Are Greek Bonds Pricing In a Massive Default?

by: Felix Salmon

Martin Feldstein reckons that the market is pricing in a “massive” Greek default:

Even though the additional loans that Greece will soon receive from the European Union and the IMF carry low interest rates, the level of Greek debt will rise rapidly to unsustainable levels. That’s why market interest rates on privately held Greek bonds and prices for credit-default swaps indicate that a massive default is coming.

And a massive default, together with a very large sustained cut in the annual budget deficit, is, in fact, needed to restore Greek fiscal sustainability. More specifically, even if a default brings the country’s debt down to 60% of GDP, Greece would still have to reduce its annual budget deficit from the current 10% of GDP to about 3% if it is to prevent the debt ratio from rising again.

I don’t think this is true. The Greek yield curve is odd: Three-month T-bills yield about 6%, the 30-year bond yields about 10.8%, and the big spike is at two years out where the yield is about 25%. Clearly the market isn’t pricing in any kind of massive default over the next three months. But let’s look at that benchmark two-year bond, since if any instrument is pricing in a massive default, it’s that one. The bond carries a coupon of 4.6% and is trading at 71 cents on the dollar.

Now what would happen to that bond if there was a massive default? Let’s be conservative and say that Greece’s debt-to-GDP ratio will be about 150%, and let’s take Feldstein’s target ratio of 60% as where the country is going to end up. In that event, the face value of Greece’s debt would have to fall by at least 60%, and probably more, given the hit to GDP which normally accompanies a big default.

Clearly, traders pricing the two-year note at 71 cents on the dollar are not pricing in any kind of event in which Greece will swap that note out for an instrument worth only 40 cents on the dollar.

In fact, any priced-in default looks decidedly modest to me. Let’s say that Greece defaults before the next coupon payment, which is due on May 20, 2012. And let’s say that traders in this risky asset want a return of at least 10% if there’s a default. Then someone buying the bond at 71 cents now is betting that it’s going to be worth at least 78 cents post-default. Which implies a pretty modest haircut of just 22% — something which would bring Greece’s debt-to-GDP ratio from 150% to a still-unsustainable 117%.

In fact, any priced-in haircut is even lower than that, since the post-default bonds aren’t going to trade at par.

My point here is that although yields on Greek debt are indeed high, they’re not anywhere near the really distressed levels that we’d expect to see if the market was expecting a massive and imminent default. If you buy a bond at 71 cents on the dollar, that’s cheap, to be sure, but there’s also an enormous amount of downside: If Greece does default in anything but the gentlest possible manner, then you’ll end up losing money.

The current price of Greek debt, then, says to me that traders are requiring hefty returns of 25% in order to pay them for taking the risk that Greece might default. They know that they’ll lose money if that happens, but they reckon that, more likely than not, Greece will muddle through — in which case they will make very good money.

The further you go out the curve, of course, the higher the default probability becomes. According to Thomson Reuters analytics, the CDS market is pricing in a 71% probability of a default in the next five years, and an 83% probability of a default in the next 10 years — both assuming a recovery rate of 41.5 cents on the dollar.

But the Greek single-name CDS market is small and speculative; we have to be a bit careful about drawing too many conclusions from where it’s trading. If you want to protect yourself against default, you’re going to pay through the nose: This particular insurance market is very expensive. But if you’re buying Greek bonds at these levels, you’re not expecting a massive default. Quite the opposite.

Disclosure: None