Factors Matter in Fixed-Income Investing

A new paper from asset manager Robeco aims to dispel myths that equity factors don’t apply to fixed income.

Illustration by II

Illustration by II

Think factor investing isn’t really a thing with bond funds? Think again.

The traditional equity factors of value, low volatility, and momentum are also driving returns in fixed income securities, despite critics’ concerns that these factors don’t translate to credit, according to new research from asset manager Robeco. The firm looked specifically at government bond portfolios and found that using factors such as value and momentum to pick these securities provided “consistently higher” returns over traditional bond market benchmarks, Robeco said in a white paper detailing its research.

What’s more, the low volatility factor both outperformed and reduced risk, according to the research. The authors of the study argue that investors can benefit from these premiums with a multi-factor government bond portfolio.

Robeco evaluated well known factors such as value (buying assets that are trading at a discount to their underlying value), momentum (buying securities that have performed well in the recent past and avoiding recent losers), and low-risk (lower risk assets generating higher risk-adjusted returns). To do its research, Robeco looked to pick government bonds based on common factors like low-risk, but it incorporated traditional ways of evaluating fixed-income securities. Robeco further refined the factors for the quirks of the government bond markets, such as for liquidity and transaction costs that have a big impact on investors’ portfolios.

It found in the study that all three factors generated higher risk-adjusted returns than the market.

[II Deep Dive: Is Active Bond Investing...Just Easier?]

With the low-risk factor, Robeco compared short-dated government bonds to those that matured farther into the future. Robeco also added a valuation measure into its assessment of low-risk securities to make sure that the rules it used to favor short-duration bonds over longer-dated ones did not include overvalued assets.

The authors — Olaf Penninga, Martin Martens, and Casper Zomerdijk — found that longer maturity bonds offered a better yield than shorter-dated bonds. However, “the value advantage of longer-dated bonds is generally insufficient to compensate for the higher risk,” the authors wrote. “By observing this discrepancy between additional yield pick-up and additional risk, investors could already have expected better risk-adjusted returns for shorter-dated bonds.”

When it comes to value, Robeco went further than existing studies, which generally measure the valuation of the bond market of one country or look at the bonds of only one maturity. Robeco took a more detailed approach, assigning a value score to each maturity, for example.

Although there is growing interest from both institutions and asset managers in using factor-based credit and government bond investments, the area is still in its infancy. That’s because academics haven’t produced the range of comparable research on fixed income factors that they have published on equity factors over the past few decades.

“One of the reasons why credit is more difficult for academic researchers and for our peers is they didn’t have access to data,” said Patrick Houweling, lead portfolio manager and researcher of quant credits at Robeco, who was not involved with the new study, in an interview. “Look at academic journals; they have been dominated by research on stocks. But [fixed income] is coming now.”

It’s all about behavior, which doesn’t change based on the asset class, he said.

“The starting point is how people behave in financial markets,” said Houweling. “Even though there is more data in equity markets, it doesn’t mean these patterns only exist in equities. The well-known factors may not use the exact metrics as equities, of course, but they capture the concepts.”

Olaf Penninga Martin Martens Patrick Houweling Robeco Casper Zomerdijk