By: Frank Armstrong, CFP, CLU
In the universe of the average investor, bonds provide two important functions. They reduce the risk of the portfolio, and provide a store of value to fund future expenditures.
We assume that the average investor is interested in maximizing his portfolio’s rate of return per unit of risk. Some institutional investors such as pension plans or insurance companies are more interested in matching assets with liabilities. For those investors a totally different approach than what we discuss below is appropriate.
Of course, we would like bonds to add an increment of total return. However, while most investors tend to focus on that, that’s really only an incidental and pleasant side benefit of the bond portfolio.
Let’s get real. Bond returns are not very exciting. They barely outperform inflation over the market cycle. The real return is so low that it’s hardly possible to meet any reasonable financial objective over the long haul with bonds.
Many investors look back with longing to the “good old days” of the 70’s when interest rates were above 15%. This overlooks the inconvenient truth that inflation was higher than nominal interest rates during much of that period. Real returns were zero or negative for long periods. Interest rates that high were a sign of a very sick economy.
There are not a lot of creative things a bond manager can do to increase total return without dramatically increasing risk. The incremental return for accepting lower credit or longer durations is simply not worth the enormous additional risk that “stretching for yield” generates.
Because going longer or accepting lower quality bonds adds a boat load more risk for a trivial additional return it’s hardly a rational approach. That risk is not compensated, and no investor should willingly accept any uncompensated risk.
Longer term bonds usually have marginally higher returns than shorter bonds. If we graphed this return against duration we would see the so called “positive yield curve”. However longer term bonds also carry a lot more risk. During most time periods the “sweet spot” on the yield curve hovers around two years. This is the point where we have the most favorable return to risk ratio.
Few investors appreciate the magnitude of these interest rate risks in long bond portfolios. In fact, long bond portfolios have as much volatility as the S&P 500 but have only a fraction of the total return. That’s not what I call a good trade off. In a rising interest rate environment, holders of long bonds can get severely punished. The longer the bond duration, the worse the havoc on capital values as interest rates increase. Take a look at long term bond returns from 1967 to 1969 and 1977 to 1980 if you want to see some depressing examples:
Year Annual Return
Bonds have the potential to return more than cash or “guaranteed” type accounts. But, no matter how you slice it, once you leave the safety of 90 day treasury bills or CD’s you are accepting some type of investment risk.
Some investors believe that by holding a bond to maturity they have avoided any investment risk. This is the financial equivalent of holding your head in the sand. Refusing to acknowledge capital fluctuation doesn’t make it go away. If, for instance, at any point along the way to maturity the value of your bond is down 25%, you have a capital loss. If you had your original investment in your hand would you still be willing to pay full price for it? Of course not! You could invest for a higher yield today if you had your purchase price back. So, you have suffered a lost opportunity. Holding to maturity does not make this lost opportunity cost go away.
Longer term CD’s are very similar. If interest rates go up during the holding period, you’ll kick yourself for the lost interest you could have gotten if you had the cash in your hand. One month CD’s will not be substantially different from T-Bills.
Markets being what they are, the “guaranteed” products like CD’s and 90 Day Treasury Bills have lower expected returns than bonds of longer duration. If they didn’t, then nobody would ever buy a bond and accept the additional risk.
Our bond strategy is to accept very limited risk because the bond holdings are the portfolio’s “safe harbor”, and additional risk is not likely to be compensated. Given that every investor has a risk tolerance level, we believe that this risk should be “spent” in an area where it will be compensated. Having a very conservative bond portfolio allows us to tilt the portfolio towards small, value, foreign, real estate, and commodities where we expect higher returns while holding total portfolio risk within the investor’s tolerance level.
In plain words risk means that the results will be variable. Even in an optimum portfolio there will be some years above the expected return and some below. Just like stocks, bond portfolios have an expected return and standard deviation. Human nature being what it is, no one ever worries about the years where returns are above the expected return assumption. The trouble comes when returns fall below the average. Then “something is wrong”, or “it’s just not working”, or “I could have done better in CD’s”. And we all know that investor hindsight is always 20-20!
DFA utilizes a variable maturity strategy that can capitalize on a positive yield curve with relatively short term bonds (under approximately five years). When positive yield curves exist, the fund manager can capture both interest and incremental capital appreciation by purchasing bonds of longer duration and selling them as they approach maturity. However, positive yield curves are not always available. If yield curves are flat or inverted, the only rational strategy is to retreat to very short durations and wait for a positive yield curve to reappear.
Over the market cycle, this strategy has been able to add non trivial additional returns to the portfolio. And the return per unit of risk (Sharp Ratio) has been quite satisfactory. While we have every reason to believe that this strategy is prudent and close to optimum, it is not a zero risk strategy. Rather it is a low risk strategy designed to add value over the market cycle.
At the fund level, for the strategy to work reliably there must be lots of highly liquid bonds with very low transaction costs. So, for instance, it’s not possible to generate any additional benefits to a junk bond portfolio using this strategy because transaction costs would eat up any additional benefit.
At the investor level, discipline is required to reap the expected long term results. We have seen the Federal Reserve push short term rates from 1% to 5% in less than a year. It’s not realistic to expect big fat juicy rewards from a bond portfolio in that kind of rising interest rate environment. Not losing money in that period is a major moral victory for a bond fund manager. Additionally, last year the two year duration was just about the worst place to be on the yield curve. The reality is that the two year point won’t be the best place to be every day for the rest of your life. But, over the rest of your life, you will probably be glad you were there.
If you believe that, unlike the majority of professional economists, you can reliably predict future interest rates, then you might speculate in the bond market. But, if you are that smart (or delusional), perhaps you should skip the bond market and trade interest rate futures. We mere mortals are better served on the short end of the duration scale in the bond market for our non risky portfolio.
If you can’t accept any risk at all in your bond portfolio, and you are willing to accept almost no real rate of return, then perhaps you should investigate Treasury Bills or CD alternatives. If however, you would like to take a small measured risk in return for a non trivial increase in real returns, then you must accept that no bond portfolio is going to outperform zero risk alternatives every day.
Risk and Return in Bonds
By: Frank Armstrong, CFP, CLU