Wednesday, December 10, 2008

"But there are lots of other things that credit default swaps are useful for."

Felix Salmon comes to the defense of CDSs. There's something strange about this constant focus on the products, and not on the people:

"John Dizard wants to kill off the entire CDS market. It does no good, he says, and quite a lot of harm, and we'd all be better off without it.

I disagree. Dizard says there are only "three possible defences for treating the CDS market as a going concern"; in fact, there are more than that, and he misses out the big one, which is that the CDS market has allowed investors, for the first time ever, to hedge their credit exposure. Yes, there's a downside to that -- which is that it becomes easier to simply buy credit protection than to do the hard work of fundamental credit analysis. But CDS by their nature are more liquid than bonds, and it will always be easier to buy credit protection than to sell a bond."

I'm not sure why insurance on mortgages and bonds is inherently indefensible. It seems sensible to me. Buying CDSs for other reasons than mimicking bonds bothers me, but only in the sense that I wouldn't personally come near them because they're risky, and more like a bet. All I ask is that buyers be clearly and truthfully explained the risks in buying them. That's it.

"What's more, CDS prices are a much better indication of credit risk than bond spreads are, for many reasons including the tax treatment of bond coupons and the fact that many bonds simply don't trade. In other words, not only are they more liquid, they're also more transparent. These are good things."

This seems true to me in theory, but I'd like to see more information on how these differences work out in practice.

"But Dizard doesn't concentrate on simple things like liquidity and transparency. Instead, he talks about CDS providing "support for capital raising", which was never something it was designed or even really used for. The whole point of credit default swaps is that they're derivatives: they're not cash instruments to be used for raising capital."

This is something I talked about on Derivative Dribble. Some people understand bonds because they're essentially loans for capital, but don't see the sense in investments that look like side bets.

"Dizard does have a good point that as spreads have widened, banks have seen increasing amounts of money tied up in CDS collateral. That's a concern, and it's a good reason to work hard on compression, netting, and rehypothecation. It's not a reason to kill the CDS market outright."

In other words, ways to free capital for other uses. However, I'd worry about this possibly lowering capital on these investments right now.

"Dizard's other big point, however, eludes me:

Price discovery is a useful economic function; that is the rationale for commodities markets. But CDS are derivative instruments, whose price is "discovered" these days as a function of equity volatility, since buying equity puts is one way to dynamically hedge the illiquid legacy books...
At high levels of default risk and equity volatility, if you hedge the one with the other you get frantic, self-defeating activity.

I'm not sure I understand this, but Dizard seems to be saying that there's a lot of capital-structure arbitrage going on: people hedging equity positions in the CDS market, and vice-versa. That's a strategy which has blown up quite consistently since the summer of 2007, and it tends to require quite a lot of leverage, so I'd be surprised if it was a major factor today. But even if it is, I still don't see why it means the CDS market should be abolished."

I actually don't understand the point at all. It sounds like a weird trading strategy, at best.

"I do understand, however, what Dizard is saying here:

If the default rates implied in investment grade CDS spreads were to occur, the only economic activity would be court-supervised reorganisation. The CDS market has been preventing efficient price discovery.

He's wrong. I just had a long conversation with Kai Gilkes of CreditSights, who confirmed for me that it's pretty much impossible, in this market, to back out implied default rates from CDS spreads. There are so many technical factors in the market, so many reasons beyond expected default that people are buying protection on certain credits, that it's impossible to isolate expected default probabilities. So I don't know what implied default rates Dizard is using, but I do know that they're unreliable to the point of uselessness, since right now CDS spreads tell us precisely nothing about expected default rates.

So yes, if you try to use CDS spreads as a guide to default probabilities, you're not going to get very far. But there are lots of other things that credit default swaps are useful for. So let's not abolish the entire market quite yet."

That's correct. Here's my comment:

Posted: Dec 09 2008 3:30pm ET
I too believe that CDSs can be useful. As to spreads on many bonds right now, the question is whether or not they are moved more by panic than clear analysis of a company's fundamentals. Right now, there's so much uncertainty, that it must contribute as well.

The real question is do you believe that these spreads are accurately predicting default rates, or simply overshooting in our current crisis.

James Grant had a good piece in the FT about default rates recently I believe.

Felix also posted a couple replies:

"Posted: Dec 09 2008 3:20pm ET
My point is that you need to combine default probabilities with an opaque risk-aversion function to get credit risk prices. We can use CDS prices as a great way of pricing risk. But how much of that is default probabilities and how much is something else, we don't know.

No comments: