Basically, we were guessing on interest rates. What we’ve come to believe is that no one can guess interest rates.
—Fred Henning, head offixed-incomeinvesting at Fidelity Investments, quoted in the Los Angeles Times, July 22, 1997
Today’s investors find it inconceivable that life might be better without so much information. Investors find it hard to believe that ignoring the vast majority of investment noise might actually improve investment performance. The idea sounds too risky because it is so contrary to their accepted and reinforced actions.
—Richard Bernstein,firstvice president and chief quantitative strategist at Merrill Lynch
The most costly of all follies is to believe passionately in the palpably not true.
—H. L. Mencken
There are two general theories about how markets work. The A. first, which is the conventional wisdom (because it is accepted by almost all investors), is that there are smart people working hard who can uncover securities that have been somehow mispriced by the market. In the case of equities (stocks) that means that there are people who can identify which stocks are undervalued or overvalued, and there are also people who can identify when the bear is about to emerge from its hibernation (it’s time to get out of the market) and when the bull is about to start another rampage (it’s time to jump in and get fully invested). The former is the art of stock selection and the latter is the art of market timing. Together they form the practice of active management.
When it comes to fixed-income investing the same two active management strategies can be employed. Timing the market would involve guessing when interest rates are going to rise (you would sell longer-term instruments and buy shorter-term ones) or fall (the reverse strategy would be used). Security selection would entail identifying which securities are mispriced (under- or over-valued) by the market. For example, a security might be rated A by the major rating agencies (Standard & Poor’s, Moody’s, and Fitch), yet an analyst might decide that it really deserves a A A rating and it will be upgraded when the agencies figure it out (get it right). The analyst, therefore, would recommend the bond for purchase because, if an upgrade in credit rating were to occur, the price of the bond would be expected to rise and a profit could be made.
Investors need to understand that just because something is conventional wisdom does not make it correct. For example, “The earth is flat” was once conventional wisdom. As author Nicholas Chamfort noted: “There are well-dressed foolish ideas just as there are well-dressed fools.”
Another great, and perhaps more relevant, example of conventional wisdom being wrong is the case of Galileo, the Italian astronomer who lived in the sixteenth and seventeenth centuries. He spent the last eight years of his life under house arrest, a punishment ordered by the Roman Catholic Church for committing the “crime” of believing in and teaching the doctrines of Copernicus. Galileo’s conflict with the Church arose because he was fighting the accepted doctrine that the earth was the center of the universe. Ptolemy, a Greek astronomer, had proposed this theory in the second century. It went unchallenged until 1530 when Copernicus published his major work, On the Revolution of Celestial Spheres, which stated that the earth rotated around the sun rather than the other way around. The example of Galileo demonstrates that even when millions of people believe a foolish thing it is still a foolish thing. As Don Marquis, writer, poet, and journalist, noted, “An idea is not responsible for the people who believe in it.”
People cling to the infallibility of an idea even when there is overwhelming evidence that the idea has no basis in reality— particularly when a powerful establishment finds it in its interest to resist change. In Galileo’s case, the establishment was the Church. In the case of the belief in active management, the establishment comprises Wall Street, most of the mutual fund industry, and the financial media. All of them would make far less money if investors were fully aware of the failure of active management. This brings us to the second theory on how markets work.
The foundation of the second theory is what is called the efficient market hypothesis (EMH). The following is a very simple definition of the EMH: Prices of securities traded in the public market are the best available estimates of their real value because of the highly efficient pricing mechanism inherent in the market. If this is true, then the conventional wisdom must be wrong.
The EMH is based on over fifty years of academic research on how capital markets work. Here is what Michael C. Jensen, Harvard professor of business administration, had to say about the EMH: “There is no other proposition in economics that has more solid empirical evidence supporting it than the efficient market hypothesis. In the literature of finance, accounting, and the economics of uncertainty, the EMH is accepted as a fact of life.” 1
If markets are efficient, which is one of the fundamental tenets of the investment strategy recommended in this book, active management is not likely to be able to add value after the expenses of the efforts are deducted. The reason is that the market price is, as Jensen stated, the best estimate of the correct price. If another price were the best estimate, the market would quote that price instead.
There are two important issues related to the EMH that often confuse investors. The first is that the EMH does not preclude the possibility that some investors will outperform the market. instead, it says that there is no way for investors to identify ahead of time who the few that might do so will be. As you will see, relying, for example, on past successful performance simply doesn’t work. There is no way, therefore, to identify ahead of time the future winners. And, unfortunately, we can only buy future returns, not past returns.
The second mistaken notion people have about the EMH is that it does not preclude the possibility of the market being wrong. For example, it might turn out that the market might have been overly optimistic about the outlook for interest rates (and thus rates were “too low”). The problem is that there is no evidence that after accounting for expenses there are investors who can persistently exploit any such pricing errors. Markets can be irrationally exuberant or irrationally pessimistic. However, unless investors can exploit such irrationality the market can be said to be efficient. Active managers not only have to cover the cost of their research, but they also must add enough value to cover all trading costs. In addition, for taxable accounts they would have to add enough value to cover the increased taxes that are often generated from their increased trading activity.
Unfortunately for investors who believe in the conventional wisdom (that active management is likely to add value), there is a body of overwhelming evidence that suggests that the markets, both for equities and bonds, are in fact highly efficient. For example, Mark M. Carhart’s study on equity mutual fund performance analyzed 1,892 funds for the period 1962–93. He found that the average equity fund underperformed its appropriate style bench-mark by about 1.8 percent per annum on a pretax basis. Carhart also found no evidence of any persistence in outperformance beyond the randomly expected (though he did find some persistence among the worst performers). 2 There are now probably hundreds of studies on the subject and they basically come to the same conclusion—the markets are highly efficient. Even those researchers that conclude the markets are not perfectly efficient often suggest that investors are best served if they act as if they were—because the costs of trying to exploit inefficiencies are likely to exceed the benefits.
Consider what the American Law Institute had to say about market efficiency when in 1992 it wrote the Third Restatement of the Prudent Investor Rule. After looking at the evidence they concluded that:
“Economic evidence shows that, from a typical investment perspective, the major capital markets of this country are highly efficient, in the sense that available information is rapidly digested and reflected in the market prices of securities.
“As a result, fiduciaries and other investors are confronted with potent evidence that the application of expertise, investigation, and diligence in efforts to ‘beat the market’ in these publicly traded securities ordinarily promises little or no payoff, or even a negative payoff after taking account of research and transaction costs.
“Empirical research supporting the theory of efficient markets reveals that in such markets skilled professionals have rarely been able to identify underpriced securities (that is, to outguess the market with respect to future return) with any regularity.
“In fact, evidence shows that there is little correlation between fund managers’ earlier successes and their ability to produce above-market returns in subsequent periods.” 3
In his excellent book, The Prudent Investor Act (a must read for anyone with trustee responsibilities as well as for anyone managing their own assets), W. Scott Simon reached the conclusion that under the Uniform Prudent Investor Act, which sets forth standards that govern the investment activities of trustees, and is currently the law in most states, “passive investing (e.g., index and [passive] asset class funds) appears to be the standard for investing and managing trust portfolios.” 4
The following are the results of just two of the many studies that could be cited on the attempts of active management to exploit market inefficiencies in the fixed-income markets. A study covering as many as 361 bond funds showed that the average actively managed bond fund underperforms its benchmark index by 0.85 percent per annum. 5 Another study found that only 128 (16 percent) out of 800 fixed-income funds beat their relevant bench-mark over the ten-year period covered. 6 Of course, being a loser’s game does not mean there are not some winners. That leaves the hope that we can identify the few winners ahead of time. Unfortunately, the evidence suggests that believing so would be the triumph of hope over experience. For example, John Bogle of Vanguard studied the performance of bond funds and concluded that “although past absolute returns of bond funds are a flawed predictor of future returns, there is a fairly easy way to predict future relative returns.” Bogle found that “the superior funds could have been systemically identified based solely on their lower expense ratios.” 7 Other studies on the subject, including those on municipal bond funds, all reach the same conclusions:
The results of a study by Morningstar demonstrate both the importance of costs and that the past performance of actively managed funds is a poor predictor of future performance. They tested funds with strong performance and high costs against those with poor past performance with low costs. “Sure enough, those with low costs outperformed in the following period.” 8
Why did all these studies come to the conclusion that bond fund managers charge Georgia O’Keeffe prices and deliver paintby-numbers results? The EMH provides the answer: The market’s efficiency prevents active managers from persistently exploiting any mispricing. And as difficult as it is for active managers to add value when it comes to equity investing, it is much harder for them to add value in fixed-income investing. Let’s see why this is true.
First, as we have already discussed, with U.S. Treasury debt, all bonds of the same maturity will provide the same return (with the exception of a few bonds that have call provisions). Thus, there will be no differentiation in performance, and, therefore, no ability to add value via security selection. If we restrict holdings to the highest investment grades, there is an extremely limited ability to add value via security selection (because credit risk is very low). That leaves interest rate forecasting as the only way an active manager might add value in any significant way. William Sherden, author of the wonderful book The Fortune Sellers, reviewed the leading research on forecasting accuracy from 1979 to 1995 and covering forecasts made from 1970 to 1995. He concluded that:
Michael Evans, founder of Chase Economics, confessed: “The problem with macro [economic] forecasting is that no one can do it.” 10 Since the underlying basis of interest rate forecasts is an economic forecast, the evidence suggests that bond market strategists who predict bull and bear markets will have no greater success than do the economists. If active managers are highly unlikely to add value via either security selection or market timing, the only conclusions we can draw are that the conventional wisdom is wrong and that the markets are highly efficient.
Once we conclude that the market is efficient, the winning strategy becomes obvious: It pays to be a passive buy-and-hold investor using only low-cost investment vehicles to implement the investment plan. The only question remaining is which instruments we should invest in. In order to determine the answer, we need to understand what the major determinants of fixed-income risk and expected return are.
In 1977 Bill James self-published the book 7977 Baseball Abstract: Featuring 18 Categories of Statistical Information That You Just Can’t Find Anywhere Else. Seventy-five people found the book of sufficient interest to buy it. 11 Today James’s annual edition (now called The Bill James Handbook) is considered a must read for all serious fans of our national pastime.
James demonstrated, through vigorous research, that certain statistics are more crucial than others in determining the effectiveness of a player. Among his many findings: A player’s batting average and the number of home runs he hits are not as important as people have assumed. James found that other statistics were more vital, namely, the total of a player’s on base percentage and his slugging average.
What James did was nothing less than revolutionize the way people think about baseball statistics and how to build a winning team. In fact, most teams today have statistical experts (called sabermaticians) on their staffs. Michael Lewis’s book Moneyball explains how Billy Beane, the general manager of the Oakland Athletics, used sabermaticians to build a winning team despite the constraint of having a very limited payroll.
Just as James revolutionized the way we think about the game of baseball by assessing which factors were the most significant in determining the impact a player had on the outcome of a game, the publication of the paper “The Cross-Section of Expected Stock Returns,” by professors Eugene F. Fama and Kenneth R. French, in the Journal of Finance in June 1992 had a dramatic impact on the field of financial economics. The Fama-French research produced what has become known as the three-factor model, and it explains virtually all (as much as 97 percent) of the variability in returns of diversified U.S. stock portfolios.
The conclusion that can be drawn from the Fama-French research is that the markets are efficient and, therefore, the vast majority of the returns one can expect from a diversified equity portfolio are unrelated to the ability either to pick stocks or to time the market. Instead, it is the degree of exposure to what Fama and French called risk factors that determines almost all of the variability in returns. The first risk factor in the three-factor equity model is the amount of exposure of a portfolio to the risk factor of the overall stock market. All equities have some exposure to this risk factor. Since equities are riskier than fixed-income investments, they provide greater expected returns. The second risk factor is the size of a company as determined by market capitalization. Intuitively we know that small companies are riskier than large companies—and they must provide greater expected returns. The third risk factor takes value into consideration. High book-to-market (value) stocks are intuitively riskier than low book-to-market (growth) stocks—and they must provide greater expected returns.
Thanks to professors Fama and French we have a similar two-factor model to explain the returns of fixed-income portfolios. The two risk factors are term and default (credit risk). The longer the term-to-maturity, the greater the risk, and the lower the credit rating, the greater the risk. And the markets compensate investors for taking risk with higher expected returns. Note that individual security selection and market timing do not play a significant role in explaining returns, and thus should not be expected to add value. If they did explain returns, then we would see evidence that active managers were adding value, not subtracting it.
The implication for investors is that the winning strategy in fixed-income markets, whether taxable or tax-exempt, is to choose the lowest-cost fund (and passive funds are likely to be the lowest cost) that meets your credit and maturity criteria. Alternatively, if your portfolio is large enough, and you can do so at a very low cost, the winning strategy is to build your own individually tailored portfolio. Note that for U.S. government securities mutual funds do not provide one of their greatest benefits, the diversification of risk (since U.S. Treasury securities have no credit risk). Thus building your own portfolio might be a good alternative.
With corporate and tax-exempt issues, however, credit risk is a consideration. That means that corporate and tax-exempt municipal bond funds do provide the benefit of diversification. Therefore, only those investors with portfolios large enough to achieve effective diversification (e.g., $500,000) should attempt to construct their own portfolios. In addition, because trading costs in the corporate and municipal bond markets are much greater than they are in the Treasury bond market, only investors who are almost certain that they will be able to buy and hold to maturity should own individual bonds.
Academic research has found that over long periods of time, while investors have been compensated for accepting the risk of owning longer-maturity fixed-income assets, this relationship has broken down beyond two to three years. Research on the relationship between risk and return has shown that for the period 1964–2003: 12
An important question for investors is how to determine the most efficient maturity in terms of return relative to risk. Nobel Prize-winner William Sharpe provided the answer when he developed what has become known as the Sharpe ratio.
The Sharpe ratio is a measure of return relative to risk. It is derived by first subtracting the average rate of return on riskless one-month Treasury bills from the average annual rate of return earned on the asset, then dividing the result by the standard deviation of the asset. The higher the Sharpe ratio, the more efficient is the investment in delivering returns relative to risk. The Sharpe ratio has been about 0.40 at the one-year maturity but falls to about 0.26 if we extend the maturity to about five years, and it continues to fall as we extend the maturity. Thus holding assets with a maturity of about one to two years is the prudent strategy for those investors wishing to maximize the risk-reward relationship. Note, however, that there may also be other considerations (to be discussed later) that should be taken into account in determining the term risk one is willing to take, including whether one is in the accumulation or withdrawal phase of one’s investment life cycle.
It is important to note that the yield curve for municipal bonds is almost always steeper than it is for Treasuries. There are three basic reasons for this difference. Perhaps the one of greatest significance is that there is a greater supply of long-term bonds from issuers than there is demand for long-term municipal securities from investors. Municipalities are generally trying to match their liabilities with the long-term nature of their assets (e.g., highways, bridges, buildings, parks). On the other hand, most investor demand is for securities with short to intermediate maturities. In order for the market to absorb all the supply, prices adjust downward (yields rise). The two other factors that cause the municipal bond curve to be steeper than the Treasury curve are the credit risk of municipal securities and the potential for the loss of their tax-exempt status.
To illustrate the differences in the slopes of the different curves consider the following: If a five-year municipal bond were trading at 80 percent of a Treasury note with the same maturity, a ten-year municipal bond might trade at 85 percent, and a twenty-year municipal bond might trade at 90 percent. The result is that the risk-reward relationship for extending the maturity of an investment is better in the municipal bond market. The higher yield may make it attractive to consider longer maturities than would be the case for Treasuries or corporate bonds. Therefore, depending on the shape of the municipal bond yield curve at the time the investment decision is made, it may be appropriate to extend the maturity of municipal bonds perhaps to an average of about five to seven years (versus the one to two years for taxable instruments).
At this point we need to ask the question: If long-term bonds are riskier than short-term ones, and the market is efficient at pricing for risk, why have investors in long-term bonds not been compensated for the greater risk they have taken (using standard deviation as a measure of risk)? There is a good explanation for this seeming risk-return anomaly—while standard deviation is a measure of risk, it is not the only one. When thinking about risk, some investors care about issues other than volatility. There are many investors, such as pension plans, that have fixed long-term obligations. Insurance companies are another good example of investors who have relatively well defined long-term obligations. In order to create a match between the term of their defined liabilities (the pension obligations due to past and current employees) and the term of their assets (thereby eliminating risk), pension plans are willing to accept the interim price risk of the assets themselves. Now let’s consider an alternative strategy.
To eliminate the price risk of holding long-term bonds, a pension plan buys three-month Treasury bills and continually rolls them over until they are needed to fund obligations. While there will be virtually no volatility in the price of their holdings, because the rate of return that will be earned over the long term is unknown, the pension plan would be running a great risk as to its ability to earn a return sufficient to fund its obligations. The result is that the investor demand for longer-maturity bonds exceeds the demand by issuers for liabilities of that length. Prices rise (and yields fall) when demand exceeds supply. In this case, the price of long-term bonds has risen sufficiently to make them bad investments for those investors not needing them to match a liability of similar length. Note that since these pension plans are not buyers of municipal bonds, this does not apply to the municipal bond yield curve.
There is another reason to consider not owning longer-term fixed-income instruments, especially if you are in the accumulation phase of investing. During the accumulation phase the main reason for holding fixed-income assets is generally to provide a safety net to anchor your portfolio during bear markets, allowing you to stay disciplined. In order for the safety net to be effective, the assets it holds must have low correlation with the risky equity portion of the portfolio. Unfortunately, the longer the maturity, the higher the correlation of fixed-income assets to equities. Also, unfortunately, that higher correlation with the equity portion of your portfolio can appear at just the wrong time. There may be times when interest rates rise, bond prices fall, and the stock market falls at the same time. Just when you need low correlation, you may get high correlation. The following are the correlations between Treasury instruments of various maturities and the S&P 500 and EAFE indices. 13 Remember that the lower the correlation, the more effective the diversification, and the lower the overall risk of the portfolio. Note also that there is basically no correlation between short-term maturities (up to one year) and U.S. equities and no correlation for up to five years between short-term maturities and international equities.
Annual Correlation Data
| Maturity | Correlation with S&P 500 Index 1964–2003 | Correlation with EAFE Index 1969–2003 |
| One month | 0.02 | -0.11 |
| Six months | 0.03 | -0.11 |
| One year | 0.05 | -0.18 |
| Five years | 0.20 | -0.02 |
| Twenty years | 0.26 | 0.09 |
Source: Dimensional Fund Advisors
As the above figures demonstrate, the risk of having relatively higher correlation between equities and fixed-income instruments can be avoided by buying short-term fixed-income instruments— they have essentially no correlation with equities.
With the preceding information, we can determine that for those investors using fixed-income assets to reduce the risk of an equity portfolio the prudent strategy is to own only very short-term fixed-income assets of the highest investment quality. Here’s why.
Since the main purpose of fixed-income assets is to reduce the volatility of the overall portfolio, investors should include fixed-income assets that have low volatility. Short-term fixed-income assets have both low volatility and low correlation with the equity portion of the portfolio. By limiting the maturity of the fixed-income portion of the portfolio to just one year, we get most of the yield benefit and accept only moderate risk (a standard deviation of only 2 percent). The benefit of lower volatility of the asset class itself, combined with the benefit of the reduced volatility of the overall portfolio, seems a small price to pay for giving up the extra thirty basis points in annual returns that have been gained by extending the maturity of the fixed-income assets to five years. Remember, in a 60 percent equity and 40 percent fixed-income portfolio, that extra thirty basis points (0.3 percent) becomes an added return on the overall portfolio of only twelve basis points per annum (0.3 percent x 40 percent).
We can further improve on this scenario by including international short-term fixed-income assets within the fixed-income allocation (as long as the assets are hedged against currency risk). The reason is that their inclusion should further reduce volatility, since not all international fixed-income markets fluctuate in the same direction at the same time or by the same amount. The lack of perfect correlation will reduce the overall volatility of the fixed-income portion of the portfolio. The subject of international fixed-income securities is covered in depth in chapter 8.
There is another fixed-income investment strategy (instead of buying and holding) that is based on the research of Eugene F. Fama and Robert R. Bliss. Their study “The Information in Long-Maturity Forward Rates,” published in the September 1987 edition of the American Economic Review, covered the period from 1964 through 1985 and examined the historical returns of Treasury instruments with maturities out to five years.
Fama and Bliss concluded that today’s yield curve contains information about future yield curves—current forward rates provide the best forecast of future spot interest rates. In other words, today’s yield curve is the best estimate we have of what future yield curves will be. For example, if today’s yield on a five-year Treasury note is 5 percent, at any point in the future the best estimate of what the five-year Treasury yield will be is also 5 percent—our best estimate is that the yield curve will not change over time. They also found that the longer the horizon, the greater the forecasting power contained in today’s yield curve.
The results found by Fama and Bliss are in direct conflict with the expectations theory of the term structure of interest rates—that yields on Treasuries of different maturities are related primarily by market expectations of future yields. In other words, the yield of a long-term bond will equal the average of the expected short-term interest rates over the same period. Fama and Bliss provided us with not only a better explanation, but also with insights that can help us improve on market returns.
Let’s assume that, based on academic research, we have determined that we want to limit our fixed-income investments to a maturity of two years. After checking our Bloomberg screen we see that a one-month Treasury bill is yielding 1 percent, a U.S. Treasury note with a term-to-maturity of one year is yielding 2 percent, and a similar instrument with a two-year term-to-maturity is yielding 3 percent. Our investment choices would seem to be to buy:
If we had a clear crystal ball that could correctly forecast interest rates we would know which choice would prove to be the best. This is mostly what active management of fixed-income portfolios attempts to do. If we knew rates were about to fall sharply we would choose alternative 3. If we thought rates would rise sharply in the near future we would choose alternative 1. Unfortunately, as we have seen, all interest-rate forecasting balls are very cloudy. However, Fama and Bliss’s work suggests a strategy with an alternative approach. The strategy is based upon the information contained in the current yield curve.
Instead of buying and holding the two-year instrument until maturity we can consider buying the two-year note and selling it in one year to buy another two-year note, which could then be sold after holding it for one year. What rate of return can we expect to earn by executing this strategy? The Fama and Bliss study tells us how to estimate that rate. Note that to keep the example as simple as possible the following calculations ignore the effect of compounding.
If the current two-year rate is 3 percent, and the current one-year rate is 2 percent, then the one-year forward rate (the one-year rate one year from today) must be 4 percent (ignoring compounding effects). The math is simple. In order to earn 3 percent for the full two years, if we earn only 2 percent for the first year, the second year we must earn 4 percent ([2 percent + 4 percent] / 2 = 3 percent). Before proceeding, it is important to note that under the expectations theory of interest rates we would conclude the best forecast of the one-year rate one year from today would be 4 percent. Fama and Bliss showed this to be incorrect. Instead, the best estimate of what the one-year rate will be one year from today is the current one-year rate of 2 percent. The higher one-year forward rate at 4 percent is not a forecast of higher interest rates in the future. Instead, it is a risk premium. In this case the risk is term risk.
Returning to our example, let’s see how it works if the current yield curve remains unchanged—the best assumption we can make. We buy the two-year instrument which today is yielding 3 percent. If we hold it for one year we will have earned 3 percent for that year. Now there is only one year left to maturity. Our assumption is that the then current one-year note will be yielding 2 percent (the best estimate of tomorrow’s yield curve is today’s). Since we will be holding an instrument with one year left to maturity yielding 3 percent, we could sell that instrument at a 1 percent profit. Our total return for the first year would have been 4 percent. We could then buy the two-year note again and repeat the process, again earning 4 percent.
The shifting-maturity approach finds the point on the yield curve that provides the greatest total return and invests in securities with that maturity, shifting maturities as the yield curve shifts. Dimensional Fund Advisors (DFA), which has successfully utilized this approach since 1983 for their fixed-income funds, adds two caveats to the strategy. The first is when considering shifting maturities, any trading costs must be accounted for. The second is that because longer-maturity bonds have more price risk, DFA imposes an arbitrary rule that a longer maturity must provide at least twenty basis points per annum in higher expected returns in order for DFA to extend the maturity. In other words, if in the above example the one-year bond had an expected return of 2 percent and the two-year of only 2.1 percent, then DFA, because they could expect to earn only an additional 0.1 percent (ten basis points), would not extend the maturity an additional year. In bondspeak this process is called “finding the sweet spot” on the yield curve. The sweet spot is the point at which the yield curve begins to bend sharply to the right—the curve begins to rapidly flatten out, no longer rising at a rate of at least twenty basis points per annum. Note that if the yield curve were either perfectly flat (all yields along the curve were the same) or inverted, the sweet spot would be one month. The more positively sloped the curve, the farther out will be the sweet spot.
What we have learned is that by adopting this very specific shifting-maturity approach investors can expect to earn higher returns than a simple passive or indexing strategy to fixed-income investments would provide. While individual investors are not able to actively trade the yield curve on a daily basis, they can apply a similar strategy. For example, Vanguard offers a short-term, an intermediate-term, and a long-term bond fund. Vanguard publishes the average maturity of their funds, allowing us to employ a shifting-maturity strategy. Let’s see how this would work. Let’s assume that the average maturity of the three Vanguard funds is two, five, and eight years, respectively. Applying our rule of thumb, we would buy the fund with the highest yield, as long as it met the criteria of providing at least twenty basis points of extra yield for each year of extra maturity (due to the tax-exemption available on municipal bonds, the hurdle rate to extend would be perhaps just fifteen basis points). Thus if the intermediate fund, with a maturity of five years, yielded at least sixty basis points (twenty basis points times three years) more than the short-term fund whose maturity is just two years, we would invest in the intermediate fund. If it did not yield at least an additional sixty basis points, we would invest in the shorter fund. We would also compare the yields of the long-term and intermediate fund and do the same mathematical comparison. Investors might perform the comparison on a quarterly basis, and shift accordingly. Investors building their own portfolios of individual bonds can, in theory, also consider doing the same thing. However, in the real world, trading costs at the retail level would likely more than destroy any benefit.
Before considering implementing this strategy you should carefully weigh the consequences of increasing maturity (going farther out on the curve). First, like any strategy it involves risks. While it has produced above-benchmark returns over the long run, it has not outperformed every year. Therefore, discipline is required. Second, if you invest in longer maturities, you will earn higher expected returns, but remember, you are also increasing the correlation of those returns to the equity portion of the portfolio, thus increasing the risk of the overall portfolio (not just the risk of the fixed-income assets).
As we have discussed, with fixed-income investments there are two main risks for which investors should expect to be rewarded (a third risk is that of liquidity, the purchasing of less liquid assets). These risks are the risk of interest rate changes negatively impacting the value of the asset (duration, or term risk) and credit risk. We also noted that since the main purposes of fixed-income investing (beyond providing liquidity for short-term cash needs and unanticipated expenses) are to either reduce the overall risk of a portfolio or to provide a sure and stable cash flow, trying to enhance returns by taking significant credit risk would not be a prudent strategy. Note that the taking of liquidity risk, on the other hand, might be appropriate if that is a risk with which you are not concerned—if, say, you are virtually certain you will be able to hold the securities purchased until maturity. As we have discussed, if this is the case, the liquidity premium might be viewed as a free stop at the dessert tray.
The following table presents the results of the study “Which Risks Have Been Best Rewarded?” The results of the study support our conclusion that the prudent strategy is to limit maturities to the short to intermediate term and to also limit the taking of credit risk to the highest investment grades. The data is consistent with longer-term historical results presented earlier in this chapter. 14 The study covers the relatively brief period 1985–2002, so we have to be careful about drawing conclusions that might only be period specific.
We can make the following observations from the data.
Has Credit Risk Been Rewarded?
| Asset Class | Mean Return | Standard Deviation |
| 1–3 Year Treasuries | 7.3% | 1.9 |
| 1–3 Year AAA/AA Corporate Bonds | 7.9% | 1.8 |
| 1–3 Year A/BBB | 8.0% | 1.9 |
| 7–10 Year Treasuries | 9.6% | 6.5 |
| 7–10 Year AAA/AA | 9.6% | 5.7 |
| 7–10 Year A/BBB | 9.6% | 5.3 |
| 7–10 Year High-Yield Bonds | 8.8% | 7.7 |
U.S. Treasuries have no credit risk. AAA/AA are bonds of the highest investment grade. Bonds rated A/BBB are bonds with the lowest rating that are still considered of investment grade. High-yield bonds are often referred to as junk bonds.
When considering this data, because the results might be period specific, we should also consider the following:
In summary, the historical evidence suggests that, from an overall portfolio perspective, investors in fixed-income securities are best rewarded for taking risk if they adopt the following strategies:
It is important to note that for individuals in the withdrawal stage of investing the recommendation to limit maturities to the short to intermediate part of the yield curve may not be the most appropriate strategy. For investors who have recurring cash flow or income needs, exposure to constantly changing rates (from remaining at the short end of the yield curve) may introduce larger risks to the portfolio in the form of funding shortfalls than the price risks introduced to the portfolio from extending maturities. This is true particularly in cases where the nominal liability stream being funded is relatively well defined and inflexible, and the investor’s intent is to hold bonds to maturity. Thus an investor heavily dependent upon cash flows emanating from their investment portfolio could conclude that the risk of changing interest rates is greater than the risks of more volatile price swings. For these investors a measure of risk such as the Sharpe ratio may not be the most appropriate.
Just as a contractor must have the appropriate tools for each job, investors must know which analytical tools are appropriate for their situation. Using the wrong tools to fix a plumbing problem could result in leaks and water damage. Using the wrong tools to assess the health of your portfolio could result in poor outcomes.
Investors who do choose to extend maturities in order to reduce the risk of falling interest rates must accept that they have increased their exposure to risk of unexpected inflation and have also increased the correlation of their fixed-income portfolio to their equity holdings. For them this might be a prudent trade-off. Note that investors who are willing to accept greater term risk should build a portfolio of individual bonds the maturities of which are tailored to meet their required cash flow need. And to reiterate, only bonds of the highest investment grade should be considered for purchase. Thus the choice of the prudent strategy is dependent on the unique financial needs of each investor, as well as on the risks about which he or she is most concerned.
Although the two-factor model explains almost all of the risks and returns of a fixed-income portfolio, there is an additional risk factor that investors should consider besides credit and maturity. The third factor is liquidity.
The more liquid an investment, the lower will be the costs related to trading the security. Bid-offer spreads are negatively cor-related with liquidity—the more liquid the security, the narrower will be the spread between the bid and the offer. In addition, other trading costs such as transaction fees, market impact costs, and dealer markups will tend to be less as liquidity improves. Thus investors prefer liquid to illiquid investments, all else being equal. The result is that investors require less liquid investments to provide higher yields. However, investors who are highly certain that they will be able to hold a security they purchase until maturity may not have much, if any, concern about the issue of liquidity. For these investors, at least for some portion of their portfolios, holding issues with a liquidity premium can be a prudent decision.
In chapter 3 we discussed the difference between “on-the-run” (newest issue) and “off-the-run” (older issue) securities. The newer security is more liquid and thus investors (or at least traders) are willing to pay a somewhat higher price (accept a somewhat lower yield). For buy-and-hold investors the higher yield of the older security might be considered a free stop at the dessert tray (especially since its maturity is also a bit shorter, making it a bit less susceptible to changes in interest rates). another example will help illustrate the point.
The state of Missouri, a AAA credit, announces a new issue of $500 million of ten-year bonds yielding 4 percent. Given the size of the issue, and the quality of the credit, the issue will be very liquid (at least relative to other municipal bond issues). Now consider an issue with the same exact credit and term risks—a $10 million, fifteen-year State of Missouri general obligation bond that was issued five years ago, and thus has ten years left until maturity. That bond may not even have traded at all in the last several months. That bond will trade in the secondary market to yield a bit more than would the new primary issue. Perhaps the yield would be 4.1 percent, or even a bit higher. Investors who are virtually certain that they will be able to hold to maturity are likely to find the higher-yielding asset to be very attractive. Thus investors who have access to prices that are at or near the wholesale (interdealer) price will often find the secondary market to be a more attractive alternative than the primary market.
We have now completed our review of the efficient market hypothesis and its implications for fixed-income investors. We have also reviewed the risks and rewards of fixed-income investing. We are now ready to discuss the various alternative instruments and vehicles that investors can choose from when building a fixed-income portfolio. We will discuss the nature of each of the alternatives and evaluate the pros and cons of each. Chapters 5 through 9 cover the world of taxable investments. Chapter 10 covers the world of municipal securities.