I am flattered that the economic of contempt guy* wrote me a long e-mail about my last CDS post.
The e-mail is after the jump.
my comments on his comments in [brackets]
I am flattered by the attention. Really, I stress, I only recently learned what a CDS is.
I wanted to respond to your comments about CDS, but I hate typing in those tiny little comments boxes, so I figured I’d email my response instead. I’ll admit to a little bias on CDS, since I was involved in all the negotiations over the first standardized CDS contracts back in the early 2000s. But I’m much more critical of CDS (especially the dealers’ role, as I’ve noted in my own blog posts) than most of the CDS market participants, so I think it evens out.
“[M]y main argument is that the way to eliminate naked purchases of CDSs is not to ban them directly but to ban cash settlement CDSs so only ignorant people who have never heard of Delphi (like me) buy naked CDSs.”
Keeping in mind that I think it’s a bad idea to ban cash settlement in the first place, I don’t see how you could ban cash settlement without also requiring protection buyers to own the underlying. Just banning cash settlement could (and would) lead to situations where there are more protection buyers than bonds available for delivery, even if every protection buyer believed when he purchased the CDS that he was, or would soon be, a legitimate hedger. If you’re going to ban cash settlement, then why allow this kind of imbalance to exist in the first place? That’s just asking a CDS settlement to become what we called during the Delphi debacle, “bond market musical chairs.”
[I see no reason to ban people from doing something that hurts only them. Under my proposal it would be stupid to buy a naked CDS. This will become clear after another default (evidently it became clear already once when Delphi defaulted). Now keeping track of who owns what seems to be constly difficult and invasive. In general I think adults should be free to do stupid things provided they don’t hurt other people. If I have a physical settlement CDS and the underlying instrument, I don’t care if I get paid in full because of the CDS or because people with naked
CDSs bid the price of the instrument up to the valued of its scheduled payments even though it is in default. I think my regulation will have the same effect as requiring people to own the instrument if they own the CDS and it seems to be much less burdensome (need to verify who owns what only after a default, so rarely, and the CDS writer and holder do it so they don’t bother the SEC). Now I don’t believe in the efficient markets hypothesis. As far as I can naked physical settlement CDS can’t ever convert into any cash money payment if there are more of them out there than the underlying instrument. I assume they will trade at a positive price for a while, because I assume investors will make every conceivable error. However, I also assume that they will wise up and essentially no one will be silly enough to buy a naked physical settlement CDS after a few more episodes of financial musical chairs. In any case, I don’t see as a legitimate purpose for an invasive regulation the aim to prevent fools from being separated from their money].
“Your argument against banning naked CDSs lacks two things: 1) any social advantage of not banning them, and 2) any explanation for why you expect stupid CDS tricks like AIGs not to happen in any markets where they haven’t happened already.”
As to your first point, the most obvious social benefit of so-called “naked CDS” is price discovery. (I know, I know, “price discovery” is usually a bullshit argument, but when it comes to the corporate bond market, it’s actually legitimate.) Corporate bonds are generally illiquid instruments, making them very difficult to price accurately. The CDS market is extremely liquid, and has dramatically improved the pricing of corporate debt. This was one of the original purposes of the CDS market. The vast majority of CDS contracts are traded by investors who don’t own the underlying bonds. Banning naked CDS would kill the liquidity that has made the CDS market so useful. Without speculators providing liquidity by trading naked CDS, legitimate hedgers would have to pay significantly more for protection.
[here Mr economics of contempt is asserting a very weak form of the hypothesis that markets aren’t efficient but they are better than nothing. I am not at all convinced. In particular I think the pricing of debt implied by the CDS market was totally wrong. If markets were efficient, then it would be very useful to look up a market price. If they are not efficient but not totally massively utterly wrong except as often as a stopped clock is right, then price discovery is socially useful. I don’t think that market price discovery is socially useful.
Recall it can be very hard to beat the market even if market prices are far from fundamentals (shown by Shiller and Summers ages ago).
Also I’m not sure that any socially useful purpose is served by knowing exactly what bonds are worth. I’d say a socially useful interaction of bonds and financial markets is that people can buy a balanced portfolio of bonds — a bond index fund when they are young and sell it when they retire. Obviously that’s not what financial operators are doing. They care about small brief miss pricing, because they are trying to beat the market with active trading strategies. I agree this is perfectly legitimate business, but I don’t think it is socially useful at all (I think the prices of actively traded assets in which extremely smart people are trying to outguess each other are further from fundamental values than are the prices of less actively traded assets).
My reason for thinking this is that volatility increases when trading volume increases. It seems absolutely implausible to me that the volatility is rapid change of fundamental values. I think it is noise and I think there is more noise in prices the more liquid markets are. Now I admit my ignorance, but I’m waiting for someone to point to evidence that I am wrong.]
I don’t understand why people are so quick to denounce “speculators” when it comes to CDS. Speculating is not inherently evil. The options market doesn’t have any sort of “naked” ban — investors write calls without owning the underlying all the time — and that market seems to work just fine. Similarly, you don’t have to own the underlying securities to buy an ETF, nor do you acquire a direct interest in the underlying securities when you actually purchase an ETF. It’s pure speculation. Speculators provide the liquidity that allows “real” investors to move in and out of their positions efficiently. In fact, buying an ETF that tracks the S&P is just like buying an index CDS contract like the immensely popular CDX contracts, which a ban on naked CDS would eliminate. If your argument for banning naked CDS rests solely on the “speculation” issue, then you have to explain why you don’t also favor banning ETFs, “uncovered” call and put options, and so forth.
As to your second point, banning naked CDS would have no effect whatsoever on the CDS contracts that brought down AIG. They’re completely unrelated issues. AIG failed because of CDS it had written on CDOs and ABS. The protection buyer in all of those contracts owned the underlying CDO or ABS (these CDS contracts aren’t actively traded), so the issue of naked CDS is completely irrelevant to AIG’s problems. AIG essentially failed because it took on crappy CDOs and MBS — just like everyone else. So restricting the use of single-name and index CDS in response to AIG’s failure would be useless at best. In the more likely scenario, it would be extremely counterproductive, as banks would be stripped of an enormous amount of regulatory capital relief in one fell swoop, forcing another round of forced liquidations to meet capital requirements.
Anyway, that’s my argument. Hope it helps.
Economics of Contempt
* I am not making a sexist assumption that he is a guy. He had a wife at a time when it was impossible for women in the USA to have wives.