In Monday’s post, I discussed the growing popularity of high-yield bonds: four Canadian ETFs covering this asset class have appeared in the last 12 months. I explained how investment gurus Larry Swedroe and David Swensen both advise investors to avoid high-yield bonds and instead stick to safer fixed-income investments such as government bonds.

Not all portfolio managers agree with that assessment, however. Today we’ll look at two well-known index investors who believe that high-yield corporate bonds can play role in a diversified portfolio.

Richard Ferri’s All About Asset Allocation, now in its second edition, is one of my favourite books on this important subject. Ferri argues that high-yield bonds aren’t as highly correlated with equities as others believe. “Some market researchers suggest that default risk is nothing more than a type of equity risk,” Ferri writes, “and therefore adding high-yield corporate bonds to a portfolio is the equivalent of adding more equity. That argument is not entirely correct.”

Ferri explains that since the early 1980s, there have been periods where the default risk of high-yield bonds was highly correlated with equities, but also periods where the correlation was zero, or even negative. “Only about 25 percent of the default risk in high-yield bonds can be attributed to the same factors as equity returns,” Ferri estimates, although the correlation is higher when you look at the really junky bonds, rated CCC or lower. He believes investors can get a diversification benefit from adding high-yield bonds to a portfolio. For the average mid-life investor, about one quarter of the overall fixed-income component would be about right: for example, in a portfolio that is 40% bonds, you might include a 10% allocation to high-yield corporates.

Alexander Green’s Gone Fishin’ Portfolio (see my review) also includes an allocation to high-yield bonds. “Despite the pejorative name ‘junk,’ these bonds offer a number of advantages,” Green writes. “A diversified portfolio of these bonds, even after accounting for defaults, has returned more than either Treasuries or high-grade corporates. And while they do tend to be more highly correlated with the stock market than other bonds, they do not move in lockstep with equities, giving you some diversification advantage.”

Green’s model portfolio allots 30% to fixed income: 10% to investment-grade corporates, 10% to inflation-protected bonds, and 10% to high-yield. He notes that these three types of bonds all behave differently. “High-yield bonds have a fairly low correlation with investment-grade bonds, like Treasuries and AAA corporates. That’s what we want when we asset allocate,” he argues. “Blending high-yield bonds with the other asset categories actually reduces your overall portfolio volatility while increasing your returns… In short, including a 10% allocation to high-yield bonds gives you an edge historically: portfolios that include exposure to this asset class have performed better than those that rely solely on investment-grade bonds for the fixed income allocation.”

So who’s right? Personally, I don’t use high-yield bonds in my portfolio: even if they do provide a diversification benefit as Ferri and Green argue, I’m partial to Swedroe and Swensen’s view that the fixed-income side of a portfolio should be a safety net, not a place to reach for maximum yield. That said, a modest allocation of 10% to junk bonds, as Ferri and Green recommend, seems perfectly reasonable for investors who understand the risks. Any more than that, however, is playing with fire.