Prior to the financial crisis, few outside of Wall Street knew what Credit Default Swaps (CDS) were. But AIG's collapse and subsequent bailout made the once-obscure derivatives famous: the insurer needed taxpayer money to cover billions of dollars of its CDS exposure. Ever since, their presence in finance has been questioned. In an article from the New York Times yesterday, Gretchen Morgenson appears to argue they should be outlawed -- or their creation heavily restrained. This raises an important question: should regulators reform CDS, leave them alone or kill them altogether?
First, for anyone reading who doesn't know what a credit default swap is, let me quickly explain. Let's say your spouse wanted to buy a life insurance policy so that, if you die, your family will be protected. Now, imagine you're a corporate bond. That life insurance policy would be called a credit default swap. An investor can buy CDS to protect against the failure of a bond, and consequently, the firm who issued it. And she can trade that life insurance policy so it could pay out to whoever owns it. That's CDS in a nutshell.
I think it's pretty easy to dismiss the "leave them alone" option. AIG and others have proven that pretty easily. Gone unregulated, firms can develop enormous exposures without the capital adequacy to back them. If interconnectedness is significant, then bailouts inevitably follow.
So should we eliminate CDS or fix the market? That depends on whether they can be regulated in such a way that they become safer. I think they can.
Morgenson quotes Brookings scholar Martin Mayer saying:
"This is an insurance instrument and it must be regulated on an insurance basis with minimum reserves, instead of making deals that don't even have maintenance margin on them," he said.
Absolutely. I couldn't agree more. Any firm that issues CDS must meet relatively conservative capital requirements to back that exposure. That's where AIG went wrong. He continues:
"And I think it is an instrument that insured depositories ought not to be allowed to hold or trade."
Here, he loses me. If I buy a credit default swap to protect an investment, then my potential loss is limited to what I paid for the contract. If the security performs well, then it becomes worthless. I don't owe anyone anything, I just lose the money I spent on the insurance. How is it harmful if an insured depository institution, or anyone else for that matter, holds or trades CDS? So long as their issuer can cover any payouts, then there's nothing to worry about.
With one caveat: I would also add liability exposure concentration limits for issuers of CDS and other derivatives. For example, you don't want any one firm to have an incredibly large exposure to one segment of the market. That could create interconnectedness issues. If residential real estate collapses, for example, one firm might not be able to pay out incredibly large sums to a huge portion of the market that holds insurance on residential real estate assets. Even with capital requirements place, such a predicament could lead to its failure and over-connectedness. If these losses were spread over many issuers, then the likelihood of those firms being able to sustain losses increases substantially. The need for a bailout would also be far less likely. This speaks to the second, less controversial, part of the so-called "Volcker Rule."
Really, CDS are a very positive financial innovation for investors who want to hedge their securities portfolios. And as long as they're properly regulated, I can't see any reason why they would be so harmful: as market demand for the CDS on a particular security increases, so does the cost of creating more for issuers, if they are required to hold capital to cover their exposure. So the market is already self-limiting if that reform takes effect. Firms aren't going to write more CDS contracts if it becomes too expensive for them to do so. Moreover, limiting firms' concentration will further prevent the possibility that too much CDS is written.