by Jonathan Bernstein
In my view, it’s a good thing that the recent Eurozone deal has basically ended the issuance of CDS on sovereign credits. Why shouldn’t big banks and insurers bring back the forgotten art of credit analysis, and look carefully at what they buy?
I know, there is an argument that CDS allow some borrowers who might not otherwise be able to obtain funds, to issue bonds. But in the end, lending money to debtors who can’t repay ends badly for all concerned.
Sovereign CDS Exist to Make Possible Loans that Really Shouldn’t Be Made
If the Eurozone wants to help its poorer members, it should do so with outright grants or other funding that is workable longterm, rather than go through a lending charade. If it doesn’t, tell the truth and move on.
As many readers know, the Eurozone structured Greece’s 50% haircut as “voluntary” so that banks writing CDS on Greek bonds would not have to pay off. Having seen that, investors won’t bother buying worthless CDS “protection” for sovereign bonds going forward. Therefore sovereign issuers who can’t sell bonds on the strength of their own creditworthiness, won’t be able to scrape through by having underwriters issue CDS to cover the risk.
A centralized CDS Exchange Would Be More Transparent But Market Size Would Have to Be Limited
Writers such as John Mauldin and Barry Ritholtz, to name two, say that we could reform the CDS market by requiring that all CDS be traded on a centralized exchange. They make a good point – an exchange traded market would be safer than the current system — but I have difficulty envisioning a deep and liquid market for exchange traded CDS. I’d rather explore ways to wind down existing CDS and ultimately ban them, but if you are going to keep them, I think the exchange traded model is the best choice.
Of course if all CDS traded on one centralized exchange, risk would be decidedly lower than in the current model. CDS writers would have only one counter party, the new CDS exchange, whereas banks now hedge by making canceling trades with other banks. This exchange traded model would also increase transparency, since the exchange and financial regulators could monitor CDS writers’ exposure in real time. The trouble is, that such an exchange could not easily protect itself from default risk in size. Let’s see how this would have to work.
Suppose you are chief risk management officer of this exchange, call it the CDSE. You’ve lived through the AIG fiasco and the Greek restructuring, so you know this much: if a large asset class gets unglued (MBS in the case of AIG, or PIIGS sovereign debt) many of the banks exposed will all experience liquidity issues, simultaneously. Lending would slow down and a systemic crisis would result.
Therefore, as the CDSE’s CRO, what would you do? You would make sure that the exchange traded CDS contract allows for very strict position limits, not only by each credit, but by asset class being protected. In this example, you would probably limit CDS writing on, let’s say, all sovereign credits that are not backed by a printing press, rather than just Greece, or even just the PIIGS countries. And these limits would stand in addition to the more traditional position limits for each contract that each CDS writer would have to abide by. The maximum hedgeable size of the sovereign CDS market might well be smaller than the current outstandings.
Without strict position limits, the CDSE couldn’t count on the ability of the CDS writer to buy back the CDS he wrote, or to post adequate collateral, during a systemic crisis; unlimited position sizes could get unmanageable. In crisis mode the cost of the protection would skyrocket to some amount commensurate with the expected bondholder haircut. Buybacks would be prohibitive if they could be accomplished at all.
Furthermore, in a banking crisis, you might have to jostle with bank regulators (like the FDIC and its foreign equivalents) to get your mitts on the amount of cash collateral you would need to feel comfortable. Therefore you’d have your own, independent credit analysts monitoring the exchange traded credits, and you would be raising margin requirements rapidly at the first hint of trouble. You can’t afford to be unprepared when real trouble hits, and your margin policy would have to be aggressively pro-active in order to protect the market before a crisis broke.
In this your model would be the way Goldman Sachs (GS) demanded collateral from AIG before the 2008 crisis was obvious to everyone. And like GS, you wouldn’t wait for S&P or Moody’s to downgrade anyone you are exposed to.
I don’t want to overdo the dark side, but I’m harping on crisis conditions for a reason. That’s when sovereign credits often go bad, and in the last 30 years, we have had three such instances: Latin America in the early 80s, Asia in 1998, and now Europe. If you are planning a CDS product to insure against sovereign risk you have to think about crisis conditions before product launch, not after.
Banks Like CDS Writing When they Can Shift the Risk
Banks don’t like the idea of exchange traded product, because it is commoditized and commands lower profit margins than the current model of over the counter, custom product. Nor would banks like the idea of tying up large amounts of cash collateral when writing the CDS under a risk regime that would work for any rational exchange.
In other words, the obvious way to make CDS profitable for the bank – or at least as profitable as banks would prefer — is to shift much of the risk onto someone outside the banking system. In the cases of AIG (and probably the PIIGS), that “someone else” turned out to be the nearest taxpayer.
Which brings me back to my opening statement. If you are going to buy bonds, know what you are buying, because in the event of systemic trouble, credit enhancement may itself become as worthless as the bonds being insured. GS and its ilk were lucky and got bailed out in 2008 when AIG failed. On the face of it the European sovereign bondholders were less lucky, and while they might get bailed out too if the coming European bank recapitalization turns out to be favorable, we don’t know yet how that will turn out.