Factor Investing in Fixed Income
Factor investing has become a cornerstone of evidence-based portfolio construction, particularly in equities. Investors tilt their portfolios towards known drivers of return such as value, size, and profitability, seeking to harvest systematic risk premia rather than rely on discretionary selection. Yet much less attention is given to fixed income factor investing, despite decades of research showing that bond returns are also shaped by clear, persistent sources of risk and return. Understanding these fixed income factors—and how they differ from traditional index exposures—can materially improve both the efficiency and the robustness of a portfolio.
This post explores the theory, evidence, and implementation of factor investing in fixed interest, focusing on the two primary drivers: term and credit. It also examines how leading firms such as Dimensional Fund Advisors apply these ideas in practice, and why investors should consider moving beyond conventional bond benchmarks.
The Foundations: Term and Credit as Fixed Income Factors
The academic roots of fixed income factor investing go back to the seminal work of Fama and French (1993), who identified two distinct risk premia in the cross-section of bond returns. These are:
Term premium: The additional return investors expect to receive for holding longer-term government bonds rather than rolling short-term Treasury bills. This premium compensates for interest rate uncertainty and inflation risk over extended horizons.
Credit premium: The excess return associated with holding corporate bonds over duration-matched government bonds. This reflects the risk of default, downgrade, and spread volatility linked to broader economic conditions.
A common misconception is that returns from the term and credit factors depend on yields increasing or remaining high. In reality, factor premia are about the compensation investors receive for bearing certain risks, not about directional interest rate or yield movements. These premia can be realised in several ways—even when overall yields decline.
Take the credit premium. A portfolio designed to capture this factor might go long corporate bonds and short duration-matched government bonds. This long–short structure neutralises general interest rate exposure, isolating the performance attributable to credit spread changes and the higher yields embedded in corporate bonds. The portfolio earns a positive capital return if spreads tighten, as corporate bond prices rise relative to Treasuries. But it can also earn a return if spreads remain wide—because the investor continues to collect the excess yield (carry) offered by corporates over risk-free bonds. In other words, so long as defaults remain low and spreads do not widen to cause significant losses to capital via price decreases, the credit premium accrues steadily over time, even in a falling yield environment.
The term premium functions in a similar way. Investors holding longer-duration government bonds instead of rolling over short-term bills are compensated for taking on inflation and interest rate uncertainty. This compensation is typically embedded in the yield curve: when the curve is upwards-sloping, the yield on a longer duration bond is higher than that of a shorter duration bond. Even if yields decline across the curve, the investor still benefits from the so-called roll-down effect. That is, as a bond approaches maturity, its yield typically declines (moving down the curve), and its price rises accordingly. This price appreciation enhances the total return, independent of the direction of overall rates.
Moreover, the convexity of longer-duration bonds contributes to their performance. Convexity means that longer duration bonds gain more in price when yields fall than they lose when yields rise by the same amount. This asymmetric payoff, combined with an upward-sloping yield curve, provides another path through which the term premium can be realised—particularly in volatile or declining rate environments.
Example: How Convexity Enhances the Return Profile of Long Bonds
To understand why longer-duration bonds benefit disproportionately when yields fall, we need to look at how bond prices respond to changes in interest rates.
The approximate percentage change in a bond’s price in response to a small change in yield can be estimated using both duration and convexity, as follows:
Price change ≈ –Duration × Yield change + ½ × Convexity × (Yield change)²
Where:
Duration measures the first-order sensitivity of the bond’s price to interest rate movements.
Convexity measures how that sensitivity itself changes as yields move.
Yield change is expressed in decimal form (e.g. 0.01 for a 1% change).
Let’s say you hold a 20-year government bond with a modified duration of 12 and a convexity of 160 (typical for a long-dated bond). Here’s what happens when yields move up or down by 1 percentage point:
If yields fall by 1%:
Price change ≈ –12 × (–0.01) + 0.5 × 160 × (0.01)²
≈ 0.12 + 0.008 = +12.8%
If yields rise by 1%:
Price change ≈ –12 × (0.01) + 0.5 × 160 × (0.01)²
≈ –0.12 + 0.008 = –11.2%
Even though the change in yield is symmetrical, the price changes are not. The bond gains more in value when yields fall than it loses when yields rise. This is the result of positive convexity, and it’s more pronounced in long-duration bonds because both duration and convexity are higher at longer maturities.
This asymmetry helps explain why long bonds can deliver strong returns during falling-rate environments. It’s not just about yield levels—it’s about how bond prices behave non-linearly in response to rate changes.
The Bottom Line
These examples highlight a key point: factor investing in fixed income is about relative compensation, not absolute yield levels. Term and credit premia represent the extra returns investors earn for bearing duration and default risk, respectively. These premia can accrue through spread compression, carry, roll-down, or convexity effects, regardless of whether interest rates are rising, falling, or flat.
What the Data Say: The Persistence of Bond Factor Premia
A large body of academic research confirms the existence and persistence of fixed income factor premia:
Term premium: Fama and Bliss (1987), Cochrane and Piazzesi (2005), and Duffee (2011) show that the slope of the yield curve predicts excess returns on long duration bonds, suggesting the presence of a positive term premium over time. This premium varies with macroeconomic conditions, risk aversion, and supply-demand dynamics (Greenwood and Vayanos 2014).
Credit premium: The excess return on corporate bonds is not fully explained by expected defaults. Studies such as Elton et al. (2001) and DeJong and Driessen (2012) find that a large portion of the spread is attributable to a true credit risk premium—investors are compensated not just for default losses but for bearing systematic credit exposure during economic downturns.
Dimensional Fund Advisors, AQR, BlackRock, and others have used this body of evidence to guide fixed income strategy design. By tilting portfolios towards term and credit exposures in a controlled, systematic way, they seek to improve expected returns whilst retaining the defensive characteristics that make bonds so useful in multi-asset portfolios.
Dimensional’s Approach to Fixed Income
Dimensional Fund Advisors (DFA) has long applied academic research to portfolio construction, and their fixed income strategies are no exception. Unlike traditional bond funds that simply track market-value-weighted indices, Dimensional designs portfolios to capture compensated risks, primarily through exposure to the term and credit factors.
Their implementation begins with a broad, investment-grade universe, excluding securities with excessive default risk or limited liquidity. Within this universe, Dimensional applies a set of systematic tilts based on current market conditions. When term spreads (e.g. the difference between 10-year and 2-year Treasury yields) are historically wide, Dimensional increases exposure to longer-duration bonds to capture the roll-down return. Conversely, when term spreads are flat or inverted, they reduce term exposure and shift towards shorter maturities, improving expected risk-adjusted returns.
Similarly, Dimensional dynamically tilts credit exposure based on current credit spreads. When spreads widen—often in periods of elevated market stress—they increase exposure to corporate bonds, provided that expected credit premiums outweigh estimated default risk. This results in time-varying exposure to credit risk, but always within a tightly controlled risk framework. The process is not about forecasting interest rate moves or timing the market. Rather, it is rooted in the view that forward-looking expected returns vary over time and can be estimated using observable variables like yield spreads.
Dimensional’s process is also informed by rigorous simulation and empirical testing. They carefully control duration, credit quality, and sector diversification, ensuring that the portfolio's overall risk characteristics remain consistent with the investor's objectives. They often favour short-to-intermediate-term portfolios, reducing sensitivity to unexpected interest rate changes whilst still capturing meaningful credit premiums.
Notably, Dimensional’s bond funds are not designed to outperform over every short-term window, but to deliver higher expected returns over time by capturing known risk premia in a disciplined, cost-effective way. Their success lies in avoiding uncompensated risks (such as concentrated issuer exposure or macroeconomic bets) whilst leaning into sources of return that are supported by decades of data and sound economic theory.
The Benchmark Problem: What Traditional Indices Miss
Most investors benchmark their bond exposure to broad indices such as the Bloomberg Global Aggregate Bond Index. These indices are constructed using market capitalisation weights, meaning that the largest borrowers—typically governments and highly rated corporates—receive the greatest representation. As a result, portfolios tracking these indices tend to be heavily tilted towards government and agency bonds, with relatively limited exposure to corporate credit or longer-duration instruments.
More importantly, the mechanics of market cap weighting can inadvertently lead investors to increase their exposure to bonds after prices have risen and expected returns have fallen. This dynamic works against the goal of systematically harvesting factor premia.
Consider the credit premium. When corporate bond spreads narrow, prices rise and yields fall. A market cap-weighted bond index will, by definition, allocate more to those bonds whose prices have increased the most—typically those with tighter spreads and lower prospective credit returns. Investors are thus increasing their weight to credit risk after the premium has diminished. Conversely, when credit spreads are wide—precisely when expected returns are higher—a market cap-weighted approach underweights these opportunities because prices are lower.
A similar pattern occurs with the term premium. When the yield curve flattens (for example, as long-term yields fall closer to short-term rates), the prices of long-duration bonds increase. This leads a market cap-weighted index to allocate more capital to those longer bonds at a time when the term premium is lower or even negative. Again, the investor ends up increasing exposure when the expected reward is least attractive.
In this way, market capitalisation weighting mechanically leads investors to allocate more capital to bonds with lower expected returns. The exposure to term and credit risks that such an index provides is not optimised for return; it simply reflects the outstanding debt in the marketplace, which may be influenced more by issuance patterns, regulatory considerations, or government borrowing needs than by efficient pricing of risk.
By contrast, a factor-based approach explicitly targets the underlying drivers of return. A portfolio that underweights government bonds and increases exposure to high-quality credit when spreads are wide can harvest the credit premium more effectively. Likewise, shifting modestly along the yield curve to capture the term premium when it is attractive allows investors to benefit from upwards-sloping yield curves or roll-down effects, without assuming undue interest rate risk.
This is not about taking more risk for its own sake. It is about taking compensated risk—leaning into areas of the bond market that offer higher expected returns based on sound economic reasoning and long-run empirical evidence. Factor-based portfolios can be designed to maintain similar volatility and drawdown characteristics to traditional indices, but with improved return expectations over time.
Why It Matters
In a world where bond yields are no longer consistently generous, investors cannot afford to leave returns on the table. Capturing the term and credit premia embedded in fixed income markets offers a structured way to improve outcomes, particularly when implemented with discipline, transparency, and low costs.
Empirical evidence and Dimensional’s own research suggest that the term premium—measured as the excess return of intermediate over short-term government bonds—has averaged around 9 basis points per month, or approximately 1.1% per year, rising to 4% annually when the term spread is unusually wide. Similarly, the credit premium—captured by holding investment-grade corporate bonds over Treasuries—has averaged around 0.84% annually, increasing to 3% per year when credit spreads are in the highest quartile (DFA 2024a, 2024b). These premia are time-varying but economically rational, and they are realised not just through spread compression but also via carry, roll-down, and convexity effects.
Factor investing in bonds is not a silver bullet, and the premia are not guaranteed. They are volatile and can turn negative over short periods. But over time, they represent the rational compensation investors earn for taking on identifiable, systematic risks.
The key is to avoid treating fixed income as a monolith. Just as equity investors have learned to distinguish between growth and value, small and large, profitable and unprofitable, bond investors can benefit from understanding the dimensions of term and credit, and from allocating accordingly.
References
Adrian, Tobias, Richard K. Crump, and Emanuel Moench. 2013. ‘Pricing the Term Structure with Linear Regressions’. Journal of Financial Economics 110 (1): 110–138.
Campbell, John Y., and Robert J. Shiller. 1991. ‘Yield Spreads and Interest Rate Movements: A Bird’s Eye View’. Review of Economic Studies 58 (3): 495–514.
Chen, Longstaff, Lesmond, and Wei. 2007. ‘Corporate Yield Spreads and Bond Liquidity’. Journal of Finance 62 (1): 119–149.
Cochrane, John H., and Monika Piazzesi. 2005. ‘Bond Risk Premia’. American Economic Review 95 (1): 138–160.
DeJong, Frank, and Joost Driessen. 2012. ‘Liquidity Risk Premia in Corporate Bond Markets’. Quarterly Journal of Finance 2 (1): 1250003.
Dimensional Fund Advisors. 2024. Fixed Income Investing: A Factor-Based Approach.
Dimensional Fund Advisors (DFA). 2024a. ‘Higher Expected Returns with Systematic Fixed Income’. Dimensional.com. https://www.dimensional.com/us-en/insights/higher-expected-returns-with-systematic-fixed-income
Dimensional Fund Advisors (DFA). 2024b. ‘Demystifying Systematic Fixed Income Investing’. Dimensional.com. https://www.dimensional.com/gb-en/insights/demystifying-systematic-fixed-income-investing
Duffee, Gregory R. 2011. ‘Information in (and not in) the Term Structure’. Review of Financial Studies 24 (9): 2895–2934.
Elton, Edwin J., Martin J. Gruber, Deepak Agrawal, and Christopher Mann. 2001. ‘Explaining the Rate Spread on Corporate Bonds’. Journal of Finance 56 (1): 247–277.
Fama, Eugene F., and Robert R. Bliss. 1987. ‘The Information in Long-Maturity Forward Rates’. American Economic Review 77 (4): 680–692.
Fama, Eugene F., and Kenneth R. French. 1993. ‘Common Risk Factors in the Returns on Stocks and Bonds’. Journal of Financial Economics 33 (1): 3–56.
Greenwood, Robin, and Dimitri Vayanos. 2014. ‘Bond Supply and Excess Bond Returns’. Review of Financial Studies 27 (3): 663–713.
van Binsbergen, Jules H., Itamar Goldstein, and Yang Lu. 2019. ‘The Credit Risk Premium’. Journal of Finance 74 (2): 527–573.