Believe those who are seeking the truth. Doubt those who find it. Andre Gide

Friday, February 4, 2011

Time to short treasuries?

U.S. Treasuries over the last two years have served as sort of a safe-haven for investors (something that still has gold bugs scratching their heads).  But with the worst of the financial crisis over, and growing evidence of U.S. and world economic expansion, there is good reason to believe that long-term real interest rates are likely on the way up (reflecting the increasing world demand for investment).

Ceteris paribus, higher real rates also imply higher nominal rates. That's bad news for treasuries. And though the Fed has promised to keep inflation in check (around 2% per annum), the market might have different ideas concerning the Fed's willingness and/or ability to deliver on its promise. Market expectations of inflation appear to have risen lately. Via the Fisher relation, one would expect this to put further upward pressure on nominal interest rates. Again, this is bad news for treasuries.

Note that I am not personally making any forecast about where interest rates are likely to go in the future. All I want to say is that IF you believe nominal interest rates are likely to continue their way upward, you may want to play this by shorting U.S. treasuries. And an easy way to do this is to go long on the Proshares Ultrashort 20+ Treasury ETF; see recent performance below (on Canadian exchanges, try ticker symbol HTD).

What could go wrong with this trade? Well, the fact remains that U.S. treasuries are likely to retain their role as a safe-haven instrument, at least for the near future. So, surprise events in sovereign debt markets, for example, may very well make TBT tumble again. And then there's the Middle East...what could possibly go wrong there?


  1. So are you saying that investors are currently overweight treasuries?

    Anyways, I pretty much agree with this trade. The safe haven destination of choice seems to be moving away from US to the Australias, Brazils, and Canadas of the world, at least that's what Bill Gross is predicting:

  2. JP: If they're American investors, they're likely just overweight.

  3. I appreciate learning about the new security - wasn't aware of that one.

    But really, I think one could end up short any American market and play this one. The risk-free rate of Treasuries is the benchmark rate for most securities.

  4. David,

    I don't blame you - a good 2/3rds of my post didn't show up.

    I'll boil it down like so. Since the Treasury rate is the benchmark rate for pricing most assets, including other bonds, stocks, options, etc. an increase in the benchmark rate will cause the prices of all these securities to drop. The sensitivities will depend on the extend to which the interest rate risk is already priced in.

  5. Prof J:

    Weird that some of your original post did not appear.

    In any case, I sort of figured that this is what you were saying. But I'm thinking that any inflation will ultimately manifest itself in higher nominal stock prices too. So shorting stocks in this environment is not the same thing as shorting nominal treasuries.

  6. >> What could go wrong with this trade?

    It will be completely wrong when your organization is rigging the market. Tell your boss stop channelling off money to the primary dealers then your trade will work.

    >> Well, the fact remains that U.S. treasuries are likely to retain their role as a safe-haven instrument, at least for the near future.

    Anyone who thinks investment based solely on Gov't promise to pay is a safe-heaven needs to have their brain examined for sanity. Government debt will never get repaid, it's never intended to be. Gov't will not default only currencies will get debased which is happening right before our eyes, thanks to great minds like you in your organization.

  7. David,

    Yes, I agree this security that you found that allows to short nominal treasuries is quite dandy.

    I've almost finished a first draft of a paper looking at the "right" measure of inflation vis-a-vis the effect of inflation on the equity premium. So far I've found no relationship betwixt inflation and the S&P 500 nominal excess return. I think this means that unless the inflation is unexpected, the overall effect has already been priced in. Inflation leads to higher nominal discount rates, but also higher nominal growth rates, thus ultimately the effects cancel (more or less).

    The interesting thing is that higher measured inflation (depending on how it's measured) lead to higher subjective risk premiums on equities. In other words, the real effects of inflation (in aggregate) on stocks appear to balance out, but people become more nervous and so attach a higher risk premium which reduces stock prices. Binky might say people become underweight in stocks...

  8. Prof J:

    No effect of inflation on the EP. What is your sample period? (Do you include the 1970s?)

    It is evidently an empirical fact that high inflation rates are correlated with variable inflation rates. Loosely speaking, there is more uncertainty associated with higher inflation rates. So I guess from this perspective, the higher premium on stocks you identify should not be surprising. I would be surprised, however, to discover that this premium was limited to stocks.

    PS. On

  9. TBT is not a good way to short Treasuries, assuming your expected holding period is more than a day.

  10. There are (now) lots of explanations online; Googling returned this as the top result:

  11. I guess what you meant to say is that TBT is not the best way to short treasuries. Perhaps so, but I'm pretty sure that TBT will generally rise along with nominal interest rates. Indeed, it's behavior appears perfectly consistent with this view.

  12. I like your post and I want to recommend many of my colleagues to read this.

  13. It is investing in securities that are currently betting against the bottom? Because if the piles of money from TNS should fly stocks.