Fixed Income / Wealth Management

This Is No Time to Bail on Bonds

By Seth J. Masters February 07, 2017
This Is No Time to Bail on Bonds

Before the 2008 crash (a period we call the Old Normal Era) bond investors viewed their core bond portfolios like a Swiss Army knife—a single tool that could perform at least three key tasks: diversify equity risk, provide stability, and deliver income.

In the New Normal Era that followed, bond investors had to choose among distinct objectives. With central banks driving government bond yields to record lows, many income-hungry investors opted for income, sacrificing stability and diversification. Many invested in the highest-yielding securities they could find, such as 30-year callable municipal bonds or high-yield exchange-traded funds (ETFs). They soon came to regret it: These high-income bonds traded off sharply when interest rates rose after the recent US election.

Now we seem to have entered a Retro Normal Era, with bonds providing strong diversification benefits once again. As the left side of the Display below shows, stocks and bonds have been uncorrelated, on average, for nearly 30 years. But for the last three months, their returns have been almost perfectly negatively correlated, with bonds falling when stocks rose. If this relationship persists, bonds are likely to rise in a stock market sell-off, providing a valuable cushion to balanced accounts.

We expect bond returns to be bumpy for the next few years, as interest rates continue rising from historic lows. Historically, bonds have usually fallen in price in the initial phase of rising rates—but bond income soon offsets the fall, aided by higher reinvestment rates.

You can see how income cushions the negative price impact of rising rates in the right side of the Display above. We estimate that interest rates would probably have to rise 0.9% across the yield curve for a municipal bond portfolio with a four-year average duration to deliver a 0% return over the next year; they’d have to rise even more for bond returns for the year to be negative. Similarly, rates would have to rise 2.1% for bonds to have a 0% return over two years, and even more to deliver negative returns for that period.

History suggests that over the long term, higher reinvestment rates will also ease the pain of short-term losses. The next Display shows what happened during the last sustained period of rising rates, from 1955 to 1980. Over those 26 years, the compound annualized return on Treasuries was 4.5%. That may be far less than their 7.7% compound annualized return over their subsequent, 36-year decline, but it’s far better than their current yields suggest.

We don’t know whether the coming years for bonds will repeat the 1955 to 1980 pattern, or chart a new path. Either way, we think bond investors whose time horizons are longer than their portfolios’ durations should ultimately benefit from rising rates, even if sharp jumps in rates create short-term losses along the way.

The views expressed herein do not constitute research, investment advice or trade recommendations and do not necessarily represent the views of all AB portfolio-management teams.

This Is No Time to Bail on Bonds
Back to a top