• Dimensional
• Philosophy
• Application
• Strategies
• Service
• Library

• Articles
• Reprints
• Videos
• Glossary
• FAQ
• Bios

Search

For Professionals

Institutional investors and Financial Advisors worldwide have access to Dimensional editorials, materials, and analytical tools:

Secure Site Login

Fixed Income Investing

Print this

E-mail this

Managers of investment portfolios are concerned about risk, return, and the amount of risk required to achieve a specific return. In the pursuit of returns, history and theory confirm that the expected return on equities exceeds the expected return on fixed income. Managers looking for superior inflation-adjusted returns begin with equities. The first line of defense in risk reduction is diversification. Managers reduce the risk of equity portfolios by diversifying. They spread the risk among domestic and foreign equities, large companies and small companies, growth stocks and value stocks, etc. Still, even a highly diversified equity portfolio may have a higher volatility than is desired by the investor. At this point, a less volatile asset class—fixed income—is introduced into the portfolio. The primary reason for adding fixed income is volatility reduction.

The Maturity Decision

Most investors recognize that short-term fixed income strategies are less volatile than long-term bond strategies. The difference in standard deviations between short-term securities like treasury bills and long-term securities like 20-Year Treasury Bonds is startling.

In fact, long bonds have volatilities approximately five times that of the One-Year Treasury Bill. In addition, short-term bonds have a lower correlation with equity portfolios than do long-term bonds. The evidence is strong, if risk reduction is the primary goal, that short-term is far more effective than long-term fixed income.

Yet long-term bonds still fill many investment portfolios. Many investors hold long-bonds in the hope of higher returns. There is often the intuition that long-term bonds must have higher expected returns than their short-term counterparts. Historically, this has not been the case.

Quarterly: January 1964-December 2001

One-Month T-Bills, Five-Year T-Notes, and 20-Year Govt. Bonds data courtesy of © Stocks, Bonds, Bills and Inflation Yearbook™, Ibbotson Associates, Chicago (annually updated works by Roger C. Ibbotson and Rex A. Sinquefield). Six-Month and One-Year Rolling T-Bills data courtesy of Merrill Lynch.

The data shows there has not been a reliable return premium for extending maturities into longer bonds. One reason this result may be surprising to many is that recent history does not conform to this long-term result. The data from the period 1982 through 1993 shows a tremendous premium for maturity extensions. In this period long-term bond returns were more than double the returns of short-term, and almost kept pace with the return of the stock market.

Yet a longer perspective would indicate this outcome was more the exception than the norm. In data going back to 1926, there is no other period for bonds like the bull market of the '80s (EnCorr). Then again, there has never been a period where long-term interest rates declined from 16% to 6% as they did over this time span. In the five-year period preceding 1982, rates climbed from 7% to 16% and bond yields trailed the returns on One-Month Treasury Bills by more than 10% per annum. For there to be a substantial decline in rates, there first has to be a substantial rise in rates. When the entire interest rate cycle is considered, return premiums disappear.

Interest Rate Forecasting

With no reliable "term premium," investors could only hope for astute active management in order to extract value out of long-term bonds. Managers possessing the ability to forecast interest rates stand to profit more from their approach by using long bonds when their forecast calls for declining interest rates. The question that naturally follows is whether or not managers can reliably forecast interest rates. Inextricably linked to this active vs. passive argument is the question of market efficiency. Many studies have been done regarding the question of bond market efficiency. Much of this research is similar in nature to efficient market studies performed on the stock market. The conclusions are similar—there is no reliable method of forecasting future bond prices, and therefore future interest rates. The bond market is efficient. If this is true, the returns of active bond portfolios should mirror the results found in active stock portfolios—an inability to consistently outperform their benchmarks. Indeed, this was the outcome found in research conducted by Blake, Elton & Gruber.¹ The researchers found that when properly categorized, bond funds underperformed relevant indexes. The amount of this underperformance was approximately equal to the fund expenses. Gross of fees, bond funds on average perform on par with their appropriate index.

Dimensional's Fixed Income Approach

Investors seeking short-term, non-forecasting strategies are not limited to "buy and hold" or indexing strategies. While these approaches may be conceptually appealing in that they do not attempt the difficult task of forecasting interest rates, they fail to capture changing opportunities offered by a changing yield curve environment. Indexing strategies are essentially static maturity approaches. The average maturity, or duration, generally fluctuates only in a small band. Bonds are held in an index portfolio only because they conform (or help the portfolio to conform) to the characteristics of the index. For example, if a Five-Year Treasury Bond is held in an indexed portfolio, it is typically held through all yield curve shapes. Even if the less risky Two-Year Treasury Bond offers a higher yield, as is the case in inverted yield curves, the portfolio will continue to hold the five-year bond. While the volatility of the portfolio can be reduced (and yields increased) by trading the five-year bond for the two-year bond, the indexing strategy generally will not execute the trade if it increases the risk of tracking error. Risk, to an indexer, is measured by tracking error. Trades that reduce volatility and increase yield may have no inherent appeal in an index approach even though they may be very attractive to the diversified investment portfolio. Buy and hold strategies, such as a laddered approach, are equally indifferent to a changing opportunity set.

Investors may be able to increase their risk-adjusted returns with an alternative approach developed by Professor Eugene Fama of the University of Chicago. This variable maturity strategy shifts the maturities of the portfolio as yield curve changes create the possibility for lower risk, higher expected return outcomes. In recognizing the bond market as being highly efficient, the variable maturity approach does not anticipate changes in the yield curve, rather it seeks to maximize the risk-adjusted returns present in the current curve. The variable approach would shorten from five-year maturities to two-year maturities in the example above. The cornerstone of this approach is efficient market theory. Bond prices (and therefore their yields) reflect all available information and follow the classic "random walk" pattern. In this type of market, the best estimate of future bond prices/yields is simply today's price/yield. New information is unpredictable. Summarized, Fama's research showed that the best estimate of future yield curves is simply today's yield curve. It is not a statement that the curve will not change, but a statement that the changes are unpredictable. The objective of the strategy is then to take what is offered by the current curve. In broad terms, this means shortening maturities in inverted curves, and extending in upwardly sloped curves.

Determining the Optimal Maturities

The Shape of the Yield Curve

Source: Hypothetical yield curves of a fictional fixed income product, presented solely as an example.

While the above figure graphically depicts broad maturity differences due to yield curve differences, it describes only one aspect of the variable maturity strategy. Even without forecasting changes in today's yield curve, there exists a myriad of potential strategies, each with a unique expected return. If the two-year bond is purchased, how long should it be held? Can expected returns be improved if the two-year bond is held for only a portion of its life, then sold, and the proceeds reinvested back in a new two-year bond? The figure below shows a sample matrix of expected returns generated for varying initial maturity choices and holding periods. For each expected return a horizon analysis is conducted. In each case, the analysis depends on three factors: the purchase price of the bond, the yield or income generated over the holding period, and the sale price of the bond. To calculate, for example, the expected return for the strategy of buying a two-year bond and holding for six months, we start with the purchase price of two-year bonds. This is observed in today's marketplace. This price will determine a yield which will be earned over the six-month holding period. The unknown factor is the sale price of the bond six months hence. To generate an expected return, an expected price/yield is needed for what will be an 18-month bond, six months from now. Using the Fama research, we use today's curve as our estimate of the curve six months from now. We use today's price/yield of 18-month bonds to estimate the price/yield of 18-month bonds six months hence.

Finding the Optimal Maturity and Holding Period

Source: Hypothetical yield table of a fictional fixed income product, presented solely as an example.

In general, yield is a poor estimate of expected return. In upwardly sloped curves, expected returns are generally higher for strategies that do not hold bonds to maturity. The variable maturity approach will highlight the optimal combination of initial maturity and holding period—in essence the best segment of the curve to hold. This combination will change as the shape of the curve changes. In each case though, the strategy seeks to make optimal use of information imbedded in today's efficiently priced yield curve.

The Use of Global Bonds

The arguments of active vs. passive, long vs. short, and static vs. variable maturities hold in developed foreign bond markets as well. With non-dollar bonds, however, there is the additional concern of foreign currency exposure. Efficient market research conducted on exchange rates found the same random walk phenomenon. Exchange rates move unpredictably. Currency exposure clearly increases the volatility of the fixed income portfolio, while there is no reliable evidence to suggest that the expected return of exchange rates is anything other than zero. Our belief is that currency exposure should be hedged in global bond portfolios. Generally, investors pursue global portfolios in order to diversify. Statistically, diversification should result in a lower portfolio volatility due to the combination of uncorrelated assets. If volatility is increased with the addition of global assets, the whole purpose of international diversification is defeated.

With currency exposures hedged away, the goal of diversification is attained. Introducing hedged foreign bonds into a domestic portfolio reduces the volatility of the portfolio. Portfolios of hedged global bonds take advantage of imperfect correlations among developed bond markets and enjoy the classic benefits of diversification. Further, given the global nature of highly rated debt issuers, this international diversification can be reached without sacrificing the credit standards maintained in domestic portfolios.

Correlations of Hedged Country Bond Indexes

Monthly: February 1991-December 2001.

Data courtesy of Salomon Smith Barney.

In terms of returns, just as investors are no longer subject to the risk of one bond market, they are no longer subject to the expected returns of just one yield curve. Expected returns across hedged bonds differ as the shape of each yield curve is different. Portfolios can be formed which are tilted toward the higher expected return countries. In this case, portfolio maturities and country weightings follow a variable approach based on the expected return matrix generated for each eligible country.

In our view, global bonds do not represent a separate and distinct asset class from domestic fixed income. Instead, the use of currency hedged non-dollar bonds along with domestic bonds allows investors to create a more diversified, less risky fixed income portfolio. Global bond portfolios are often better suited than their domestic-only counterparts to achieve the primary goal of fixed income in the investment portfolio—reducing its volatility.

Conclusion

Dimensional's fixed income strategies are designed (but not guaranteed) to provide superior risk-adjusted returns. The objective of each strategy is to maximize returns for a given maturity range through the use of a disciplined non-forecasting approach. The global portfolios are simply extensions of the domestic strategies with a larger universe of eligible instruments. Fully hedged global bond strategies increase expected returns and decrease volatility through improved portfolio formation.

  ¹ Christopher R. Blake, Edwin J. Elton, and Martin J. Gruber, "The Performance of Bond Mutual Funds." Journal of Business 56.3 (1993): 371-403.

Fama, Eugene F. "Forward Rates as Predictors of Future Spot Rates." Journal of Financial Economics 3 (1976): 361-377.

Fama, Eugene F. "Inflation Uncertainty and Expected Returns on Treasury Bills." Journal of Political Economy 84.3 (1976): 427-448.

Fama, Eugene F. "Inflation, Output, and Money." Journal of Business 55.2 (1982): 201-231.

Fama, Eugene F. "The Information in the Term Structure." Journal of Financial Economics 13 (1984): 509-528.

Fama, Eugene F. "Short Term Interest Rates as Predictors of Inflation." The American Economic Review 65.3 (1975): 269-282.

Fama, Eugene F. "Term Premiums and Default Premiums in Money Markets." Graduate School of Business at the University of Chicago and National Science Foundation, July 1985.

Fama, Eugene F. "Term Premiums in Bond Returns." Graduate School of Business at the University of Chicago and National Science Foundation, November 1983.

Fama, Eugene F. "Term Structure Forecasts of Interest Rates, Inflation, and Real Returns." Graduate School of Business at the University of Chicago and Graduate School of Management at UCLA, January 1988.

Fama, Eugene F. "Transaction Cost and Treasury Bill Returns." Graduate School of Business at the University of Chicago, February 1984.

Fama, Eugene F. and Robert R. Bliss. "The Information in Long Maturity Forward Rates." Graduate School of Business at the University of Chicago, UCLA, Center for Research in Security Prices, and National Science Foundation, April 1985.

Fama, Eugene F. and Kenneth R. French. "Business Conditions and Expected Returns on Stocks and Bonds." Graduate School of Business at the University of Chicago and National Bureau Economic Research, January 1989.

Fama, Eugene F. and Kenneth R. French. "Common Factors in the Returns on Bonds and Stocks." Graduate School of Business at the University of Chicago, Revised July 1992.

Fama, Eugene F. and Kenneth R. French. "Common Risk Factors in the Returns on Bonds and Stocks." Graduate School of Business at the University of Chicago, First Draft April 1992; Revised July 1992.

Fama, Eugene F. and Kenneth R. French. "Forecasting Returns on Corporate Bonds and Common Stocks." Graduate School of Business at the University of Chicago, National Science Foundation, and Center for Research in Security Prices, December 1987.

Fama, Eugene F. and Michael R. Gibbons. "A Comparison of Inflation Forecasts." Journal of Monetary Economics 13. 1984: 327-348.

Fama, Eugene F. and Michael R. Gibbons. "Inflation, Real Returns and Capital Investments." Graduate School of Business at the University of Chicago and Graduate School of Business at Stanford University, December 1980.

Fama, Eugene F. and G. William Schwert. "Asset Returns and Inflation." Journal of Financial Economics 5 (1977): 115-146.

Fama, Eugene F. and G. William Schwert. "Inflation, Interest and Relative Prices." Journal of Business 52.2 (1979): 183-209.

Dimensional Fund Advisors Ltd. is authorised and regulated in the United Kingdom by the Financial Services Authority (FRN: 150100). For professional investors and advisers only. You should obtain relevant and specific professional advice before making any investment decision. Past performance is not an indication of future performance. Please read the Important Information.

Back to Top