Chart of the day: Greek bonds and CDS

By Felix Salmon
February 25, 2010
today's NYT piece -- that CDS spreads gapped out, with a nasty effect on Greek bond spreads -- isn't really borne out by the facts. What this chart says to me is that both CDS and bond spreads increased pretty steadily over a period of two or three months, as perceptions of Greece's creditworthiness deteriorated. And that Greek CDS spreads were pretty flat on either side of the introduction, in September 2009, of the iTraxx SovX Western Europe index.

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

GR_CDSBND0210.gif

Many thanks to the wonderful Stephen Culp for putting this pretty chart together for me. It shows pretty clearly, I think, that the narrative in today’s NYT piece — that CDS spreads gapped out, with a nasty effect on Greek bond spreads — isn’t really borne out by the facts. What this chart says to me is that both CDS and bond spreads increased pretty steadily over a period of two or three months, as perceptions of Greece’s creditworthiness deteriorated. And that Greek CDS spreads were pretty flat on either side of the introduction, in September 2009, of the iTraxx SovX Western Europe index.

More generally, this chart shows that the fixed-income markets are working in exactly the way that they’re meant to work, and that the CDS market is the Greece grease enabling them to do so. In an efficient bond market, the secondary-market yield on any given credit’s bonds is very close to the rate at which that credit can issue in the primary market. It can fluctuate up and down, but that changes the price of credit more than the availability of credit.

That’s exactly what we can see in this chart: Greece was able to issue the whole time, at steadily-higher spreads. And anybody who remembers the credit crunch of 2007-8 knows that that is just about the best possible outcome that can be expected, especially when you’re dealing with a borrower who (a) has €25 billion of financing needs in the next three months; and (b) has been so good at hiding its true debt and deficit figures in the past that they’ve lost the trust of the markets.

In markets which, as we’ve seen, are prone to panic, it’s pretty obvious what one might expect to happen in such a situation: investors would be likely to start selling their Greek bonds en masse, and there’s no way that Greece could borrow another penny on the open market.

In the event, neither of those two things happened — thanks largely to the CDS market. You don’t need to sell your Greek debt if you can hedge it in the CDS market instead, where there’s evidently a pretty deep group of investors willing to accept hefty insurance premiums over the next few years and bet that there won’t be an event of default. It’s a much easier way of making money than buying Greek bonds outright, which requires a lot more cash up front.

In turn, the ease of hedging marginal Greece exposure in the CDS market made helped to ease the fears of investors active in the primary market, and Greece has continued to be able to issue debt without interruption.

So rather than demonize the CDS market and blame it for Greece’s current woes, let’s place the blame firmly where it belongs — with Greece itself, and its profligate ways. And let’s thank the CDS market for adding enough liquidity to the bond market that Greece’s fiscal woes didn’t become a major liquidity crisis.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
Comments
4 comments so far

Have rules been established to require the entities writing these CDSs to have sufficient reserves to pay them out if Greece defaults? Or could this be AIG all over again, with Germany forced to take the role the U.S. Treasury did?

Posted by expatsp | Report as abusive

I find the fact the scales are the same, and so the CDS on Greek bonds is basically the same as the spread with bunds quite interesting.

So a buyer of secondary Greek debt gets X bp higher yield on Bunds, but pays it all back to get the safety of Bunds (in fact safer, as Bunds have about a 45bp CDS cost).

This seems to make sense, assuming you believe the payers of CDS will pay up in the event of a default and Bunds are correctly priced.

Do you know if CDS pay out if Greece was to leave the euro and renegotiate its debts into Drachma?

Posted by mjturner | Report as abusive

These two charts just show as you said the deterioration of Greece’s creditworthiness. It shows a certain level of correlation but it does not show anything, about causation, which would support the thesis pro or against CDS. We would need to see the volume of CDS.

Posted by M.G.inProgress | Report as abusive

Agree with MG. I don’t see how these charts tell us anything about the effect of naked CDS positions on the cost of borrowing for Greece. The CDS and bond spreads are linked by an arbitrage-based parity relationship, they have to move in tandem just like spot and futures prices for an index or commodity. The chart tells us about changes in the cost of borrowing, nothing more.

Posted by rajivsethi | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/