Two questions being asked right now about bonds is when will interest rates rise and how will it affect bond performance? Of course, these questions aren’t unique to today. Investors always ask about interest rates and the potential impact to bond returns.
There’s just one tiny problem. Nobody knows how far, how fast, or even when interest rates will move. Predicting the markets is fruitless in the short term.
That doesn’t mean you can’t come up with a realistic expectation for future bond returns.
This brings up an important question.
When setting expected long-term returns, most investors rely on past performance. So how useful are bond’s past performance for setting expected returns?
Not very.
Outside of using it to piece together the data below, to check if a fund lived up to expectations, or to see how bonds performed during a specific time period or event, past performance has its limits. And sometimes, it can do more harm than good by overinflating expectations.
Current yield is a better measure.
To show this, I plotted the monthly starting yield for 10-year Treasuries (from Multpl.com) going back to 1920 against the subsequent 10-year annual returns and got the chart below.
What you get is the elongated shotgun pattern that tends to follow a straight line from the bottom left corner to the upper right corner of the chart. You do get a few stray pellets mixed in but close enough counts here.
I wasn’t expecting perfection because changes in interest rates impact bond prices which affect the overall return.
If you owned an individual bond and held it to maturity, your annual returns would be the starting yield. However, bond funds don’t have a maturity date but instead rely on price returns and yield. If you held the bond fund for a similar ten-year period (as the duration of a single bond), the funds annual total returns tend to approximate the starting yield. The chart shows that the changes in bond prices don’t play a big role in long-term bond returns.
Sure, some years the difference between yield and returns were bigger than others. The two widest spreads were the periods starting in 1971 and 1976 because of the dramatic change in interest rates due to the spike in inflation:
- Oct. 1971 – returns fell short because rates went from 5.93% to 11.75%
- Dec. 1976 – yields fell short because rates went from 6.87%, peaked at about 15%, then fell to 9.26%
Overall, 10-year Treasury returns beat the starting yield by an average of just two-tenths of a percent. (The accuracy depends on how far and how often interest rates change. Ignoring things we know we don’t know about the future, like changes in inflation over the next ten years, starting yields work as a decent guide.)
What you don’t see are low starting yields that produce very high returns or high starting yields producing very low returns. Simply put, starting bond yield is a good predictor of future bond returns.
While 10-year treasury bonds produced 8% returns since 1980 and 5% returns going further back, that information isn’t very helpful today. If the current 10-year yield is 2.3%, you could earn 8% next year. You could also lose 8% next year. But you’ll never earn 8% annually over the next 10 years with a starting yield of 2.3%. Bonds just don’t work that way.
You should expect close to 2.3% annual returns from 10-year Treasury bonds over the next decade. That probably doesn’t sound very exciting, but it helps to remember why you own bonds in the first place – to be a stabilizer when stocks act a little crazy. Owning both stocks and bonds should make the lower returns more bearable.
Bond yields may eventually move toward the historical average, but when and how fast is unknown. Until then, the best way to temper your expected returns is to focus on the current bond yield, plus or minus a percent or two.