they didn't sell enough risk

TO BE NOTED: From The Atlantic Business Channel:

"Mar 28 2009, 8:17 am by Charles Davi

The Kling and I on credit default swaps

Arnold Kling and I will probably never agree when it comes to credit default swaps (CDS). Kling and I have had words in the past over CDS, and so have Kling and Felix Salmon. But so long as spirited debate proves interesting to us and our readers, I'm happy to participate in that hallowed, nerd-sport-of-choice: arguing over the internet.

Kling seems convinced that because he cannot conceive of a way to hedge credit risk using the long end of a CDS (the protection seller's end) it follows that CDSs have no "natural seller." In short, his position is the following:

"[N]o institution was in a position to sell credit default swaps as a natural hedge against its other business."

Why would both ends of a CDS need to hedge some risk in order for the CDS to be economically beneficial? I fail to see how hedging is the sine qua non of economic utility. If that were the case, who is a natural buyer of bonds? I'm sure Kling is incapable of answering that question because as a matter of pure logic, any answer to that question is an answer to his, since selling protection through a CDS is economically equivalent to buying the underlying bond (ignoring CDS collateral, which complicates the matter).

In any case, it seems futures and forwards are acceptable means of speculation, but CDS are not. In the case of fuel and other energy and commodity derivatives, there are those in the market who have bona fide economic exposure to the underlying risk. For example, an airline might enter into a swap or a forward contract to lock in a price for fuel, so that it can plan around that price and won't be brutalized by volatility in fuel prices. The other end of the trade could very well be an entity with no bona fide economic exposure to fuel prices. Rather, that entity wishes to speculate on the movement of energy prices. Both benefit through contract in that both get what they want: the airline wants stable fuel prices and the speculator wants the opportunity to profit by expressing a view on the movement of fuel prices.

The same applies to CDS. Certain entities in the market have bona fideeconomic exposure to credit risk. For example, banks. In order to shed this risk, banks will contract with another party, the protection seller, to absorb this credit risk. The bank wants to unload its credit risk and the other party wants to speculate as to the probability of default on and, more generally, the movement of credit spreads relative to the underlying credit. And so, both parties get what they want and the transaction is, at a minimum, economically useful ex ante."

And Felix Salmon:

"CDS: The No-Natural-Seller Meme

I was on a panel last night with Simon Constable of Dow Jones Newswires, and I'm sure that to our lay audience a peculiar exchange in the middle of the conversation must have sounded a bit like dolphin squeaks. He was trying to demonize credit default swaps, and said with great finality and self-assuredness that if you wanted proof positive that they were the spawn of the devil, all you needed to do was examine them objectively, as he had done, and you'd see that they had no natural seller.

I then interjected that of course credit default swaps have natural sellers: any bond investor is a natural seller of CDS protection. We started going around in ever-decreasing circles of mutual incomprehension, until the moderator happily put an end to that particular discussion and moved us on to the next topic. But now Arnold Kling has resuscitated the meme:

There is no institution which, in the ordinary course of its business, takes a position for which selling credit default swaps is a natural hedge...
Credit default swaps allow companies to trade the default risk on, say, a mortgage-backed security. The holders of that security have a natural interest in buying protection. But nobody has a natural interest in selling protection.

I thought I'd dealt with this back in December, but evidently not, so let me try again, this time quoting a little of my Wired article on the Gaussian copula function:

If you're an investor, you have a choice these days: You can either lend directly to borrowers or sell investors credit default swaps, insurance against those same borrowers defaulting. Either way, you get a regular income stream--interest payments or insurance payments--and either way, if the borrower defaults, you lose a lot of money. The returns on both strategies are nearly identical, but because an unlimited number of credit default swaps can be sold against each borrower, the supply of swaps isn't constrained the way the supply of bonds is, so the CDS market managed to grow extremely rapidly.

The point is that there are a lot of very sophisticated bond investors out there, and much of the time they could replicate the risk and return of buying a bond by putting together certain trades in the CDS market -- and get much better liquidity that way. It's not easy to find bonds from certain issuers, but you can always find a broker willing to buy credit protection on any given name.

A bond investor isn't really hedging anything, so it's true that if and when a bond investor starts selling default protection, then he isn't offsetting some opposing risk. But it's simply not true that in order to make a derivatives market work, both sides have to be hedging something. In the CDS market, you can simply have one person, who doesn't want risk, selling that risk to another person, who does want it.

Generally speaking, it's a good thing for banks to sell down their risk, even as it's also a good thing for institutional fixed-income investors to buy risk: that is, after all, their job. Part of the problem in this financial crisis, as I was talking about earlier, is that banks started persuading themselves that they'd sold so much risk that there wasn't any left, even as the amount of risk they had on their balance sheets continued to balloon. That was a serious failure of risk management: they didn't sell enough risk, largely because they couldn't find any buyers (except for AIG, sometimes) for the super-senior risk tranches that they were prone to keeping on their books.

But those buyers weren't absent because there were no natural sellers of CDS; they were absent because the yields on offer on those super-senior tranches were so ridiculously low that no one in their right mind wanted to buy them. So long as there are bond buyers, there are natural sellers of default protection. They might not be hedging anything, in a narrow sense, but they are large, and numerous, and extremely useful when it comes to providing liquidity and price discovery. By all means regulate the CDS market; by all means move CDS trading onto an exchange. (Although that might not make as much of a difference as many people hope.) But let's not kid ourselves into believing that there was no reason for the CDS market to exist in the first place."