Treasury Bonds: The Short of the Century

What is the best way to ride the inflation wave?

There are some alternatives to hedge against inflation. The best know are buying TIPS (still a good opportunity but with limited upside and pegged to core inflation – a measure that does not include the ever growing food and gasoline), buying gold, or, even better, buying into timber assets, particularly international ones, as in the U.S timber assets are quite possibly overpriced already (look at the multiples at which Plum Creek Timber (PCL) is trading, for example).

In my view, however, the risk/reward calls for more aggressive action: shorting the ten year (or longer maturity) treasury bonds. This could be done through RYDEX INVERSE GOVT LONG BOND STRAT INV [RYJUX], but if you can, I would recommend actual direct shorting of the T-Bonds.

Marc Faber called it the short of the century.

Yields on treasuries across the maturity curve are substantially below the historical averages and although they have widened slightly in recent weeks, are still close to all time lows; therefore, treasury prices are close to all time highs (as yield and price are inversely related).

Let’s look more closely at the risk/reward profile of this trade, which looks great:

10 year treasury yields are at 4.19% (6/23/98); the “real” interest rate according to the treasury department is 1.75%, based on a core inflation rate of 2.44%. This is really not much of real return, so it can’t go down much further, particularly if one considers that if inflation was appropriately measured (more on that below) it would be much higher than 2.44%; the “real” real return is probably close to zero. So the downside is very limited.

The upside comes from the inflationary ball that is building up around the world for a variety of reasons (fuel prices on the rise, food prices on the rise all sorts of commodity prices on the rise) and which Ben Bernanke has greatly amplified in the U.S. due to (in my view) very misguided monetary policies: lowered interest rates (expanding the monetary mass) and the devalued dollar. The devaluation of the dollar further fuels the commodity price run and has an immediate direct impact has imported goods (denominated in other currencies) comprise an important component of the CPI.

It is not hard to imagine inflation reaching high single digits or even double digits; this would naturally lead to a increase in the return demanded by investors in treasuries just to keep their real return a the same level. A 5% percent increase in the yield (which would require a similar increase in long term inflation rate expectations) would lead to a 30% decrease in the value of the 10 year treasuries (3%x10=30%; the change in the price of a bond is roughly equal to the simetric of the change in yield multiplied by the duration of the bond which in the case of treasuries because of their bullet amortization is the same as the term of the bond). Not bad!

I would not recommend to try this at home, but for traders who can sell futures for the 10 year treasuries, and are putting down perhaps 5% of the notional amount, with that leverage the returns could be 6X (of course the downside here would be more meaningful as well).

Now another way to look at the situation and reach similar results, without requiring the same extent of acceleration in reported inflation, would be for investors to stop looking at the “official” core inflation but to realize that in fact “real” inflation has been and is higher. I read somewhere the following funny but insightful comment stating that the situation with the way inflation is measured in the U.S. currently is a little bit like parents who invent Santa Claus stories for their kids but over time end up believing in them themselves!

Bill Gross’ [PIMCO] June investment outlook newsletter has some interesting insights on inflation figures in the U.S. Some of the key changes that have been made over time on how inflation is measured and monitored in the U.S. which have had a moderating impact including:

1. Replacing house purchase prices for owners’ equivalent rent (this got ride of the rampant house price inflation in the last 10 years or so, although current it would have deflationary impact).
2. Including adjustments in the weighting of the CPI to take into account substitution effects. i.e. if chicken prices increase and turkey prices don't, I will reduce my consumption of chicken and increase my consumption of turkey.
3. Including adjustments to take into account increases in the quality of products overt time. i.e. the computer I bought today may cost the same as the one I bought five years ago, but the cost per MB has decreased a lot. Actually, 46% of the CPI comprises products durable goods and others subject to hedonic adjustments, even textbooks!
4. Introduction of a distinct concept of core inflation and headline inflation where the first excludes food and gas. Core inflation is relevant to monetary policy makers in that it excludes items which often have temporary fluctuations.

The slight problem with these is that oil and gas prices have been steadly going up for the last six years, so there is nothing temporary about it. Most other countries do not use these “sophisticated” methods, although one country is now revising its methodology to include some of these - Argentina, whose government is suspected of having manipulated their inflation figures for a while with unofficial estimates of inflation running at about twice the official figures (most forecasters expect inflation in Argentina to raise by 25-30 pct).

Bill Gross estimated a while back in 2005 that without the quality and substitution effect inflation in the last decade or so would have been about 1% higher and real GDP growth 1% lower. Also, consider that headline inflation has been running at least 1% higher than core inflation for several months/years; add these up and you are talking about a real difference!

Disclosure: Author holds short positions in T-Bonds

Comments

[fxtrader07]

poor timing , i would say. t-bonds will be the short of the century at one point. but not just yet. Inflation isn understated, true. and the printing presses ran way too easy, also true. BUT: the fed hasn't been overly expansive in its monetary policy over the past years, contrary to what many believe and assume. Second, the demographics (retiring baby-boomers) and the deleveraging and asset-deflation that is going on will exert strong anti-inflationary pressure over the coming 2-3 years. We may well see 10-year yields of 2.5-3% before we get a 5%+ print. remember japan? 1-2% long term yields were certainly regarded by many as a 'short of the century', too. but look, how poorly that short has worked out for more than 5-6 years?
Shorting t-bonds seems to be the obvious 'surefire' trade these days. a tad too obvious, perhaps?
at the turn of 2009/2010 the time might be right. right now i think it will tie up money for quite some time right now when there are way sexier opportunities to deploy it.

[djzvue]

your anti-bond argument has been out there for so long that I believe Marc Faber now suspects it is flawed. The analogy of blackhole physics may be more apt now--the destruction of credit and financial systems (so much more than just cyclical falloffs of growth), moving now from housing into commerical credit, pensions, municipalities, etc could be such a powerful destructive centripetal force that the more visible sparkles of inflation (generated by desperate remedies and their effects) still at work on the periphery will in time be sucked in and extinguished. The struggle between these forces seems almost cosmic, the leveraging you recommend highly risky, anything but shooting fish in a barrel. Money mgr Gary Shilling, who has been good on deflation and the bond market in the past, is now saying buy long T-bonds. He's not a theorist but worth listening too. Thanks for laying out your view.

[User 211635]

It's important to remember that long bonds are also a place people put their money during flights to quality/away from risk. So, in addition to inflationary pressures, these rates have recently been highly correlated with the broad equity markets. When people feel bullish, they pull money out of fixed income and put it into equities, which drives interest rates up. When markets have bad days, you see the reverse.

Another way to play this, by the way, is a new ETF: TBT

[User 215889]

For TIPS inflation adjustment, the Treasury Direct site notes: "The index for measuring the inflation rate is the non-seasonally adjusted U.S. City Average All Items Consumer Price Index for All Urban Consumers (CPI-U), published monthly by the Bureau of Labor Statistics (BLS)." This is not the core rate; it does not strip out food and energy.

[User 215991]

The duration of a T-note is NOT equal to its nominal maturity unless the bond has zero coupon. Coupled with the error about the inflation rates used for TIPS, one concludes this is a poorly researched article.

[CharlieBottle]

I stand factually corrected on both cases and also stand 100% behind the logic and resonablensess of the conclusions of the article:

- Tips are pegged to CPI-U which includes food and gas. Please note that this is in no way central to the arguments in the article. I stand by the main argument that inflation is not accuratelly measured for the several reason pointed, and that the focus on core inflation (which is systematically tauted byt the FED and in the media) MAY (I repeat MAY) have some impact on investors perception of inflation and therefore the required return on their bond investments. This just one a several arguments provided and not the main one by any means.

- In the example provided where I mention duration it should read maturity. Maturity is a VERY rough approximation of the sensitivity of a bullet bond price to interest rates (it has the advantage of being very intuitive which is good for an article of this nature); in longer maturity bonds it does differs samewhat from duration.

Duration (McCaulay Duration) as you point out needs to take into account not just the principal repayment but interest coupon; it is the sum of the cash flows discounted by the yield multiplied by the time cashflow divided by the price of the bond (i.e a PV wighted average maturity). In this case, assuming a 4% coupon and that the bond trades at par it would imply a duration of about 8.4 years, and therefore an estimated change in the price of the bond of 25.2% rather than 30%.

Off course Mcauley duration is also an approximation, to be more precise one should use modified duration and to be even more precise one should use convexity, so you can decide how precise you would like to be. The numbers were used to give an idea of the order of magnitude and this in no way changes the merits of the argument, I believe.

Thank you for adding precison to the discussion.

With regards to the second comment (by "djzvue") on the use of leverage, I just wanted to clarify that he must have misread, on the contrary I did NOT recommend the use of leverage and did in fact highlight the increased risk of levergage in such a trade.

[blink0404]

Interesting take. I believe that investors are just truly confused and uninformed. How could you possibly predict how to short treasury bonds when investors actions are not corresponding with market signals. When the Fed makes statements that they will be tougher on inflation, yields shoot up. When the opposite happens, yields go down. this is exactly the opposite of what should be happening.
Check out this article for a more in depth analysis of what investors are doing and how they are getting screwed by the treasury bonds, which off a lower return than current inflation levels.
www.greenfaucet.com/bo...

[fund mgr]

The discussion is missing the key point of this "short of the century". The Fed can control the fed fund rate, but it cannot control the yield of the US treasury, which is determined by the market. The US treasury rate is further driven by the short-term rate which is FF and the inflation expectation of the market. Thus even if Fed does not raise FF - which translates to higher forecast for future inflation i.e. the residual effect of expansive monetary policy, the long rate will go higher. On the other hand if Fed raise FF rate then long rate will also go higher as well, however the spread between short and long rate might flatten. So which point is not made clear? The long rate for the UST has no way to go but ... where? Safe for a financial crisis again emerging in the near future, which might happen too, you can never say never with this financial climate, however the Fed is determined to save the financial institutions (and the system) so any flight to quality will be relatively short lived. Full disclosure - I am short the UST.