Naked CDS.  Let me define it first.  A Credit Default Swap (CDS) is a bet that some debtor, say the government of Argentina, is going to default on their debt. When you buy a CDS you are buying a claim to a payment made in the event that there is a default.  When you sell a CDS you are betting that there will be no default and you won’t have to make that payment.

The conventional role of a CDS is an insurance instrument.  If you hold Argentinian bonds and are worried about default, you can buy CDS to insure against that risk.  (In the event of default you are out one bond but you get compensation in the form of your CDS payout.) Naked CDS refers to an unconventional role:  selling a CDS contract to someone who doesn’t actually hold the bond.

There is a very interesting argument that naked CDS can poison the financial well, and I believe it is in this paper by Yeon-Koo Che and Rajiv Sethi.  (I haven’t actually read the paper but I discussed it with Ahmad Peivandi and I think I get the gist of it.  Fair warning:  don’t assume that what follows is an accurate account of the paper, but whether or not it is accurate, it’s an interesting argument.)

Suppose that Argentina needs to issue new bonds in order to roll over its debt. The market for these bonds consists of people who are sufficiently optimistic that Argentina is not going to default.  Such people have a demand for bets on the solvency of the Argentinian government.  Argentina wants to capitalize on this demand.

In a world without naked CDS the government of Argentina has market power selling bets.  You  make your bet by purchasing the bonds.   Market power enables the  Argentinian government to mark up the price of its bonds, selling to the most optimistic buyers, and thus raise more capital for a given issue.  This reduces the chance of a default.

A market for naked CDS creates an infinitely elastic, perfectly competitive supply of bets.  Someone who is optimistic that there will be no default can now bet their beliefs by selling a naked CDS rather than purchasing bonds.  If the Argentinian government tries to exercise its market power, they will prefer to trade competitively priced CDS bets instead.  Thus, the market for naked CDS destroys the government’s market power completely.  There are welfare effects of this.

  1. The downside is that the government is less likely to raise enough to roll over its debts and therefore more likely to default.
  2. But the upside is that people get to make more bets.  Without competition from CDS, there are people who are willing to bet at market prices but are excluded due to the exercise of market power.  This deadweight loss is eliminated.

But how you evaluate these welfare effects turns on your philosophical stance on the meaning of beliefs.  One view is that differences in beliefs reflect differences in information and market prices reflect the aggregated information behind all of the traders’ beliefs. If competition drives the price of bonds down it is because it allows the information behind the pessimists’ beliefs to be incorporated.  If you hold this view you are less concerned about 1 because investors who would have bought bonds at marked-up prices must have been at least partially fooled.

Another view is that beliefs are just differences of opinion, more like tastes than information.  If the price of beer is low that doesn’t make me like beer any less. If this is your view then you really worry about 1 because those pessimists who drive the price down aren’t any better informed than the optimists. The concern about 1 is a rationale for banning naked CDS. But by the same argument you also care a lot about 2. Every bet between people with different beliefs, people who agree to disagree, is a Pareto improvement.

The bottom line is that arguments against naked CDS based on 1 probably also need to account for 2.