Enhance Value

Enhancing Returns in Passively Managed Bond Funds

Enhance Value

While the title of this article may seem like an oxymoron, it definitely does not refer to the activity of seeking out mispriced bonds, for the bond market is efficient, with prices incorporating investor expectations of future cash flows and the risk-adjusted discount rates applied to them. The fact that there are only a handful of managers who can claim to have delivered long-term alpha is testament to the fact that active bond managers have had a difficult time adding value in excess of their costs. Even for these few, there is no justification for assuming that alpha will be obtained going forward, and there is always the risk of negative alpha, as we saw with Bill Gross in 2011 with an incorrect call on Treasury bonds. Thus, for most investors, a low-cost passively managed bond fund is a prudent choice. Regarding the efficacy of individual bond portfolios, we addressed that in this article.

Most bond index funds simply track an index defined by a market segment (e.g. long-term investment-grade corporate bonds) by buying a large collection of bonds within that segment. The most popular bond index funds track the Barclays Aggregate U.S. Bond Index which includes both governments and corporates of short, medium, and long-term. For many investors, this is a perfectly sound choice for the bond side of their portfolio, and we have recommended it numerous times for our own clients in 401(k) plans that are not managed by us. One noteworthy change that has occurred in this index over the last decade is that it has become more heavily weighted in short-term government because that is where the majority of new debt has been issued, and we have no problem with that.

Many years ago, while working on behalf of Dimensional Fund Advisors, Nobel laureate Eugene Fama asked the question of whether more can be done to add value to a passively managed bond fund. His affirmative answer is based on one simple assumption—that the best predictor of tomorrow’s yield curve is today’s yield curve. Recall that the yield curve delineates the relationship between the yield on Treasury bonds and the length of their maturity. Below is an example based on bond yields as of 9/30/2014. We can describe it as level for the first year (at close to zero) and then steadily increasing thereafter.

Fama’s “variable maturity” strategy evaluates the yield curve to determine where the highest return potential is offered. For example, in a linearly increasing yield curve such as the one above, a fund that is allowed to purchase bonds up to five years in maturity will maximize its return by purchasing five-year bonds not only whenever it has excess cash from deposits, coupon payments, and maturities, but it may also make sense to sell the bonds it has held for more than a year or two to buy new bonds. Of course, this should only be done if trading costs can be held to a low enough level to not overcome the benefit of moving to higher yields. If the yield curve were inverted (a relatively rare phenomenon), then it would make sense for our hypothetical fund to stick to lower maturities, which again, will increase turnover in the fund, so trades must be executed in a cost-conscious manner. In a flat yield curve environment, lower maturities are preferred because they provide the same yield with lower volatility. There are also yield curves where the best course of action is to buy bonds in the middle of the allowable range. The chart below shows a hypothetical yield curve where the optimal course of action is to purchase the 3-year bond, based on the anticipated 1-year total returns.

 

The crux idea behind the variable maturity strategy is that since the value of the bond changes over time based on the yield curve, it is not only the current yield that is important but also the anticipated change in yield as time-to-maturity decreases. For a section of the yield curve that is steeply increasing, the percentage increase in the value of the bond will be high as the bond moves down the slope over time. Without getting too much into the mathematics of bond returns, suffice it to say that the anticipated total return (current yield plus the benefit from the yield change) can be calculated across the yield curve and the optimal purchase point located. Again, this is all based on the assumption that today’s yield curve is the best predictor of tomorrow’s yield curve. There are some practitioners who may question that assumption because today’s yield curve has implied future yields incorporated within it. For example, if a 1-year bond yields 1% and a 2-year bond yields 2%, then we may expect that a year from now, the 1-year bond will yield about 3% so that an investor who buys the 1-year bond and rolls it over will get the same overall return as the investor who bought the 2-year bond. In answer to this potential objection, the variable maturity strategy is a dynamic one that adjusts to changing yield curves, and yield curves do not usually change drastically over short periods of time. To be clear, the variable maturity strategy provides no guarantee of beating the simpler buy-and-hold approach every single time it is utilized, and it will incur higher trading costs. Nevertheless, in an efficient bond market, we have sound reason to expect it to outperform the buy-and-hold strategy over a long period of time.

Now that we have an idea of the theory behind the process of the variable maturity strategy, we should examine the results it has delivered in practice. All four of the DFA bond funds that IFA advises for our clients utilize the variable maturity strategy, but only three of them have data going back more than twenty years for both the fund and its benchmark. The bar chart below shows how the funds have performed against their prospectus benchmarks over the 23-year period ending 12/31/2013. The length of the period is based on the inception date of the DFA Five-Year Global Fixed Income fund.

For the first two of the three funds, the variable maturity strategy appears to have essentially covered the expense ratio of the fund. For the DFA Five-Year Global fund, the outperformance may be more attributable to risk taken on by the fund (e.g. default risk) that is not reflected in the benchmark. At this point, we may ask the question of whether variable maturity should be considered an active strategy. By our definition of an index fund, the answer would be no because it is a rule of portfolio construction that is held constant regardless of market conditions. Nevertheless, the high turnover ratios of the funds that utilize it suggests active, but we have never agreed to the idea that passive requires the minimization of trading.

The other primary explanation for the good performance relative to benchmarks of the DFA bond funds is DFA's flexible approach to trading which provides a price advantage over approaches that demand immediacy in trading. Both actively managed and traditional index funds target specific bonds either because they are deemed to be "undervalued" by the active fund manager or they are one of the constituents of the targeted index. The problem is that, unlike stocks, at any given time, a particular bond may not be on offer. This means that a motivated buyer (a demander of liquidity) will pay a high cost for executing the trade. Dimensional's approach allows for many different bonds to be eligible for purchase at a given time, which means that they are likely to trade at or better than market prices. Dimensional's own internal study1 of bond trades completed during 2011 and 2012 found that both their buys and sells gained a price advantage of approximately 0.14% for corporate bonds and 0.06% for government agency bonds.     

If you would like to learn more about how IFA utilizes bond funds to help our clients meet their financial goals, please give us a call at 888-643-3133.

1Shao, David and Twardowski, Dave, "Trade Price Advantages of Flexible Fixed Income Portfolios", Dimensional Research, February 2014.
David A. Plecha, "Fixed Income Investing", Dimensional Research, June 2002
Bryce D. Skaff, "Fixed Income Strategies", Dimensional Research, December 2003