of situation. Bad collateral would drive out good collateral. Once someone starts to figure out how to affordably detect collateral quality, he would begin to hoard high quality collateral and trade with only low quality collateral. Or, even more aggressively, he might begin to explore ways to short the low quality collateral.
Gorton and Ordonez talk about “aggregate shocks” inducing the production of information about credit quality. But I’m not sure we need any shock at all. All we need is a few guys to see an opportunity to decide to trade on the decay of information about the collateral.
In other words, you just need a few guys at hedge funds or on trading desks at investment banks to find a way to acquire or produce information about credit quality. They might not even need accurate information—just a hunch will do.
In the initial stages of this process, a few well-informed traders enter the market with predatory trades based on credit quality. Recall how much research guys like John Paulson and Michael Bury did about the securities they shorted. These guys had decided that it wasn’t too expensive to produce information. They were arbitraging ignorance.
Once other market participants realize that some people on the other sides of their trades are well-informed predatory traders, the MBS collateral flips from “information insensitive” to “information sensitive.” Lenders can no longer assume that they aren’t receiving low-quality collateral.
But not everyone can develop into a well-informed trader. For many it makes more economic sense to withdraw altogether—to stop lending against the collateral pool corrupted by predatory behavior. The result is that liquidity dries up and credit contracts. Margin calls can make the problem even worse, as lenders demand higher margins or additional collateral to make up for declining market value.
Doesn’t this sound like a pretty good description of our financial crisis?