And I get to pwn him, in the nicest possible way of course, in pointing out the erroneous thinking underlying this blog post.

Dr Krugman's analysis is this: "Has the risk of a US government default risen? Probably. Nonetheless, the people buying these contracts are crazy. A world in which the US government defaults would be a world in chaos; how likely is it that these contracts would be honored?"

What Dr Krugman has just revealed is why he he might not cut it as a speculator. Or at least, a speculator in the credit markets.

And he may even want to give it a rest commenting on credit default swap matters until he has straightened out some fundamentals. Firstly, he appears to be laboring under the the belief that CDS are insurance contracts.

Whilst it is true that in extreme circumstances they behave like insurance contracts - by paying out the buyer of credit protection if a credit event occurs - in no other respect are they like insurance contracts. In fact, the entire legal infrastructure supporting CDS has been designed to keep them from being recognizable as insurance contracts, and instead have them treated as financial derivatives. CDS contracts are synthetic bonds - the economics of selling protection is literally no different from buying a bond. If the bond issuer defaults, you get the shaft. The key difference is that as a bondholder, you lose money you had already invested. As potection seller, you have to stump up the amount payable - as it turns out, more or less equivalent to that which a bond holder will have written off.

So back to Krugman. With this question and answer - Has the risk of a US government default risen? Probably. Nonetheless, the people buying these contracts are crazy - Krugman has, ironically, validated the speculative trading strategies he is deriding as "crazy". If the default probability has risen, bought protection has increased in value, and therefore these are money-making trades. Correspondingly, when default probabilities rise, so too do the yields on bonds. So another way of looking at the speculation on Uncle Sam CDS is that people are betting that Treasuries issued in the months and years to come are going to pay a higher coupon. Given the fact that we are in a liquidity trap - something Paul Krugman is very hot on - it strikes me that going long protection (short credit) is an apparently decent one-way punt. And even if the premise turns out to be incorrect, and Treasuries continue to offer anemic yields, the downside to the CDS play certainly appears pretty limited.

Dr Krugman's latter rhetorical question - A world in which the US government defaults would be a world in chaos; how likely is it that these contracts would be honored?" - also, ironically, further validates the trading strategy. Because it is so unthinkable that the US would actually default - it is a tail-risk that near-as-damnit does not exist - the trading strategy of betting on an increase (or decrease) in default probability is the rarest of trades that can almost certainly never turn catastrophically wrong. And if it does, there are frankly far more serious issues than the solvency of your counterparty to worry about.

Dr Krugman's basic misconception - that CDS contracts need to be considered as insurance products - is unfortunately quite widespread. Joe Nocera, in his celebrated article on AIG, wrote this: "But because credit-default swaps were not regulated, and were not even categorized as a traditional insurance product..." (emphasis mine). Not regulated is one thing, indeed one very big thing, but not categorized as insurance is both tautological (were CDS insurance products, they would have been regulated) and wrong-headed. The Wall Street firms that make markets in CDS are not insurance companies. CDS are, functionally, far more like securities. When a default occurs, insurance-like mechanisms are triggered, but these mechanisms are designed in accordance with and governed by applicable securities laws and regulations. Just ask ISDA.

Heh.