By: Frank Armstrong, CFP, AIF
In my first column, I discussed some of the insights offered by modern financial research, and I promised that this time around I’d offer an example of how we can translate theory into practice. After all, financial theories are only useful to investors if they can be applied to real world problems. By using no-load mutual funds as building blocks, even investors of very modest means can ape the practices of our largest institutions. Over the course of the next few months we’ll use modern financial theory insights to vastly improve a typical portfolio, lowering its risk level and improving its expected returns at the same time.
Let’s start with a portfolio that puts 60% of its assets in stocks and 40% in long-term bonds, the so-called “balanced” mixture used by many corporate pension plans and retiree investment accounts. The S&P 500 is a good index for stocks, representing very large domestic companies, and for bonds we will use the 20-Year Treasury Bond index.
Now this may seem like a pretty naive and simple strategy. Anybody could do this, right? But it actually sets a very high benchmark. This approach doesn’t employ any fancy managers, doesn’t make any projections, and doesn’t benefit from any “inside knowledge” or trading skill. But it did manage to beat all but three of the 66 balanced or asset allocation funds with a 10 year life in the Morningstar universe for the ten years ended June 97. So much for the value of active managers! The balanced portfolio gives us a good starting point, but we can do even better.
First, let’s introduce a tool that will give us a quick and dirty way to evaluate asset classes. It’s called the risk/reward line. The risk/reward line is plotted in two dimensions: return is graphed on the vertical axis, and risk on the horizontal axis. We begin to create the risk/reward line by plotting the “zero risk” asset, the Treasury bill. Next, we plot the S&P 500 index. By convention, the risk/reward line is the line that connects these two points. Any point above (higher return) or to the left (lower risk) of another is considered better. Portfolio managers love to tout their returns if they fall in the “northwest quadrant,” the portion of the graph that is both above and to the left of the risk/reward line. As you can see, our starting portfolio’s risk and reward does not, unfortunately, plot out to the northwest of the risk/reward line. The problem is those long-term bonds.
Long-Term Bonds are Not Efficient
Investors usually think that bonds are “safe,” and that stocks are “risky”. But, in fact, the volatility (a pretty good measure of risk) of 20-year Treasury Bonds has been higher than the S&P 500 for the last 20 years. Meanwhile, it has only produced about half the total return of the S&P 500. In technical terms, bonds are not “efficient”, that is they don’t deliver a very high rate of return for each unit of risk.
So, when we plot it out on a risk-reward line, the results look pretty dismal. You can see at a glance that this is a pretty grim part of the chart. If we are going to take all that risk, we ought to get a higher return.
The Yield Curve
Most investors are familiar with the concept of the yield curve: in normal times, the longer the duration of a bond, the higher its yield should be. But, most investors haven’t considered that while the yield moves up nicely from one day to about two years, there isn’t much more yield gain between the two-year and thirty-year points! Even a large change in interest rates will have no impact on the value of a 30-day Treasury bill, but a very small change in interest rates will send the value of a 20-year Treasury bond gyrating. So, risk increases dramatically as a bond’s duration increases.
Little Diversification Benefit
Another dimension to consider when we look at an asset class is the likelihood that if another asset class changes in value, it will change in lock step with it. If the two asset classes move together, there is not much diversification benefit in holding both in the portfolio. Unfortunately, domestic stocks and long term domestic bonds have a nasty habit of moving together. When bonds are having a bad year, stocks are most likely suffering too.
Long-term bonds aren’t looking like a very neat asset class. They have high risk, produce low returns, and don’t provide much in the way of diversification. Let’s dump them!
Presuming our investor needs a store of value to make required income distributions, or just doesn’t want to accept the full measure of equity risk, what should he do with bonds?
Shorten Duration of Bonds
If we shorten the duration of the bonds, we don’t give up much in total return, while we unload lots of risk. But we pick up on one more benefit. When interest rates rise, bonds tank, usually taking stocks with them. But short-term bonds actually yield more! The shorter the duration of the bonds, the less they give up in capital value, and the sooner they recover to par. So short-term bonds provide a nice diversification benefit to the portfolio. (In technical terms, short-term bonds have positive correlation to inflation, and rising interest rates, but negative correlation to stocks and long-term bonds. Believe me, this is a good thing indeed!)
Impact on Portfolio
Moving the average duration of the bond portfolio from 20 years to 2 years has a wonderful effect on the risk level, while giving up almost nothing in total return. Not a bad day’s work!
Long Term vs. Short Term Bonds
By: Frank Armstrong, CFP, AIF