I have long been skeptical of how fair bond fund prices are — or more accurately said, the potential ability for knowledgeable investors to “game” bond fund prices — in fixed income asset classes where liquidity isn’t great. Two asset classes that immediately come to mind are municipal bonds and high-yield corporate bonds. I finally got around to testing this proposition using daily returns data for a handful of different bond funds in these two asset classes. The findings confirm my suspicion.
If bond fund prices (or, equivalently, net asset values) were fair on a day-to-day basis, you should generally see that a prior day’s return for a fund tells you nothing about what today’s return will be. In other words, past returns shouldn’t predict future returns. The good news is that this is a testable proposition.
I pulled daily returns data for five different large bond funds from two different fund families. The first three funds have the majority of their portfolios invested in less liquid securities like municipal bonds and high-yield corporate bonds. The last two tend to own fixed income securities with better trading liquidity.
If each fund’s prices are fair, then we should see that historical daily returns are uncorrelated with future daily returns (i.e., the past doesn’t predict the future). If a given fund’s prices are “stale,” then we should see that historical daily returns are correlated with future returns (and, most likely, positively correlated with future returns). In this world, an investor could look at prior days' returns to determine whether he or she should buy or sell the fund today, effectively gaming the system.
Are Prior Day’s Returns Highly Correlated?
Anywhere you see a “Yes” in the table indicates that historical returns were highly correlated with future returns, and vice versa for where you see a “No.” As I suspected, we see that the three funds that focus on either municipals or high-yield corporate bonds appear to have stale fund prices. It’s good to see, though, that the two funds I analyzed that focus on more liquid securities appear to have fair prices.
For the first three bond funds, the statistically significant correlations were positive, which means that if yesterday’s returns were positive for the fund, it’s more likely that today’s will be as well. Of course, the opposite is true if yesterday’s returns were negative.
What Should Investors Do?
The primary takeaway is to be aware of this if you’re investing in bond funds (and exchange-traded funds, as well) that own relatively illiquid securities. It’s possible other investors could be taking advantage of this knowledge if the fund is not charging redemption fees to counteract this issue (or isn’t restricted to approved advisors/investors in a manner similar to what DFA does). This is also, in my judgment, just one more reason why low-credit-quality securities, which tend to have poor liquidity, do not make much sense for individual investor portfolios. On the one hand, you need the diversification that a fund provides, but on the other, the fund’s prices may be able to be gamed.