For the past year or so, perhaps the most persistent question I get involves choosing bond maturities. Typically, investors are certain that interest rates will rise, which leads them to believe they should keep the maturity of their fixed income holdings as short as possible. However, such an approach isn't likely to prove optimal.
In addressing the issue, the first thing that I point out is that the market surely knows about the likely direction of interest rates. Thus, those expectations should already be built into current prices (or, the yield curve), meaning you can't expect to profit from that information. If investors are risk neutral, the yield curve will be a reflection of future expected yields. If they're risk averse, they'll demand a risk premium (which you can earn).
My next step is to provide a lesson from history -- as George Santayana said: "Those who cannot remember the past are condemned to repeat it." As the table below demonstrates, history provides us with a remarkably similar situation to the current one. The table provides the yields from two Treasury yield curves, the current one (as of April 18) and the yield curve as we began 2004. The figures in parentheses are the spread between that maturity and the one-month Treasury bill rate.
Note how similar the spreads are. In both cases, the yield curves were very steep, which reflects two things:
- The expectation that rates will rise
- A risk premium that investors accept for bearing term risk (an uncertainty premium)
In both cases, the yield curve reflects investor expectations that the Federal Reserve would/will tighten monetary policy, leading to a rise in rates. They may also have been concerned about the potential for rising inflation (caused by the Fed's prior loose monetary policy). In another coincidence, the price of an ounce of gold had risen about 20 percent in 2003. Because of those concerns, investors were reluctant to extend maturities beyond the very short term.
At least in the period 2004-06, history played out about as expected. The Fed raised the federal funds rate 17 times, from 1 percent to 5.25 percent. Yet, long-term (20-year) Treasuries were the best performer, returning 5.8 percent. This compares to the relatively similar returns of 3.0 percent for one-month bills, 2.5 percent for one-year notes, 2.3 percent for five-year Treasuries and 2.6 percent for intermediate Treasuries (1-10 years).
Of course, we don't know what the future holds -- all crystal balls are cloudy. While rates seem likely to rise, they may not rise as much as they did in the 2004-06 period. For example, then yields on one-month bill rates rose 3.9 percent and one-year yields rose 4.3 percent. However, five-year yields rose just 1.3 percent, 10-year yields rose just 0.3 percent (outperforming short maturities), and the 20-year yield actually fell 0.3 percent.
Here's an example I use to illustrate that you may still be better off extending maturities when the yield curve is steep and rates can rise. Let's assume that one-year rates are 1 percent and two-year rates are 2 percent. A two-year note will (without considering the benefits of compounding) earn 4 percent. That means that the investor in the one-year instrument must earn 3 percent in the second year to break even. Thus, the interest rate on a one-year security would have to rise at least 2 percent (from one to three) in a year to break even with the return offered by the two-year instrument.
Investors staying away from bonds and staying in cash have paid a steep price in the form of foregone yields. Given that the nominal yield curve remains very steep, it seems prudent to maintain discipline with Treasuries, government agencies, CDs and municipal bonds. For investors with enough assets to make it worthwhile, it may make sense to build laddered portfolios with an average maturity of four to five years -- reflecting a ladder of about eight to 10 years. Investors who are more risk averse to the negative effects of inflation should lean to the shorter end of that eight- to 10-year range. Those for whom inflation is less of a risk (for example those still employed) should consider the longer end of the range. The laddered approach balances the dual risks of investing in nominal return bonds -- reinvestment risk and term (inflation) risk.
Also keep in mind that for tax-advantaged accounts the preferred investment will generally be real return bonds (such as TIPS).
More on MoneyWatch:
Why You Shouldn't Be Scared of Low Interest Rates Is Inflation Risk Overstated? Are Stocks Overvalued? John Hancock: Does It Add Value? What Rising Correlations Mean for Market Returns
Three ways I can help you become a wiser investor: