How Can I Trade the 30-Year Treasury Bond?

With considerable hassle. But we'll describe easier ways to gamble on a big drop in rates.
Author:
Publish date:

What are the ins and outs of trading the 30-year Treasury bond? Lots of pros say its yield will be 5.25% or so in 12 to 18 months. What would the total return be if one bought $50,000 via an online brokerage this week at a yield of about 6.30%, and sold 12 months later at 5.30%? How much would the price change as the yield declined? -- John Barbati

John,

Ask a simple question, get a simple answer? Not so easy, I'm afraid. For individual investors, this topic is a quagmire, to say the least. We'll try to unravel it for you and give you some good options.

You're right that lots of people -- including

Pimco's

Bill Gross, the best-known and one of the most successful bond fund managers -- think that Treasury yields are topping out. Individual investors can certainly profit from such a move, but there may be better ways to do so than buying the current 30-year Treasury bond.

Your question raises at least several issues, and we'll address each. Throughout, please remember that anything that performs extremely well when interest rates are falling will perform extremely badly if rates rise. There are plenty of smart people forecasting higher rates, too. (

Morgan Stanley Dean Witter's

Stephen Roach, who forecast this year's bear market, comes to mind.)

Buying bonds in quantities of under $1 million face value is costly.

Jim Cramer

can

trade the 30-year Treasury because he can buy it in lots of at least $1 million. And since he's got expensive data terminals, bond brokers can't pull any fast ones with him on price. That presumably isn't true for most of the rest of us.

The bond quotes we can see -- on

TheStreet.com's

ticker, elsewhere on the Web, on TV or in the newspapers -- are a) delayed, b) a bid price (you're going to pay a higher offered price) and c) for what's known as a round lot of $1 million face value or more. Smaller quantities cannot be transacted at that price, and there are no clear rules limiting how much more than that a broker can charge you for a smaller quantity.

Having said that, an honest broker shouldn't charge you more than an extra point for an odd lot. When you sell, you'll also receive a point less than you would for a round lot. So your price return will lag the quoted price return by about 2%.

Ideally, you'll do the transaction through a broker whom you trust to give you a fair price. (Treasury securities can be bought

directly from the government, but only at auction, and the next 30-year bond auction isn't till February.) As for doing the transaction through an online broker, earlier this year we

found that individuals' ability to trade bonds with a mouse remains limited.

Total return is sweet, if you can get it.

If rates drop by 1 percentage point, or 100 basis points, the total return on the long bond would be in the neighborhood of 20%. But because of liquidity reasons, you won't enjoy that entire gain.

Problem is, the current benchmark 30-year Treasury bond will not be the benchmark 30-year Treasury bond a year from now. The 30-year Treasury bond whose yield everyone quotes is always the one most recently issued. Currently, that's the one issued in August. The Treasury Department issues a new 30-year bond twice a year, in February and August.

A year from now, traders will be calling the bond that's currently the benchmark the "double-old bond" or the "two-old bond." The benchmark will be the one issued next August. What does that mean to you? If rates go down and prices go up as you expect, the double-old bond almost certainly won't trade at as high a price as the new benchmark, because it won't be as liquid. That's a few more basis points of total return you won't collect when you sell.

Achieving whatever the total return on the bond turns out to be over the next year requires reinvestment of the coupon income.

To realize the total return that the 30-year Treasury will produce over the next year, you not only have to stick with the active issue (which does not make sense, since the cost of trading into it twice will wipe out a lot of your gain). You also have to reinvest the coupon income.

The active bond has a 6.125% coupon, so it will pay $30.625 per bond twice over the course of the year. If you buy 50 bonds (total face value $50,000), your semiannual coupon payment will total $1,531.25, too small a sum to reinvest in the bond.

Don't get me wrong -- you'll do great in the current bond if rates fall sharply, even if you don't reinvest the coupon income. I'm just pointing out that with bonds, realizing the stated total return is complicated. You can ballpark what the total return would be on a bond over a period, assuming a certain change in interest rates, by adding the price change over the period to the coupon rate and dividing the sum by the original purchase price. (See chart.)

You'll get better results from Treasury Strips than from the long bond if rates fall.

Treasury Strips are zero-coupon bonds. Instead of paying coupon income, the bonds are priced at a deep discount to face value. They outperform coupon bonds when interest rates are falling for a very simple reason: There's no coupon income to be reinvested at lower yields. You can get them only through a brokerage (not through Treasury Direct), and transaction costs will take at least as large a toll. Also, price indications on Strips are harder to come by than for benchmarks (though we have

reviewed a system that provides them). But the gains on a 100-basis-point drop in yields will cover the transaction costs many times over.

There are ways to express your conviction through mutual funds, which may be less expensive than buying individual bonds.

Does all this talk of transaction costs and price opacity make you nervous? You can very effectively take advantage of a big move down in interest rates via mutual funds, and pay a stated expense ratio rather than an unstated broker's markup.

American Century

sells an array of no-load "Target Maturities Trust" funds --

(BTMTX)

2000,

(BTFIX) - Get Report

2005,

(BTTNX)

2010,

(BTFTX)

2015,

(BTTTX) - Get Report

2020 and

(BTTRX) - Get Report

2025 -- designed to match the return of a Treasury zero-coupon bond maturing in those years. The longer the maturity, the better the fund will perform if rates fall. (A

column earlier this year focused on these funds' substantial quirks.) Any fund classified by

Lipper

as a general U.S. Treasury fund should also do well if rates fall, but the longest Target Maturities Trust Funds will do better.

Finally, there's the choice of futures and/or options on futures.

But that's a lesson for another time. How about next week?