Treasury Shouldn't Duck Longer Bonds

By Allan Sloan
Tuesday, March 14, 2006

One of the rules of Wall Street is, "When the ducks quack, feed them." In other words, give the customers what they want. If technology stocks are hot, you sell people technology mutual funds. If investors fall in love with biotechnology, you sell as many initial public offerings of biotech firms as you can manage. This generally doesn't work out terribly well for the buyers -- but it's great for the sellers.

Now, let's pretend that you and I are running the U.S. Treasury, and we're looking at the prospect of enormous budget deficits as far as the eye can see. We've got to borrow money to fill the gap between what Uncle Sam takes in and what he spends.

We also see that pension funds and other buyers practically trampled each other last month in their eagerness to buy the first 30-year bonds the Treasury has issued since 2001. They bought $14 billion of these at an interest rate of only 4.5 percent. That's awfully low, considering that U.S. pension funds and similar long-term investors generally have return targets well above 4.5 percent. How they expect to meet their targets by buying 4.5 percent bonds is a mystery to me.

So if you and I were running the Treasury, what would we be doing? I'd sure be listening to all those bond buyers quacking. Not only would I be getting ready to sell more 30-year bonds while the market's hot, I'd be getting ready to sell 50-year bonds, too.

Yes, I've written about Treasury bonds a lot lately, and I may have told you more about them than you care to know. But what can I say? I'm a T-bond junkie (there, my secret's out). And as a taxpayer and the father of taxpayers, I think the United States would save lots of money by selling 50-year bonds (as Britain did in January) while the bond market's acting strange.

Here's the strangeness. On Monday, the yield on 30-year Treasury bonds was actually lower than on a 10-year note: That's 4.76 and 4.77, respectively. Thirties yielding less than tens? That's not supposed to happen. Longer-term securities generally yield significantly more than shorter ones because the longer a bond is outstanding, the more damage inflation does to the value of the original investment. In fact, the reason offered by the Treasury in October of 2001 for killing the 30-year bond was that the interest on 10-year notes was significantly lower.

The fact that the yield on the 30-year Treasury is not way above the yield on the 10-year indicates there are a lot of hungry long-term buyers out there. So why not toss them 50-year duck feed? I ran this idea by a Treasury spokesman. His response: "There are no plans for a 50-year bond." He declined to comment further.

Had the spokesman offered me the industrial-strength explanation, he'd probably have said that the Treasury wants to be predictable and steady and isn't a market timer. He'd have asserted that this approach has made Treasury securities more attractive to the world's buyers, thus lowering Uncle Sam's interest costs. (Am I a mind reader? No, I heard the full pitch on several occasions when I asked the Treasury why it hadn't brought back the 30-year bond.)

Plenty of bond mavens disagree with the Treasury's approach. "This is a very good time for the government to be issuing 50-year bonds," said Joseph Rosenberg, one of the nation's premier bond investors.

Rosenberg, chief investment strategist for Loews Corp., is a former member of the Treasury Borrowing Advisory Committee. He said there's no evidence that the Treasury's bond-selling approach has saved any money. "Why should [members of the committee] market-time for their own businesses but not for the government?" asked Rosenberg, who said he was speaking as a taxpayer, not on behalf of his employer.

In fact, the Treasury market-times constantly. When it eliminated the 30-year bond, the argument was that it would be cheaper to sell three sets of 10-year notes. That was a market-timing bet. When the Treasury sold 30-year bonds on Feb. 9, it was betting that would prove cheaper than issuing three sets of 10s. That's a bet, too.

Of course, no one can possibly know if selling 50-year bonds in today's market -- or in next winter's market, given that it took the Treasury nine months to go from discussing the resumption of the 30-year bond to actually selling it -- would ultimately be a good deal for taxpayers. There are no sure things in life. Maybe long-term rates will someday fall to the twos from their current level, and a 50-year, 4.5 percent Treasury bond will be as above market as the 14 percent, 30-year bond issued in 1981 is now. But I doubt it.

It feels funny for me to talk about debt that would be around a lot longer than I will. But I hate to see us taxpayers pass up this chance. Quack, quack, quack.

Sloan is Newsweek's Wall Street editor. His e-mail

View all comments that have been posted about this article.

© 2006 The Washington Post Company