The jury is still out.
A version of this article was published in the June 2017 issue of Morningstar ETFInvestor. Download a complimentary copy of ETFInvestor here.
Go is a board game played on a 19-by-19 grid where each player makes one move per turn in an effort to capture territory on the board. A typical game lasts approximately 150 moves with an average of 250 potential variations per move, suggesting a level of game complexity of 10360, according to calculations from Victor Allis, a computer scientist. By contrast, a chess game's estimated complexity is 10123. In May, AlphaGo, the Google-made Go program, defeated the world's top-ranked Go player, Ke Jie. AlphaGo's victory adds another trophy to the shelf for artificial intelligence. Will algorithmic approaches to building bond portfolios get the best of flesh-and-blood bond managers, too?
In setting out to answer this question, I will evaluate the performance of a select group of strategic-beta fixed-income exchange-traded funds. I compare these funds with comparable peers selected from two groups. The first comparator comes from the cohort of traditional market-cap-weighted ETFs that is most broadly representative of the fund's Morningstar Category. Where possible, I've also compared these funds with a comparable actively managed Morningstar Medalist within the category. Thus, I am framing these hybrid strategies' performance in the context of their more vanilla passive and strictly active category peers.
Generally speaking, I've found that man is still besting machine in most categories. Though the strategic-beta ETFs examined here have outperformed their market-cap-weighted counterparts, sometimes by a wide margin, they haven't kept up with our favorite actively managed funds in four of the five categories I study here.
It is critical to note that the sample size here is small and the sample period is short. There were just seven strategic-beta fixed-income ETFs with at least three years of return history as of April 2017. Also, during the period, there were only three stress periods in the bond market: a commodity sell-off that affected emerging-markets debt (spanning September 2014 to December 2014), a sell-off in the energy space that affected high-yield bonds (June 2015 to February 2016), and the post-election rate rally (August 2016 to December 2016). So, given that this has been a relatively calm period, it is difficult to discern if these strategies will pass muster in a more protracted stress period.
The analysis that follows examines the performance of this sample of seven strategic-beta fixed-income funds against their closest market-cap-weighted ETF siblings and medalist active bond funds in their categories.
There are two strategic-beta ETFs that invest in inflation-linked bonds. Unlike their more vanilla ETF peers, they do not track market-cap-weighted indexes. Instead, they seek to maintain prescribed duration-risk profiles. They are FlexShares iBoxx 5-Year Target Duration TIPS TDTF and FlexShares iBoxx 3-Year Target Duration TIPS TDTT. They seek to maintain a target duration of five years and three years, respectively. Each charges a fee of 0.20%. Both funds produced virtually identical returns against their market-cap-weighted cousins over the three-year period ended April 30, 2017: PIMCO 1-5 Year U.S. TIPS ETF STPZ and SPDR Bloomberg Barclays 1-10 Year TIPS ETF TIPX. The similarities in these funds' performance profiles are hardly surprising as these funds invest in just one kind of bond, U.S. Treasury Inflation-Protected Securities.
As far as duration is concerned, the FlexShares funds have performed as advertised, staying close to their respective duration targets. The duration of market-cap-weighted products moved in a fairly narrow range as well. STPZ's duration ranged between 1.9 years and 3.0 years, and TIPX's fluctuated between 3.1 years and 5.1 years during the same period.