Active versus Passive: Fixed-Income

Actively-Managed Bond Funds Underperform…
Given the nature of fixed-income markets (see here, here, and here), one might assume that the majority of active managers should be able to outperform. However, this claim does not hold up. The below chart indicates that although active managers occasionally beat the Bloomberg Barclays US Aggregate Bond index (or simply, “the agg”), most do not in most years.

The plot thickens in our next chart, which shows that active managers generally underperform the same index. However, these managers have bursts of outperformance which consistently coincides with an increase in interest rates.

And our final chart below smooths the above chart even more by extending the rolling performance period from 12 months to 60 months. Again, some episodes of outperformance, but those times are the exception rather than the rule.

…Because The Wrong Benchmarks Are Being Used
Besides implying that most intermediate-term bond managers underperform, the above charts also suggest that the actively managed universe of intermediate term bond funds consistently maintain both a lower duration and credit quality than the benchmark, which means that perhaps the wrong benchmark is being used. Vanguard has done much analysis on this topic and reaches similar conclusions (see: Active Bond-Fund Excess Returns: Is It Alpha…Or Beta?). Anecdotally, I can say that many fixed-income funds that I come across benchmark to “the agg,” despite no semblance of similar holdings or characteristics. The below chart does a good job of illustrating that many benchmarks are misspecified. Adjusting for a fund’s holdings reduces tracking error.

Of course, all of the above charts simply relate to intermediate-term bond funds and not other popular sectors such as high-yield/bank loans or any international bonds. However, the SPIVA data is not too supportive of active fixed-income managers’ value in these sectors and I was hard-pressed to find any supporting charts that showed anything different (at least for anything more than brief windows of time).

Many funds that I know and use are benchmarked to the agg, LIBOR + x spread, or some other arbitrary benchmark. When I ask fund sponsors what they benchmark to, they usually say something to the effect of, “Well we benchmark to the [you name your index]. It’s not a perfect fit or even a good one, but that’s what it is.” A huge number of intermediate bond funds are benchmarked to the agg, but are not trying to beat it or even hold similar assets. This is also true for many other bond funds and indices. Comparing funds to arbitrary indices is not a good way to select funds, as the comparison is meaningless and can easily be gamed.

What To Do?
Of course, if the agg outperforms various other sectors over time and thus represents an opportunity cost, then why not just invest in the agg? I think that is a fine choice for very long-term investors. However, since the agg and other high-quality benchmarks’ outperformance is neither universal nor consistent AND (as we saw in the last post) credit has periods of outperformance and underperformance AND returns can be approximated ahead of time, I believe investors can do better. To apply Klarman’s framework, bonds are better “trading sardines” than “eating sardines.” Even a brief glance at the below comparison of the agg and high-yield index shows that credit returns swing like a pendulum. Buy when attractive, avoid when not.

source: Bloomberg

Several of the above charts above refer to the number of funds outperforming or underperforming, which ebbs and flows but generally more underperform. However, the data does not show how persistent outperformance or underperformance is for individual managers, nor does it show the degree of out/underperformance for individual managers. There are funds that have consistently beaten their stated benchmark and the agg over long periods of time, which means that some managers can outperform over multiple rolling periods with persistence. Even if the majority of active managers fail to beat their benchmark, there is a group that can beat their benchmark. I will not get into it here, but many of the funds that I have found have some common structural traits.

Lastly, in both order and importance, is that much of the data we have dates back to the early 1980s. Active managers outperform during periods of rising rates (see second chart) and rates have basically fallen for 35 years. This is not a prediction that the trend will reverse, but a recognition that passively managed funds have benefited from a tailwind that cannot be repeated to the same degree (rates only have 2-3% to fall until zero, not 15-20%).

The Bottom Line
So, what does this all mean?

  • Most fixed-income funds fail to beat their benchmark.
  • However, the benchmarks are often arbitrary and/or misspecified.
  • Fixed-income returns are both lumpy and predictable, so an active approach can add alpha relative to a benchmark.

One final thought is that the above all relates to relative performance. There is a lot of debate over whether to take an active or passive approach and performance metrics, but those are really secondary questions. Fixed-income is often used for an absolute return (ie. provide $x of income) or to simply diversify equity exposure. Investors should define the role of fixed-income in their portfolio before moving on to more complex and nuanced issues.