Keep Non-Core Bond Investments in Their Place

But the story isn't pure performance-chasing; there are also fundamental reasons to look at supporting-player bond offerings. High-quality US bonds are typically the most sensitive to changes in Treasury yields, making them more vulnerable than high-yield and other riskier bond types if interest rates were to rise over a sustained period. Balance sheets have generally been improving while bond defaults, though on the upswing recently, are still extremely low. Those factors have provided a tailwind for high-yield offerings. Meanwhile, as Morningstar's Bill Rocco noted in this article, the fiscal pictures of many emerging-markets nations have improved during the past decade, even as the US debt load has increased. More recently, so-called risk assets like junk bonds and emerging-markets debt have looked better from a valuation standpoint, according to some experts.

Higher Returns, But at What Price?
At the same time, the magnitude and velocity of inflows into risky bond types makes me worry that some investors could be overdoing these categories as a percentage of their fixed-income portfolios. That's a particularly big concern for retirees who are looking to their portfolios to fund near-term living expenses and might not have an adequately long time horizon to recover from outsized losses in what should be the safe portion of their portfolios.

The average world-bond and multisector-bond funds have standard deviations that are twice as high as that of the Barclays Aggregate Index (a benchmark for high-quality core bond funds) during the past decade, while high-yield bond funds' volatility has been nearly 3 times as high during that stretch. Investors in emerging-markets bonds, today's darlings, have had to put up with a standard deviation of returns that's more than 3 times as high as the Barclays Aggregate Index's.

The year 2008 provides an extreme depiction of the losses associated with risky bond types when investors flee to quality. The typical junk-bond fund lost 24% that year, while multi-sector offerings lost, on average, 15% of their value. Emerging-markets bonds also struggled, losing 18% in 2008. Meanwhile, the typical intermediate-term bond fund lost just 5% that year, and some intermediate funds, such as Barclays Aggregate index trackers and PIMCO Total Return, even gained.

During the past decade, adding non-core bonds to a portfolio anchored in high-quality stocks and bonds has definitely improved the portfolio's return potential but, not surprisingly, that return improvement has been accompanied by higher volatility and real losses. Those bigger losses owe to the fact that non-core categories such as high-yield bonds are more correlated to the equity market than are high-quality bonds, as I argued in this article. Even emerging-markets bonds, while not closely correlated with high-quality fixed-income investments, have a mild correlation with the US equity market. Thus, investors who pair non-core bond investments with their equity portfolios aren't getting the same diversification benefit that they would from a higher-quality bond portfolio.

For example, a portfolio with 50% in a total stock market index fund and another 50% in a total bond market index would have gained 5.5% on an annualised basis during the past decade; its worst three-month loss, in 2008, would have been negative 16.6%. Meanwhile, a portfolio with 50% in a total stock market index fund, 40% in a total bond market index fund, and 5% each in high-yield and emerging-markets bonds would have gained nearly 5.9% annualised during the past decade with a worst three-month loss of negative 18.9%. A portfolio with 50% in equities and 25% apiece in high-yield and emerging-markets bonds (no high-quality bond exposure) would have generated the best returns during the past decade--7.3% on an annualised basis--but only at the expense of much higher volatility; its worst three-month loss would have been negative 28%, also in 2008.

This (admittedly back-of-the-envelope) analysis demonstrates that adding non-core fixed-income asset classes has been fruitful during the past decade, but also that going overboard subjects a portfolio to the potential for outsized losses from which it could be difficult to recover without an adequately long time horizon. Of the portfolio mixes I outlined above, the one with 50% apiece in high-quality stock and bond funds has better returns during the past five years than any of the portfolio combinations that include high-yield and emerging-markets debt. That suggests that investors need a time horizon of more than five years, and preferably closer to a decade, if they're using non-core fixed-income assets in their portfolios; it also suggests that investors need to have the temperament to put up with inevitable bumpy patches that accompany non-core fixed-income types.

In addition, the risk-adjusted performance of the above portfolios demonstrates that non-core bond assets are best used as a small component of a fixed-income portfolio anchored in high-quality bonds, and that non-core position should step down as you get closer to needing your money. Even if retirees and soon-to-be retirees don't keep their bond portfolios ultraconservative, as author Bill Bernstein suggested in this video, it's still worth remembering that bonds are there for ballast much more than they are for return generation.