Arnold Kling  

Derivative Exchanges and Systemic Risk

Panic Puzzle... Congratulations, Paul Krugman...

One suggestion that often crops up is to create an exchange for derivatives, such as credit default swaps. In my view, this probably would solve the problem of counterparty risk. However, the more important problem, systemic risk, would be left unsolved, or perhaps worsened.

A credit default swap is an insurance contract or bet that is tied to a company's credit performance. If I think that XYZ's bonds may default, I might want to buy a credit default swap that pays off in the event of an XYZ default.

Counterparty risk means that if I buy my default swap from ABC, I have to worry that ABC might not perform in the event of an XYZ default. Having an exchange step in to guarantee the transaction would be an improvement.

However, the larger problem is systemic risk. That is the risk that the contingency plans of individuals cannot be executed collectively in practice.

Suppose we had an exchange-traded default swap on XYZ. What are the sellers' contingency plans if they think that XYZ may be headed toward default and the option that the sellers have sold may be in the money? My guess is that the each seller's plan is to hedge its short option position by dynamically hedging, which means taking short positions in the bonds and stock of XYZ. These short positions will get more and more aggressive the closer XYZ gets to default.

The problem with these contingency plans is that they cannot be executed collectively. Collectively, they start a run on XYZ that drives it out of business before everyone can execute the hedge.

The problem with credit default swaps is not (just) that they are traded over-the-counter. The more serious problem is that they entail huge systemic risk. Getting rid of counterparty risk does not solve the problem. The issue is not that any one party is unable to meet its obligations. The issue is that in the aggregate, credit default swaps allow the market to sell more default risk protection than it's possible to produce. This false sense of security leads people to pile on risk, and then when the crisis comes, their protection fails.

Comments and Sharing

COMMENTS (7 to date)
TC writes:


Worries about credit default swaps might be overblown. On Friday afternoon the auctions of Lehman's credit default swaps went off without a hitch. Everyone was worried that the counterparties to the credit default swaps wouldn't be able to perform their end of the contract.

But built into these contracts were collateral requirements. And so this unregulated market of CDS worked.

It seems the government should not regulate CDS since that market is working much better than the regulated markets.

Ian writes:

This theory depends on the willingness of buyers to pay for something they weren't getting. "Insurance" from a party that lacks the means to settle is not insurance at all, and it seems unlikely that clever financiers would fail to understand that.

A much richer story is the one you told before. Those buying CDS probably did so for their regulator's satisfaction, not their own. Using CDS, banks and others were able to take just as much risk as they would have taken without regulatory encumbrances.

E. Barandiaran writes:

I'm back from Chile and Argentina. It seems that you're still struggling to explain the crisis. In this post you refer to systemic risk and write
"However, the larger problem is systemic risk. That is the risk that the contingency plans of individuals cannot be executed collectively in practice."
I hope you can elaborate on this definition because at least there is something missing in it. In all markets for all plans of individuals to be executed collectively, prices have to adjust. Why do you think price adjustments will not balance supply and demand in financial markets? How are they balanced in financial markets when there is a "shock"?
Also I'd like you to explain how your definition of systemic risk relates to other definitions in the literature. I don't remember anyone using your definition.
Finally, I'm waiting for you to comment on both the Nobel Prize to PK and Tyler's post on it. One more request, please comment on Tyler's post about what a bank is--he has been written about the financial crisis for many weeks and he doesn't know what make banks different from other financial institutions!

floccina writes:

Has anyone talked about making it easier to foreclose as a partial solution to the problem? I say this because I believe that the high cost of foreclosing decreases the value of these in doubt mortgages.

Erik writes:

The exchange generally sets the margin requirements. I think your point would argue for setting the margin requirements very high. Perhaps 50% for selling a CDS?

Matt writes:

Isn't this argument relevant to any strategy that employs dynamic hedging?

What are the parallels here between CDS hedging and Portfolio insurance?

Are there any situations in which dynamic hedging wouldn't lead to a cycle of increased volatility if used widely enough?

Arnold Kling writes:

Matt, my intuition about CDS is precisely that they work (and fail) like portfolio insurance.

Comments for this entry have been closed
Return to top