Why a Greek default won’t ever be priced in

By Felix Salmon
June 17, 2011
skeptical that Greek bonds were pricing in a massive default, despite the fact that the likes of Martin Feldstein were saying that they were.

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

Back on May 26, I was skeptical that Greek bonds were pricing in a massive default, despite the fact that the likes of Martin Feldstein were saying that they were. But even if they weren’t back then, we’re getting closer now. The numbers, courtesy of Peter Rudegair: Greek CDS spreads were 1,400bp on May 26; now they’re more like 1,900bp. Greek 10-year bonds were yielding 16.4% back then; they closed today at 18.3%, with prices at about 50 cents on the dollar. European stocks have lost 5% of their collective value since May 26, and the Thomson Reuters default-probability calculation is now over 90% for Greece.

Which means it’s time for an article saying that default is not priced in yet at all:

Today’s rising bond yields, stratospheric insurance costs and heavily pressured stock prices may only be a taste of what could come if euro zone leaders fail to halt Greece’s decline and ring-fence it from others.

In short, a lot of markets and bond holders have not priced in some of the worst outcomes…

“The main case that people are assuming is that when push comes to shove Europe and the IMF will step up,” said John Stopford, head of fixed income at Investec Asset Management…

Stopford says asset price moves actually only reflect caution.

“The market is taking risk off rather than necessarily pricing in catastrophe,” he said.

There are three factors at work here. The first is that no one has any idea what would actually happen in the event of a Greek default. In order for markets to be pricing in a default, traders would have to be buying debt now on the expectation that if Greece defaults the price of its debt would not fall further. I don’t think anybody’s doing that yet — as Stopford says, they might be taking risk off, but they’re not expecting catastrophe. If Greece were actually to default, I’m pretty sure that markets would fall further.

The second factor is the short-termism of market reporting. Journalists have a tendency to look at moves rather than levels when it comes to markets, since it’s the move which is the news of the day, rather than the level at which assets are trading. The conceit of market reports is that if an asset price moves, then it’s likely to have done so for a reason, probably related to some news or other. Meanwhile, if an asset price doesn’t move, then there’s a good chance that there was no news of interest at all. Neither of these things are true. But the result is that pundits are much more likely to think that the market is pricing in catastrophe after a big fall in bond prices from say 90 cents to 70 cents on the dollar than they are after a rise from 45 cents to 55 cents.

Finally, the markets are bad at pricing in catastrophic events even after they happen. Lehman collapsed in September 2008; it was another six months before the market reached its bottom. If Greece defaulted tomorrow and the bonds didn’t move very much, that wouldn’t mean a default was priced in, so much as that the market was paralyzed and didn’t yet know what the consequences of Greece’s actions were going to be. Look at Greek bond prices three or four months after a default: that will give you a much better indication of what the effects of the default were on markets.

In general, the market is much better at pricing in small events from Mediocristan than it is big events from Extremistan. A company releases quarterly earnings of 17 cents per share, or the monthly payrolls number comes in at 210,000 — these things can be priced in. Lehman goes bankrupt, Russia defaults, the House fails to pass the TARP bill — these things you just need to live through in order to know how markets will react. So long as the Eurocrats are all still unanimous that a Greek default is unthinkable, you can be sure that markets will fall if and when such a thing happens.

10 comments

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/

Some thoughts:

1. CDS are rising like crazy, so clearly they’re signalling higher risk of default. What would happen if there was no CDS instrument? I’d venture to guess the 10-year Greek bond rate would be much higher than the near-18% it currently is at.
2. If an unplanned default happens, for all intents and purposes, CDS is made moot and rates will rise, once outcomes have been determined. But if a planned, orderly default is processed calmly, certainty is injected into the system, and rates will fall.
3. Yield curves are inverted: that should be a fairly important indicator of serious issues, particularly when the 2-year rate is 75% higher than the 10-year rate. In a way, bonds have priced short-term operations as likely to result in a default.

Posted by GRRR | Report as abusive

If it helps, here’s some perspective on the likelihood of default (as measured by CDS spreads) and 10-year bond yields.

At least Greece is still tracking along with the general relationship shown in the chart at the link, which suggests that bond yields are indeed pricing in the likelihood of default.

Posted by politicalcalcs | Report as abusive

Here are the links that were scrubbed from the comment above:

CDS spreads vs 10-year bond yields:

http://tinyurl.com/3jpb3xy

CDS spreads vs Probability of Default within 5 years:

http://tinyurl.com/3lxm5kn

Posted by politicalcalcs | Report as abusive

i am struggling to make sense of this post. is the point really what the headline says “why a greek default won’t ever be priced in?” price into *what*? greek bonds? the s&p500? gold? pork bellies? the price of rice in china?

it seems obvious to me that *some* chance of default is priced into both greek bonds and greeks CDS from greek bond yields/prices and greek cds spreads. are markets correctly assessing the probability of default or the magnitude of the losses (the recovery value) if a default occurs? maybe, maybe not, but that is a question of degree, not a binary question.

felix’s quote from the reuters article doesn’t seem to say anything other than one guy’s opinion; i’m not sure why he cites it. i also have trouble following the argument felix is trying to make about how there is still room for greek bonds to fall further in price. sure, but does that mean that no probability of default is priced in? of course not.

agian, if the main point is that more needs to be priced in to the bonds, ok, that is a reasonable argument to make, and felix should be shorting greek bonds and/or buying cds. but then felix should tell us what the right price should be (what is the right probability and recovery value?). that’s the hard part…

Posted by alphabet | Report as abusive

I’m continuously baffled as to why the parties currently collecting 15%+ on Greek bonds regard default as “unthinkable”; what the heck do they think they’re getting that rate for, their roguish good looks? They’re replaying the same act that the subprime lenders tried – “whaddya mean there’s a risk of default? We just loaned somebody with no credit $500k at 10%, and that can’t POSSIBLY be risky!”

Posted by al2o3cr | Report as abusive

Felix, I think you’ve missed the point of Jeremy’s analysis. It’s not that the market is failing to price in the effect of a default on Greek bonds: when a two-year bond trades at half its par value, investors are expecting a haircut. The point is that investors have not priced in the *consequences* of a disorderly Greek default. If Greece suddenly stops servicing its debt, investors would start discounting the same outcome in Ireland, Portugal and even Spain. Also, Greek banks would fail, and other European lenders would suffer heavy losses. The ECB would probably need to be recapitalised. And risk premia around Europe would soar. It’s not Greek debt we need to worry about — that’s already repriced. It’s the second-, third- and fourth- order consequences that are the reason to worry.

Posted by petertl | Report as abusive

What al2o3cr and petertl Said. It’s not that they are not pricing in a default–it’s a question of what you mean by “default.”

No one expects Greece to just Walk Away from its debentures forever. Even 2001 Argentina didn’t do that.

What the market expects (because they’re not innumerate) is some combination of (1) haircut, (2) maturity extension, and (3) currency risk.

You can reasonably claim that they expect more of (1) and (2) and less of (3) than they should–I would be inclined to agree. But the market knows that every effort (even silly ones) is (are) being made to avoid (3). The feet are still pushing the brake to the floor, even as the skid extended and the cliff comes closer. Even if you and I agree that Merkel, Sarkozy, and Papandreou are not Piper Perabo, the market is still hoping they could be.

The question is which scenario gives a better expected return, having already paid EUR100 for the bond? Maybe I clipped a coupon or two, and maybe the next one will be paid–but not necessarily on time, and not necessarily in EUR.

So I might take EUR50 right now from a vulture fund to offload the multiple risks. Or I might hold the bond, at which point there are marking questions. (See non-GSE US MBSes, many of which are still, er, optimistically marked.) Or I might hedge with American banks. (See Kash again today: http://streetlightblog.blogspot.com/2011  /06/us-bank-exposure-to-greece-part-3.h tml) and try to ride it out until I know if the car is going off the cliff or just going to dangle precariously on it for a while.

Default–full-stop non-payment of principal or interest at any point–is never going to be fully priced in because it’s not going to happen. Greece is not going to be 1917 Russia.

But it might be 2001 Argentina–restructuring offers, payment delays, major haircuts. And the market is pricing those in quite well for the moment. And those flows leave a variety of possibilities that present second- and third-order economic reasons (duration shifting and convexity expectation changes) for transactions.

It’s those–not a full default–that need to be reflected in the pricing for the market to clear. And it is those that we are seeing reflected.

The market is working as well as it ever does. If you want to worry about that, well, that’s another discussion.

Posted by klhoughton | Report as abusive

Greek bonds are pricing in a default with recoveries at 58-59% of face value. When bonds have a price curve that is inverted, with long bonds at about the same dollar price regardless of coupon level, the long bond prices are indicative of likely recovery levels in the event of a default.

And yes, it’s just the market’s best guess on something uncertain, but it does tell us what the opinion of marginal money is regarding a possible default.

Posted by DavidMerkel | Report as abusive

Greek default is a matter of time not if. This is self evident. Any holder of Euros in Greece with a brain has already moved the money to a safe location and that means there is no money in Greek banks which means that Greek companies can’t roll their debt over as it comes due because there is not money in the banks to borrow. That isn’t going to change unless political unity came to the Euro zone for which the chance is nil. The only buyers of Greek debt are the bureaucrats that run the ECB and they are doing this against the wishes of the German voters a situation that will end fairly soon.

Of course smart money knows this and is just waiting with a finger on a hair trigger to run screaming for the exits. As always not everyone can solvently make it through the exits the players are just brave enough to play chicken.

Posted by Truth_Teller | Report as abusive

Whoever is not pricing it in clearly has a couple screws loose…. my thoughts: http://www.singledudetravel.com/2011/07/ greece-greed-graft-and-the-grim-reaper/

Posted by borisSDT | Report as abusive