- Fixed-income instruments are largely used within a portfolio to reduce volatility and provide a more consistent distribution stream for clients. Holding non-callable instruments backed by the U.S. government offer significant protection in times of financial crisis while reducing the long-term opportunity cost of bonds.
- U.S. government instruments with maturities from one to five years present the most favorable risk/reward profile. Additionally, the term premiums for extending maturities begin to decline for longer-term bonds.
- Mutual fund managers among the high-quality short-term (HS) and high-quality intermediate-term (HI) bond funds underperformed their corresponding government indexes and index funds over an entire decade. The average top quartile fixed-income managers in the HS and HI classes also underperformed their corresponding index funds.
- We analyzed the credit composition of the ten top-performing actively man-aged portfolios across the HS and HI mutual funds. Much of the value-added returns from these actively managed portfolios seem to stem from additional credit and call-option risk.
- There seems to be a direct inverse relationship between investment performance and fund expenses. The higher the investment return, the lower the fund expense ratio. Differences in expense ratios explain much of the differences in net returns.
- Further analysis of the top performing funds reveals that non-active strategies such as an indexing or variable maturity approach may be an investor’s best option.
With interest rates near their lowest levels since the early 1960s, the role of fixed-income in investment portfolios has been a greater concern for many advisors. In this environment, they may find it increasingly difficult to reduce portfolio volatility in equity-tilted portfolios while continuing to provide consistent income for client distributions without taking on levels of risk that run counter to a client’s risk tolerance. Additionally, advisors may feel that the portfolio stability provided by fixed-income instruments comes at the cost of diverting assets from other higher-return-ing asset classes. An advisor’s ability to add value using fixed-income instruments depends on identifying the specific role of fixed-income within a diversified portfolio and subsequent implementation of an efficient strategy to achieve it.
The first part of this article identifies the type of fixed-income instruments that best reduces portfolio volatility. We conclude that U.S. government-backed fixed-income instruments are able to effectively reduce risk within a client’s portfolio. We also examine the degree to which extending maturities among U.S. government instruments is worth the added portfolio risk. This investigation supports previous findings indicating that U.S. government instruments with maturities between one and five years present the most favorable risk/reward profile and that the term premiums for extending maturities begin to decline for longer-term bonds. The last part of the article examines which fixed-income investment management strategy best achieves these goals. Returns data comparing the average and top quartile mutual fund bond managers among the high quality short-term and high quality intermediate-term asset classes indicate that a non-active fixed-income management strategy is more effective than active approaches.
Over the long term and during market upswings, fixed-income instruments exhibit positive correlations with the equity market. Thus, during most of an investment lifetime, the diversification effects of fixed-income come at the expense of lower expected returns. This is an increased opportunity cost for owning bonds. To compensate for lower returns, some advisors increase their fixed-income allocation with instruments that exhibit riskier equity characteristics such as corporate or high-yield bonds. It is during market down-turns, however, when fixed-income instruments serve to significantly reduce portfolio volatility and many times outper-form riskier assets.
During severe U.S. stock market crashes, most equity markets tend to move down in concert. This contagion effect causes a reduction in the diversification benefits of many equity asset classes, but Gulko (2002) finds that during such market conditions there is a decoupling effect between the general equity markets and U.S. government fixed-income securities. A stock market crash is defined as a day the S&P 500 index loses more than five percent of its value. Table 1 indicates that during one-day stock market crashes from 1945–2000, Treasury bonds decouple from equity prices and have provided portfolio stability.
On average, the correlations between Treasury bonds and the S&P 500 for the ten days leading up to a crash have been positive (r = .26). During the crash, the relationship becomes negative (r = –.45) as a flight to quality ensues. Shortly after a crash, the relationship reverts to a more historical one. During these one-day crashes, the S&P 500 has lost on average 5.7 percent, while T-bonds have returned 3.70 percent. Gulko also finds that corpo-rate bonds and mortgage securities often decouple from Treasuries and decline during severe market downturns. More recently, during the entire month of September 2001, the S&P 500 and long-term corporate bonds declined –8.08 percent and 1.52 percent, respectively. In contrast, one-year Treasuries and five-year T-notes returned 1.09 percent and 2.53 percent, respectively. Advisors trying to attain extra yield within their fixed-income allocation may inadvertently own the wrong type of fixed-income instruments to protect against a hostile equity environment.
Once an advisor identifies the type of fixed-income instrument that best serves to mitigate portfolio volatility, he or she must determine the appropriate maturity. Is there a benefit to holding shorter-term versus longer-term securities? Table 2 presents correlations between the S&P 500, long-term corporate bonds, and U.S. government securities across various maturities from 1964 to 2003. Over this 40-year period, the correlations with the general equity market of the one-month T-bills (r = –.034), six-month T-bills (r = .021), and one-year T-notes (r = .036) were negligible. Five-year T-notes (r = .18) and long-term government bonds (r = .24) exhibited a more positive correlation with the S&P 500. Long-term corporate bonds (r = .32) had the strongest positive relationship with equities. A similar pattern appears in split-sample periods. The correlations among the U.S. fixed-income instruments increase as the maturities are extended. Hence, shorter-term fixed-income maturities show greater independence from equity markets.
Many investors may feel longer-term bonds provide higher returns. During the bull market in bonds from 1983 to 1993, long-term bonds significantly out performed shorter-term maturities, and they almost matched the returns of the S&P 500 (15.26 percent versus the S&P 500’s 16.09 percent). From 1982 to 1993, long-term interest rates declined from 16 percent to 6 percent. During the previous five years, from 1977 to 1981, interest rates rose from 7 percent to 16 percent and long-term government bonds significantly underperformed shorter maturities.
When this entire interest rate cycle(1977–1993) is considered, however, there was no significant term premium for extending maturities past one year.
Table 3 indicates that the six-month T-bill and the one-year Treasury had a significant term premium when compared with the one-month T-bill (at the 5 percent level of significance, this is the level used throughout the study). Although the five-year and long-term government bonds had higher returns, the significant increase in volatility prevented their premiums from being statistically significant. The long-term bond was more than 3.5 times more volatile than the six-month T-bill, which helps to show why the shorter-term securities exhibited more favorable Sharpe ratios during the entire interest-rate cycle. The one-year and five-year Treasuries had Sharpe ratios of .31 and. 32, compared with .19 for the long-term government instruments.
Table 4 shows that from 1964 to 2003, longer-term bonds have not significantly outperformed shorter-term maturities. Additionally, on a risk-adjusted basis, there does not seem to be a return premium for extending maturities past one to five years. For the entire 40-year period, five-year Treasury notes had an annualized return of 7.70 percent and a standard deviation of 6.60 percent. Long-term government bonds returned 7.56 percent with a standard deviation of 11.65 percent. The long-term government bond exhibited over 4 times the volatility of the one-month T-bill and over 3.5 times the volatility of the one-year T-bill. Only the six-month and one-year instruments had a significant term premium. In addition, the shorter-term securities had the most favorable Sharpe ratios.
Analysis of the four different consecutive ten-year periods in Table 4 also finds similar results. During these periods, we show the net change in a five-year par yield in order to indicate the directional change of the prevailing interest rates. The results indicate that extending maturities past five years does not improve the returns on a risk-adjusted basis. Only during one ten-year period (1984–1993) were the five-year and long-term bond premiums statistically significant. And this was during a period that followed an extended period of very poor returns for longer-term, fixed-income instruments. Both the six-month and one-year securities had significant term premiums in three of the four sample periods.
These results indicate that shorter-term securities consistently provide the highest Sharpe ratios and capture a significant amount of the term premium. There is no significant premium for extending maturities past five years. Among the selected securities, results indicate that the one-year Treasury offers the most favorable risk-adjusted term premium. This finding is consistent with Domian, Maness, and Reichenstein (1998) and Ilmanen (1996). They find that average premiums and Sharpe ratios rise through about one year, peak at three years, and begin to flatten past five years. In addition, they find that the return premium falls as maturities extend beyond 15 years. Thus, the greatest rewards for extending securities occur at the short end of the bond market.
Examining rolling returns from 1964 to 2003 (Table 5), we find similar results. The table describes the percentage of five-year monthly rolling periods (n = 421) in which one fixed-income instrument outperformed another. One-month T-bills outperformed long-term bonds 38 percent of the time. In addition, six-month T-bills and one-year Treasuries outperformed long-term bonds 45 percent of the time. Thus, longer-term securities are significantly more volatile and carry a significant risk of underperformance versus shorter-term securities. It is, however, worth extending the maturities of these securities past six months to a year. Results indicate an opportunity cost of remaining in a one-month T-bill for extended periods. The six-month T-bill and the one-year Treasury outperformed the one-month T-bill 100 percent and 94 percent of the time, respectively. Advisors employing fixed-income securities to mitigate portfolio volatility and provide a more consistent income stream for their clients would be better served investing in short-to intermediate-term government fixed-income instruments.
So why invest in longer-term bonds and bond funds? The intuitive answer is that investors enjoy the added yield they receive in longer-term securities. Obviously investors demand a higher rate of return for taking on the added risks (such as inter-est rate and default risk) with the longer maturities. These investors will require a premium for tying up their funds for a longer time, while the borrowers will pay a premium in order to obtain the longer-term financing. This is the heart of the liquidity preference theory.
But as we’ve shown above, the added premium does not appear to fully compensate an investor for the risk he or she takes. Although a detailed discussion is beyond the scope of this article, these findings pro-vide support for the preferred habitat model that states that investors prefer to buy bonds within certain maturity ranges. For example, certain institutions and individuals prefer short-term bonds and view risk as volatility of returns.
Larger institutions such as defined-benefit pensions and life insurance companies prefer longer-term bonds and consider them less risky because they better match their liability structure. They prefer to immunize their portfolio by matching the duration of their assets and liabilities. This structuring of a bond portfolio may have its benefits, but it requires a constant rebalancing of one’s portfolio as interest rates change. For individual investors, the trans-action costs associated with such a strategy may completely offset any gains involved with such efforts.
Effective Bond Strategies
Fixed-income investment management strategies can be characterized as non-fore-casting or active approaches. Although U.S. government fixed-income securities are among the most liquid and safest assets in the world, an active manager needs to consistently find undervalued government securities or correctly anticipate interest rate changes in order to achieve superior returns. Within this context, research has shown that over the long term, active fixed-income mutual fund managers do not consistently provide excess market returns
(Philpot, Hearth, Rimbey, and Schuman 1998; Blake, Elton, and Gruber 1995; Blake, et al. 1993). In addition, Brooks and Gray (2004) detail the inability of the nation’s most distinguished economists to accurately forecast interest-rate changes. A naïve strategy that simply chose the cur-rent yield as the best estimate of the future six-month yield on long-term government bonds was superior to expert opinion.
Using Morningstar Principia, we selected all high-quality short-term (HS) and high-quality intermediate-term (HI) bond mutual funds with at least a ten-year performance history. A further description of the selection criteria is described in the appendix. Mutual fund returns were sepa-rated into quartiles based on their perform-ance history and compared against the average manager return and an index mutual fund. Because our investigation has shown the significant benefits of shorter-term U.S. fixed-income securities, we will include a comparable U.S. government fixed-income index for the HS and HI categories.
Assessing the differences in average returns between the top quartile and general bond managers should indicate the top managers’ ability to add value and the degree of efficiency in the fixed-income markets. By separating the active managers into quartiles, we are able to address concerns about presenting average returns that wash out the returns of the superior man-agers. This also allows us to measure the degree to which the top quartile managers outperformed their peers and corresponding index. Large differences between the top-performing managers and the index would indicate that the market is providing opportunities for the astute manager to take advantage of certain mispriced assets or information regarding future interest rates.
Small differences in return would indicate that the particular market is fairly efficient and that expenses may play a significant role in differentiating performance.
In both the HS and HI categories, the average bond fund return was less than the government index and a corresponding index fund. For the ten years ending in
8/30/2004, the average HS bond manager returned 5.20 percent versus 5.76 percent for the Lehman Brothers 1–3 Year Government Index and 6.17 percent for the Van-guard short-term index fund. HI bond managers underperformed the Lehman Brothers Intermediate Government Index and the Vanguard intermediate-term index fund, 6.36 percent versus 6.62 percent and 7.89 percent respectively.
Even though active managers in both groups invest in riskier corporate securities, on average they did not outperform an index of similar duration made up of “risk-free” government securities. Results also indicate that for the HS and HI categories, the government index outperformed all but the top-performing quartile of fixed-income managers. Surprisingly, the bond index funds finished in the top first percentile of all bond mutual funds in the HI category (n = 549) and in the top sixth percentile in the HS category (n = 338).
The average of the top-quartile fixed-income managers in the HS and HI classes also underperformed their corresponding index funds by .18 percent and .68 percent annually over an entire decade. Although underperformance by the top managers was by a slim margin, an advisor choosing these managers ten years ago would have likely expected the top managers to have outperformed an index fund by a significant amount. If not, why take on the extra risk and exposure? These top HS and HI managers outperformed a much safer government fixed-income index by only .23 percent and .59 percent annually. Com-pared with their peer average, the top-quartile managers had an advantage of .79 per-cent in the HS and .85 percent in the HI categories. These results indicate that even if an advisor is able to select a manager who outperforms 75 percent of his or her peers, an active management approach among the HS and HI bond classes may still underperform an index strategy. In spite of the attempts by active managers to add value over the long term, historical returns data in these highly efficient asset classes do not support their ability to match, much less surpass, the returns avail-able for index funds.
Top-quartile managers’ risk-adjusted returns do not compare favorably with a corresponding index fund and less risky government index. The Sharpe ratios of the Lehman Brothers 1–3 Year Government Index (1.05) and the Lehman Brothers Government Intermediate Index (.83) are the highest in their respective HS and HI categories. Hence, top-quartile man-agers did not provide superior risk-adjusted returns relative to a U.S. government fixed-income index. The short-term (.89) and intermediate-term (.69) index funds also provide favorable Sharpe ratios when com-pared with the different quartiles. Not surprisingly, Sharpe ratios across the quartiles consistently decline among the lower-per-forming quartiles. Results indicate that advisors choosing active HS and HI fixed-income managers to provide superior risk-adjusted returns face serious obstacles.
What about the “exceptional” fixed-income managers? To address this, we ana-lyzed the credit composition of the top ten performing actively managed portfolios across the HS and HI mutual funds. Because these managers are classified within the same Morningstar style boxes, their market-excess returns should be due to manager skill rather than simply expos-ing their portfolio to riskier assets. Table 7 provides the average credit breakdown as of 8/30/2003 of the top ten active managers in each of the HS and HI categories com-pared with the Vanguard short-term and intermediate-term index funds.
The top ten active HS managers had 7 percent less exposure to government securities and 12 percent more exposure to AAA–A rated securities than the Vanguard short-term index fund. These managers did have 5 percent less exposure to BBB rated securities. Cumulatively, however, the top ten managers in the HS category had more credit-risk exposure than the index fund. A similar pattern was found in the HI cate-gory. The Vanguard intermediate-term index fund had over 16.5 percent more exposure to Treasury securities and less combined exposure to AAA–BBB rated securities than the top ten managers. In addition, these active managers had over an 8 percent average exposure to below-invest-ment-grade rated securities. Much of the value-added returns from these actively managed portfolios seem to stem from additional credit and call-option risk. By populating their portfolios with a greater amount of lower grade, callable, or longer-term securities, bond managers attempt to increase their expected returns over the index. The results in Table 6 also point to the important role mutual fund expenses had on the top-quartile performing funds. The highest-return quartile had the lowest aver-age expense ratio. Moreover, this pattern continued with lower-quartile returns associated with higher expense ratios. As with equity mutual funds, there seems to be a direct inverse relationship between invest ment performance and fund expenses. The higher the investment return, the lower the fund expense ratio. Differences in expense ratios explain much of the differences in net returns. Differences in gross returns (that is, the sum of net returns and expense ratios) were much narrower than differences in net returns. The difference in gross returns between the top quartile and the average fund decreases by over 34 per-cent. Among the HI fund category, this difference decreases by over 49 percent. Thus, within these asset classes, one of the most important decisions an actively man-aged bond mutual fund makes is not its anticipation of the directional change in interest rates but rather how much to charge investors.
Further analysis of the top-performing funds reveals that a variable maturity strategy was the top-performing mutual fund in the HS category. Developed by Fama(1984), a variable maturity approach is a fixed-income strategy that uses the infor-mation in the yield curve to buy the securi-ties that have the best risk/return profile for the investor. Although this strategy selects securities based on the future rates, it is a structured approach that does not actively attempt to find inefficiencies in the bond market or forecast future interest rate movements. In addition to the variable maturity fund, the Vanguard short-term index fund was the 18th best-performing HS fund (n = 338).
In the HI category, the Vanguard inter-mediate fixed-income index fund was the 6th best-performing fund and another vari-able-maturity fund was the 21st best-per-forming fund (n = 549). Within the high-quality short- to intermediate-term bond market, there do not seem to be significant opportunities for outperformance by find-ing undervalued securities or forecasting interest-rate movements. Rather than being dependent on consistently finding undervalued securities or outmaneuvering other fixed-income managers, these non-actively managed funds simply and efficiently capture the market rates of return within their respective asset classes. If the underlying reason for holding fixed-income assets is to reduce portfolio risk, then a non-active fixed-income strategy should be the strategy of choice within the HS and HI classes.
Conclusions and Implications
Advisors have a wide array of available investment options to construct a diversi-fied portfolio. Fixed-income instruments are used within a portfolio to reduce volatility and provide a more consistent distribution stream for clients. This paper attempts to serve as a starting point for advisors by identifying the types of fixed-income instruments that best complement primary portfolio needs. We then analyzed various management strategies within the high-quality short-term and high-quality long-term bond funds to assess the most effective way of implementing a fixed-income strategy within a portfolio.
Some advisors use bonds that exhibit equity characteristics or take on extra credit and option risk in order to increase the expected returns from their fixed-income allocation. By taking on extra risks, advisors may undermine the fundamental reason for holding fixed-income instru-ments. Although under normal economic conditions these types of fixed-income instruments may have a higher expected return than government securities, investors take on extra risk factors that may have been more effectively allocated else-where. The decoupling effect of U.S. gov-ernment securities to equities returns during market downturns offers effective diversification benefits when it is needed most. Thus, holding non-callable instru-ments backed by the U.S. government offers significant protection in times of financial crisis while reducing the long-term opportunity cost of bonds.
income securities must then decide on the length of the bond maturities for their port-folios. Our findings support previous research (Domian et al. 1998, Ilmanen 1996) indicating that the most favorable risk/reward ratio for extending maturities is within the shorter end of the bond market. We find that most of the term premium for extending maturities occurs at around one year and that extending maturities past five years does not provide a more favorable risk/reward ratio. Sharpe ratios consistently decline for maturities past five years. Our research also shows that over significant time periods, shorter-term securities frequently outperform longer-term securities. Advisors seeking to provide excess returns by taking on more term risk may find that the extra return is not worth the risk. These advisors would most likely receive higher rewards by simply switching to equities. As Domian et al. point out, the debt-equity decision is more important than the bond maturity decision.
An advisor can employ many fixed-income management strategies within a portfolio. Our research indicates that active bond managers in HS and HI bond funds did not outperform a corresponding index fund over the past ten years. Additionally, even the top quartile of these managers could not match the risk-adjusted returns provided by an index fund or an index made up of less risky U.S. government securities. A breakdown of the bond qual-ity of the top ten actively managed funds in each category indicates that they take on more credit risk exposure than a compara-ble passive index fund. Results also lend support to previous findings (Potts and Reichenstein 2004, Reichenstein 1999), indicating that differences in fund expense ratios were pronounced among the best-and worst-performing funds and may likely explain much of the differences in fund performance.
Future research should focus on the different types of fixed-income strategies within the HS and HI asset classes to further identify how these strategies would have fared across different interest-rate scenarios. Although shorter-term U.S. govern-ment securities should play an integral role in a client’s portfolio, individual client situ-ations and requirements also influence the selection and strategy of fixed-income instruments. For example, certain clients may have significant tax issues that warrant the inclusion of municipal fixed-income instruments. Treasury inflation-protected securities (TIPS) may also be an option for tax-exempt investors concerned about real returns. In addition, recent studies (Kothari and Shanken 2004, Roll 2004) have begun to identify the importance of TIPS within diversified portfolios. Additionally, the role of longer-term securities in bond immu-nization should be examined. Ultimately, although we have established a foundation within which to implement a fixed-income approach, all these issues need to be considered to provide a more personalized client solution.
Blake, Christopher R., E. J. Elton, and M. Gruber. 1993. “The Performance of Bond Mutual Funds.” Journal of Busi-ness 66, 3 (July): 371–403.
Blake, Christopher R., E. J. Elton, and M.Gruber. 1995. “Fundamental Economic Variables, Expected Returns, and Bond Fund Performance.” Journal of Finance 50, 4: 1229–1256.
Brooks, Robert and J. B. Gray. 2004. “History of Forecasters.” Journal of Portfolio Management 31, 1 (Fall): 113–117. Domian, Dale L., T. S. Maness, W. Reichenstein. 1998. “Rewards to Extending Maturity.” Journal of Portfo-lio Management 24 (Spring): 77–92.
Fama, Eugene F. 1984. “The Information in the Term Structure.” Journal of Economics 13 (December): 509–528. Gulko, Les. 2002. “Decoupling.” Journal of Portfolio Management. 28 (Spring): 59–66.
Ilmanen, Antii. 1996. “Does Duration Extension Enhance Long-Term Expected Returns?” Journal of Fixed Income 6, 2 (September): 23–36.
Kothari, S. P., and J. Shanken. 2004. “Asset Allocation with Inflation Protected Bonds.” Financial Analysts Journal 60 (January/February): 54–70.
Philpot, James, D. Heart, J. N. Rimbey, and C. T. Schuman. 1998. “Active Man-agement, Fund Size and Bond Mutual Fund Returns.” Financial Review 33, 2: 115–125.
Potts, Tom and W. Reichenstein. 2004. “Predictability of Fixed-Income Fund Returns.” Journal of Financial Planning. 7 (November): 62–71.
Reichenstein, William. 1999. “Bond Fund Returns and Expenses: A Study of Bond Market Efficiency.” Journal of Investing 8, 4: 8–16.
Roll, Richard. 2004. “Empirical TIPS.” Financial Analysts Journal 60 (Janu-ary/February): 31–53.
McLean Asset Management Corporation (MAMC) is a SEC registered investment adviser. The content of this publication reflects the views of McLean Asset Management Corporation (MAMC) and sources deemed by MAMC to be reliable. There are many different interpretations of investment statistics and many different ideas about how to best use them. Past performance is not indicative of future performance. The information provided is for educational purposes only and does not constitute an offer to sell or a solicitation of an offer to buy or sell securities. There are no warranties, expressed or implied, as to accuracy, completeness, or results obtained from any information on this presentation. Indexes are not available for direct investment. All investments involve risk.
The information throughout this presentation, whether stock quotes, charts, articles, or any other statements regarding market or other financial information, is obtained from sources which we, and our suppliers believe to be reliable, but we do not warrant or guarantee the timeliness or accuracy of this information. Neither our information providers nor we shall be liable for any errors or inaccuracies, regardless of cause, or the lack of timeliness of, or for any delay or interruption in the transmission there of to the user. MAMC only transacts business in states where it is properly registered, or excluded or exempted from registration requirements. It does not provide tax, legal, or accounting advice. The information contained in this presentation does not take into account your particular investment objectives, financial situation, or needs, and you should, in considering this material, discuss your individual circumstances with professionals in those areas before making any decisions.