I find the phenomenon of momentum to be fascinating. My introduction to personal finance came first in the form of broad market ETFs, but I’ve spent years now thinking of factor tilts. While many walk away from the space saying “just buy small cap value”, “this is all just data mined” or “the world is different today”, I find the nuances within the space to be fascinating.
A fair amount of the literature on these tilts stem from a handful of papers such as Fama and French’s 3 factor model in the 90s, the work of scholars like Asness in momentum (or Jagadeesh and Titman before him), the low beta work of Frazzini and Pedersen, and so on. A lot of the attention naturally goes to the equity side of things where returns are greater. However, AQR and Dimensional both have bond funds in which they claim to have higher expected returns. Rather than reading just papers from these firms, I’ve decided to take a look at the existing literature to get a feel for the systematic bond space.
An interesting paper I came across in this sub-discipline is “Momentum in Corporate Bond Returns” by Jostova et al, initially drafted in 2008.
Some highlights:
- Portfolios are sorted into decile portfolios based on returns from formation to t + 6 months.
- The asset pricing model used to measure momentum alpha controls for Fama-French SMB and HML factors, credit spreads, term spreads and market beta.
- They find that the portfolio alphas between P10, the top decile momentum portfolio, and P1, was 70-80bps.
- Perhaps the most interesting finding, for me anyways, was that they find the results to be statistically insignificant for IG bonds. In other words, there’s a relationship between credit risk and the significance of the momentum factor. They found insignificant alpha when considering just the IG segment of their portfolios.
This only adds to my intrigue with the momentum factor. I can’t think of a theoretical explanation, in terms of discount rates, that would square the lack of momentum in IG corporates. I fall in the camp of folks who think that value stocks may have higher discount rates applied to the same cash flows given the economic characteristics of firms, or I at least find that explanation plausible. Momentum is a head scratcher but when you see it empirically across asset classes and geographies, you eventually become a believer regardless of theory. Sure, behavioural theory does exist, but I also find it intuitively odd to believe that people are being irrationally exuberant or pessimistic about the returns of….non investment grade illiquid corporates and are just exceedingly rational when it comes to IG. Who knows.
Another part of the difficulty in understanding factor premia in this space is that there is a lack of liquidity. In their Journal of Portfolio Management article, for example, Alquist, Israel and Moskowitz document the debate over whether there is a small cap premium or if this is just the illiquidity premium in disguise (“Fact, Fiction and the Size Effect). The authors document a similar issue here– corporate bonds are fairly illiquid in nature.
DISCLAIMER: I am not a financial advisor. Nothing on this website is to be constituted as financial advice. All content here is solely for educational purposes. They solely represent the opinion of the author.
Bells and Tails 