Merrily Rolling Down the Yield Curve

Your humble and chronically bemused blogger drank the Kool-Aid today and added a substantial amount of a specific type of bond to his IRA holdings, transforming zero-yielding cash into a zero-coupon Treasury bond of 8 years duration. This was done while continuing to hold a substantial percentage of our family savings in gold and while being aware that the MIT Billion Prices Project is showing a price inflation rate that makes my 2.64% implied annual yield a loser in real terms.

(Do I contradict myself? Very well, then I contradict myself.)

Here is the thinking behind my latest speculative foray into bonds.

Basically this is a play for modest capital gains.

First, one must understand that when one buys a bond, one's broker will quote you a yield. But the reality is that the transaction is one in which price is the reality. The yield you are quoted is, forsooth, a derived value. It is not a toxic derivative as in CDO-squared sorts of derivatives, but it is a derivative value. In the case of most bonds, which pay coupon interest periodically before the borrower is supposed to return the invested principal, computation of effective yield is more complex than most people realize. One has to make assumptions about such matters as reinvestment returns on the interest payments, for example. Furthermore, standard bond tables assume that the interest is reinvested at the coupon rate of the bond. In a world where short-term money yields nothing, however, that calculation ends up overestimating the real return that a lender receives.

If one's goal in purchasing a bond is to attain a capital gain, the purest bond is one that pays no interest and thus does away with the reinvestment problem. This bond, commonly called a zero-coupon bond, trades at a price with a yield that can be computed with a compound interest calculator. Here is how to make lemonade with zero coupon bonds. First, one should have a non-callable bond. You have to be confident of the maturity date.

This is where Treasuries are almost unique in bond land.

Next, you want a positive yield curve, which is one in which the annual yield is higher as the duration of the bond increases.

So when I called up the broker today, I asked for the prices and yields on 7 and 8 year Treasury bonds. They were about 84 and about 81 respectively, correlating with yields of 2.44% and 2.64%. I purchased some 8 year (2019 maturity) bonds at around 81. What this means is that assuming los Federales are still in business in 8 years, they promise to pay me $100 for each $81 worth of bonds today, but they will pay me nothing till then. Annualized that's about 2.6% yearly. However, if I want to make a similar investment but want $100 back in 7 years rather than 8, I will have to fork over $84 today.

Let's now look forward one year. Let us say that for whatever reasons, the yield curve is unchanged. In that case, I will then be the owner not of an 8 year bond, but rather I will own a 7 year bond. All things being equal, the value of my bond will have risen from 81 to 84. Taking the fractions into account, my 2.6% bond today will rise 4.0% in value (from 81.18 to 84.47). It's a form of financial alchemy, in a sense. All I need to get a 1-year return almost equal to the return from lending money to the government for 30 full years is for the interest rate structure to simply not change.

This sort of activity is called rolling down the yield curve. It is one of the SOPs among bond portfolio managers.

In making this bet, I am taking the point of view that the Economic Cycle Research Institute is correct in its call for Treasuries rather than stocks or industrial commodities as preferred investments in the months ahead. In this scenario, we should remember that as recently as November last year, the 5-year Treasury traded with a yield below 1.1%. And that was merely in the setting of a growth slowdown, at a time when gold and silver prices had already been surging for months. What if there is a full-blown global industrial downturn, with oil in the $60-80/barrel range and copper at $3/pound or less? And perhaps silver at $25/ounce again, or lower? Who knows where the herd will take Treasury prices in the seemingly unending cascade of financial asset inflation that has been occurring for decades?

In that (quasi-)recessionary scenario, there is the potential for this boring debt instrument to return 5-10%, possibly in as little as 6 months.

Of course, there are numerous risks to this approach, most obviously the one that price inflation makes this 2.6% yield look trivial. My answer to that is ownership of gold so long as interest rates are at or below the general price inflation rate, no matter what the economic cycle is doing. I would then supplement that by adding economically sensitive, "weak dollar" assets when economic activity appears to be troughing or accelerating. But in a world where there are central authorities who are determined to make savers pay to recapitalize financially weak financial institutions by forcing them to receive rates on money in the bank that are below the rate of price inflation, one thing is mathematically certain.

Pitiful as a yield of 2.6% may be, it is greater than zero.

In a world where little except the air we breathe is cheap, beating zero with the chance for capital gain due to a possible major deceleration of economic activity seems reasonable to me.

Copyright (C) Long Lake LLC 2011