Are We Witnessing a Melt-Up In Long-Term Bonds?

Since the financial crisis, investors are constantly on edge that we’ll see another melt-down in the stock market. On the other hand, there are investors such as Jeremy Grantham of GMO, that have been predicting the opposite — a melt-up in stocks before we see an eventual crash as investors notoriously take things too far.

But what if investors are looking for a melt-up in the wrong place? What if the real melt-up is already occuring in one the most boring asset class of all — long-term U.S. treasuries?

Long bonds have been on fire this year as they’re currently one of the best investments of any asset class. The iShares 20+ Year Treasury ETF (TLT) is up over 28%. From a historical standpoint, were that number to hold up through year end, it would be one of the best years ever for long bonds. Here are the top ten annual returns going back to 1926:

LT bonds I

The performance in 2014 cracks the top five highest of all-time.You’ll notice that every single result in the top ten has occured since 1982 when interest rates started their steady decline from double digit levels. The average yield during these best performing periods was nearly 7%, juicing the total returns.

It seems everyone expects yields to shoot higher eventually, but I wanted to see what the performance numbers looked like in the past when rates stayed low for a long period. From 1926 to 1960, long bond yields were basically stuck in the 2-4% range. The current yield is in that range so historical returns could offer an idea about future performance. These were the ten best annual returns from that previous low yielding period:

LT bonds II

This lower-for-longer period of interest rates led to a much more muted upside. The average return of the ten highest years in the 1926-1960 time frame was around 10% versus an average of 27% for the ten best in the 1926-2014 period.

Long bonds have shown fairly decent returns over the very long-term, going all the way back to 1926, of 5.7% annually (2.8% on an inflation-adjusted basis). But from 1926 to 1960 they only returned 3.2% a year (1.8% after inflation). This is not a prediction, just something to be aware of when setting return expectations.

People have been predicting the demise of bonds and a rise in interest rates for a number of years now. Anything is possible, but rates don’t have to go up just because they did in the past. There are legitimate reasons that we could see rates continue to fall in 2015 and beyond which could mean long bonds outperform once again. If we really were in the melt-up phase that’s exactly what would happen.

However, were this trend in rates to reverse, it’s worth noting that treasury bonds don’t see huge crashes like stocks do because of the way they’re structured (see: What Does the Bursting of a Bond “Bubble” Look Like). The largest annual loss in long bonds was only around 15% in nominal terms. Inflation is the real long-term killer of bonds.

So if you’re looking for a melt-up in 2015, it’s quite possible it could be in bonds and not stocks as most people assume.

Further Reading:
What’s an investor to do about bonds?
Looking beyond interest rate risk in bonds
Back-testing the Tony Robbins All-Weather Portfolio

Subscribe to receive email updates and my monthly newsletter by clicking here.

Follow me on Twitter: @awealthofcs

Download PDF

Full Disclosure: Nothing on this site should ever be considered to be advice, research or an invitation to buy or sell any securities, please see my Terms & Conditions page for a full disclaimer.

What's been said:

Discussions found on the web
  1. 10 Wednesday PM Reads | The Big Picture commented on Dec 17

    […] Are We Witnessing a Melt-Up In Long-Term Bonds? (A Wealth of Common Sense) • Shale is to OPEC as the Apple II was to the IBM mainframe.  (BusinessWeek) see also The Oil […]

  2. Jim Haygood commented on Dec 17

    Bonds experienced a regime change around 1951, when the Treasury-Fed accord expired.

    Consider: inflation advanced from -10% in 1932 to +19.9% in early 1947. Over the same period, the 10-year yield actually dropped from 3% to under 2%.

    Whereas, when inflation reached double digits again in 1981, the yield soared to over 15%. Night and day difference.

    Bonds were much less volatile pre-1951. Under the assumption that long-term price stability would prevail, bondholders just didn’t react much to either inflation or deflation. They believed such perturbations to be temporary, and that Treasuries would always mean-revert to around 3% yield (nominal and real).

    Now we’re all trend traders … and the monetary base is in the trillions. Save us, Mr. Yellen!

    • Ben commented on Dec 18

      That’s a good point. Although bonds are much more stable than stocks, it’s now easier than ever for investors to make a bet one way or another on their inflation/deflation narrative. ETFs & MFs make it easier than ever, whereas back then you had to actually buy individual bonds.

      • Jim Haygood commented on Dec 18

        While there’s really no choice with corporates, I actually do buy Treasuries for individual accounts. One can choose the exact maturity wanted, rather than holding a fund which may be shifting duration within an allowed range according to the manager’s views.

        Unfortunately owing to the OTC nature of the bond market, brokers seem unable to provide real-time intraday market values on Treasury holdings for individual accounts. It’s positively 19th century.

  3. Bob Carlson » The Great Bond Market commented on Dec 22

    […] on stocks in 2014, bonds have been having a great year. The forecasts were completely wrong, as this post details. In January you’ll see a lot of stories about how great an investment bonds were in 2014. As […]