If you need yield, avoiding conventional junk-bond indexes is a wise move right now.
I dislike most junk-bond exchange-traded funds. With their coupons counted as ordinary income, they're the tax man's best friend. And their underlying holdings are much more illiquid than the typical ETF's, creating the potential for big premiums/discounts and heinous trading costs, much of it hidden to investors. But if you held a gun to my head and forced me to pick the best junk-bond ETF, I'd pick PIMCO 0-5 Year High Yield Corporate Bond Index ETF HYS. In truth, I rather like this fund.
Let's start with an historical observation. As Exhibit 1 indicates, short-duration junk bonds have done better than the broad junk-bond market, with lower volatility. The outperformance is even better after you adjust for the differences in duration, or sensitivity to interest rates. Decades of falling interest rates put the wind behind the longer-duration index. In fact, the junk-bond market only beat duration-matched Treasuries by about 1.7% annualized; short junk bonds beat duration-matched Treasuries by 4.3%.
- source: Morningstar Analysts
I wouldn't read too much into this single set of data. Asset classes do well over extended periods of time for one-off reasons that won't necessarily hold in the future. However, the historical risk-adjusted outperformance of short-maturity junk bonds looks like it has structural origins and is therefore likely to continue.
You can see why in today's yield spreads. Exhibit 2 shows the options-adjusted spreads, or OAS, of junk bonds by maturity. The OAS is a measure of a bond's yield advantage over a similar-duration Treasury, and it accounts for the options the bond issuer has at its disposal. Strangely enough, the further out you go, the lower the spread, bottoming at the 9-10 year tranche and then ticking up in long-maturity tranches. In the corporate-bond market, you see the logical relationship of credit spread rising with maturity. The dip in the 9-10 year tranche can't be explained by varying credit quality; you see the same relationship in the yield curve of junk bonds with the same credit ratings, becoming sharper as you go down the credit-quality spectrum.
- source: Morningstar Analysts
I suspect what's happening is that junk-bond fund managers and individual investors are stretching for yield, especially in new issues, pushing down yields in the 9-10-year tranche. It's gotten to the point where sub-five-year bonds have yields competitive with longer-duration bonds, and the junkiest of the junk bonds are trading at somewhat absurd valuations. As of this writing, the BofA Merrill Lynch U.S. High Yield Master II Index's effective yield is 6.6%; the BofA Merrill Lynch 0-5 Year U.S. High Yield Constrained Index's is 7%.
There are few if any offsetting advantages to the broad market over the short-maturity tranche. Both have average credit quality of B1. But not all credit quality is created equal--seasoned bonds, making up much of the shorter-maturity bonds, have done better than equivalently rated new issues. Additionally, short-maturity bonds have lower duration, so rising interest rates will barely hurt this fund.
This is all useless information if junk bonds are so overvalued that it doesn't matter that shorter-maturity bonds are merely less overvalued. The bears cite record-low absolute junk-bond yields, ignoring the fact that the broad market's options-adjusted spread is still about 6%, an average valuation relative to history. Looking at absolute yield rather than spread is justifiable if you're also willing to make the strong judgment that interest rates are too low and will unexpectedly go up over your time horizon. That's not a game I'm willing (or able) to play with confidence. Besides, even if interest rates do rise, being in short maturities means I won't be hurt much.