Or someone else who is equally on top of the bond market.
[UPDATE: an anonymous commenter quickly backed some market values out of the swaptions market. Bravo! I’ve added the commenter’s calculations to the table below, which only had the first two columns when I first wrote it.]
My problem is this. Suppose you ask me to predict what the 10-year nominal Treasury bond rate will be four years from now. I would give you this probability distribution:
| Percentile | Interest rate (my view) |
Interest rate (market view) |
|---|---|---|
| .05 | 2.0 | 1.1 |
| .20 | 3.5 | 1.9 |
| .50 | 6.0 | 3.2 |
| .80 | 8.5 | 5.0 |
| .95 | 10.0 | 7.1 |
What this table says is that I think that there is a five percent chance that the ten-year will be at 2 percent or less four years from now. Call that the Roubini scenario, because it means we are in one godawful recession. I think there is a twenty percent chance that the ten-year will be at 3.5 percent or less. I think there is a fifty percent chance that it will be at 6 percent or less–based on something like an expected inflation rate of 3.5 percent and a real rate of 2.5 percent. I think there is a twenty percent chance that the ten-year will be at 8.5 percent or more, because of all the debt we are running up. And there is a five percent chance that we will see double-digit interest rates, again because of all the debt we are running up.
What I need Felix to do is find the best yield curve data and the best estimate of implied volatility and back out what the market thinks these probabilities are. [commenter did this]
My guess is that the market is skewed much lower than I am, both in terms of the expected value of the ten-year and in terms of volatility. What that means is that I should be buying puts on long-term Treasuries. One way to sort of do that is to take out a mortgage on my house. But I think I an be persuaded that the transaction costs on doing that outweigh the tax benefits, so that puts on Treasuries are the way to go.
[Based on commenter’s numbers, I was right about the expected value, but I am not so right about the volatility. There’s roughly as much implied volatility in the market as there is in my numbers. What that says is I should not be buying puts so much as just flat-out shorting Treasuries. As another commenter points out, there is an exchange-traded fund, TBT, that does that. I should emphasize that (a) the point of this blog is not to make investment recommendations; and (b) I have not put any of my own money into TBT, as of this date.]
My larger question is: what is the market thinking? Ordinarily, when I disagree with the market, I assume that I am the one who is stupid. But I have a hard time drawing that conclusion right now. Is there anyone out there who can tell a persuasive story to justify the current Treasury yield curve and implied volatility?
READER COMMENTS
anonymous
Mar 27 2009 at 9:45am
Arnold,
Long time reader, first post.
Below is an implied distribution from the swaption (options on swaps) market, adjusted for 20bps of 10yr treasury spreads:
0.05 1.07
0.20 1.92
0.50 3.21
0.80 4.95
0.95 7.11
Mark
Mar 27 2009 at 11:53am
Arnold,
Look at the price changes in ETFs that are shorting the ten-year treasuries. TBT is an example.
global macro trader
Mar 27 2009 at 12:06pm
Here’s how i understand it: treasury yields are low because the Fed has pledged to buy the stuff and promised to keep the yields capped.
p.s: I’m a big fan of this blog. Thank you very much for taking the time out to write it!
Methinks
Mar 27 2009 at 12:15pm
Could it be because the market thinks that the U.S. economy will continue to perform better than other economies, maintaining strong demand for U.S. treasuries?
hutch
Mar 27 2009 at 12:33pm
i don’t know how we can call it a true market when the fed buys and talks about buying so much. wouldn’t the rates be much higher if it weren’t for fed purchases?
Charlie
Mar 27 2009 at 12:50pm
Can someone please explain where annonymous’s calculations come from? I think I understand how to back out a distribution, but where do I find quotes from the swaption market?
Peter Lentz
Mar 27 2009 at 2:00pm
No detailed rationale, but from an intuition standpoint, you are in good company. Willem Buiter:
“It is surprising that even at a horizon of 5 years or more, the markets are not yet pricing in a distinct possibility of double-digit inflation in the US. The announcement of QE in the US did weaken the external value of the US dollar, but long-term sovereign interest rates fell for the maturities targeted by the Fed (two to 10 years) and did not rise materially for longer maturities. At some point, probably not too far into the future, the future inflation expectations effects of QE that is unlikely to be reversed when required to maintain price stability, should overcome the immediate demand effect of the Fed’s QE on the prices of longer-term nominally denominated US sovereign debt instruments.”
http://blogs.ft.com/maverecon/2009/03/fiscal-dimensions-of-central-banking-the-fiscal-vacuum-at-the-heart-of-the-eurosystem-and-the-fiscal-abuse-by-and-of-the-fed/#more-925
Christian G. Warden
Mar 27 2009 at 2:25pm
Thanks, Arnold. Can you do the same thing for longer-term treasuries? Should we be buying TBT or PST?
Elvin
Mar 27 2009 at 5:11pm
And don’t forget foreign demand. I believe several Wall Street economists estimated that foreign demand for treasuries kept interest rates at least 1% lower than they should be.
ajb
Mar 27 2009 at 5:29pm
I think the market realizes that the longer the crisis, the worse the rest of the world gets relative to the US. We’ll be bad but they’ll be worse and foreign investors know it.
Parmenion
Mar 30 2009 at 4:35pm
Risk premium – you lose money in the likely bad states of the world (US=Japan) and make in other, less worrisome states.
Comments are closed.