Skip to main content
the buy side

Are U.S. long bonds still a good long-term investment because inflation is subdued? Or are the 30-year Treasury bonds best avoided owing to rising inflation and interest rates? These questions are especially relevant now, after their recent dip.

Let's look at three investment criteria I have noted before: long-term return, maximum losses from peak to trough, and maximum time of recovery for those unfortunate enough to have bought at a temporary peak.

Over the past 20 years, as long rates declined, bond prices rose from $75 (U.S.) in 1990 to about $121 today, or about 2.5 per cent annually. To this, we have to add the coupon, say mid-single digits, for 6 per cent to 8 per cent total return, depending on the period held. It's good, but doesn't even resemble the rise of bonds in the 1980s, when long rates fell from more than 20 per cent to mid-single digits, and bond returns were more than 15 per cent annually.

And how much aggravation would an investor have to take, and for how long, to get this expected long-term return (assuming inflation doesn't rise here)?

Again, let's look at the past. There have been several periods of steep spikes in bond prices during market scares as investors fled equities, followed by sharp falls. These declines have been of the order of 20 to 25 per cent, lasting from one to two years.

The recovery time, however, was often twice that. If a bond investor was unlucky enough to buy bonds at the crest of the wave, they often had to wait several years to get back to even, although they did get paid a coupon while they waited.

All things considered, long bonds clearly provide above-average aggravation for a moderate return, if bought close to their peak.

Are they likely to keep rising in future? And even if they are, are they worth buying now after their recent fall?

My answers are probably yes, and probably no. The future will likely bring slower economic growth, which means that even a larger base of money will circulate more slowly to produce a slower growth of money supply. So bonds should still do all right, and the most successful bond experts I'm tracking agree: Van Hoisington and Gary Shilling.

But as for timing, bonds may have further to fall in this cycle: They are down only 13 per cent, which is half their usual loss, from a peak in August.

Meanwhile, China is losing patience with U.S. money printing, and U.S. credit rating agencies, having been burned by their lax ratings in the 2008 financial crisis, are more careful now. Moody's Investors Service just warned that the proposed U.S. tax package will worsen already-stretched finances and increase the likelihood that Moody's will change its outlook on the U.S. rating to negative.

If any major credit agency does downgrade its outlook, U.S. bonds may tank further, which might be the only thing to push the U.S. Congress to stop borrowing and spending.



So yes, I think long bonds should be good in the long term. But based on their record and on current prospects, their recent decline may not be over and it's too early to buy them. The bottom is likely at least 10- to 13-per-cent lower, or a year or two away, or both. Look at short-term U.S. corporate bonds, instead, where you can still find some excellent value, with proper research.



Avner Mandelman is a director of Venator Capital Management and author of The Sleuth Investor. Amandelman@venator.ca

Interact with The Globe