Controlling Risk in Systematic Active Fixed Income Portfolios
For investors in the Systematic Active Fixed Income space, alongside producing systematic alpha, one of the most important objectives is controlling risk. Find out more about the types of systematic portfolio risk and how to navigate them.
Systematic investing takes a disciplined, diversified, and scientific approach to risk, based on the objective application of quantitative models informed by historical data. This form of investing has grown in popularity over the past two decades, particularly for equity portfolios. But it’s been slower to take hold among fixed income portfolio managers, who have long taken a quantitative approach to risk, but a more qualitative fundamental approach to many other aspects of the management process. Essentially, this traditional approach to credit portfolio management uses quantitative models to manage beta, but relies on fundamental analysis to generate alpha.
Many factors have contributed to the slower adoption in credit portfolios relative to equity. Compared to equities, credit markets are more complex, less transparent, and less liquid, complicating the implementation of systematic strategies. However, recent developments have reduced these frictions, paving the way for credit portfolios to reap the benefits of systematic investing.
Mitigating Portfolio Risk
Effective risk management is one of the cornerstones of a successful systematic strategy – specifically, concentrating most of the portfolio risk in intentional alpha-producing factor exposures. Unintentional risk exposure should be avoided to the extent possible, and even intended exposures should be managed to prevent any one risk exposure from dominating portfolio risk.
Unlike idiosyncratic risk, which is due to issuer-specific effects, systematic portfolio risk is the return volatility that can be traced to the movement of modeled risk factors. These risks could include changes in interest rates, exposure to currency fluctuations in multi-currency portfolios, and changes in industry sector spreads.
Interest rate risk can be measured crudely by portfolio duration, which gives the sensitivity to a parallel shift in the yield curve. However, given the primary importance of the yield curve in determining bond prices, it is important to model non-parallel changes to the yield curve as well.
Different approaches can be adopted to model such risk. For example, Litterman and Scheinkman1 showed that three factors — yield curve shift, twist, and curvature — can account for well over 90% of variation in yield curve returns. Instead, we prefer the key rate duration2 approach, which models exposures to yield changes at a number of key maturities along the curve. Although from a mathematical perspective, it is preferable to use a small number of orthogonal risk factors, many investors appreciate models that correspond intuitively to the way they view the market, even at the expense of some redundancy and correlation among the factors.
Credit sector spread risk also has several possible approaches. A portfolio’s spread duration gives its sensitivity to a parallel widening or tightening of spreads, but again this single number fails to capture all the possible ways in which spreads can change. They can widen for issuers in one industry while tightening in another. Therefore, many practitioners partition their portfolios and measure exposures to spread changes by industry as distinct, albeit correlated, risk factors.
Furthermore, even within a given industry, a parallel shift in spreads is not the most typical type of systematic change. Spreads often follow a pattern of relative changes in which bonds with wider spreads widen (or tighten) more, proportionally to their initial spreads (as shown by Ben Dor et al.,3). Sensitivity to this type of systematic spread change can be measured by contributions to ‘duration times spread’ (DTS), which measures portfolio sensitivity to relative spread changes across a market segment. Constraints on active DTS exposures by industry thus serve to limit tracking errors due to systematic changes in corporate bond spreads.
For benchmarked portfolios, a comprehensive measure of risk is tracking error volatility (TEV), which looks at how volatile the return difference is between the portfolio and the benchmark. Systematic TEV can be modeled based on the differences between portfolio and benchmark exposures to risk factors, using estimates of factor volatilities and correlations from historical data It can be effectively controlled by placing constraints on how much deviation between portfolio and benchmark risk factor exposures is allowed. Similarly, placing constraints on exposures to individual bonds and issuers can limit the TEV resulting from other idiosyncratic and default risks.
The Bottom Line
If close controls are imposed on all of these parameters, portfolios should be expected to closely track the returns of the benchmark. What’s more, it follows that any attempt to improve portfolio performance relative to the benchmark – by taking different exposures to systematic risk factors and/or issuers – increases the risk of underperformance. That’s why there are different types of portfolios available for investors with different risk appetites.
More About SAFI
The data-driven insights in our Systematic Fixed Income strategies are informed by systematic signals delivered in the form of indices developed by the Barclays Quantitative Portfolio Strategy team, or QPS, which is well-recognized as an innovator in quantitative fixed income research. Their innovative signals and portfolio optimization methodologies form an important input to the process we, at State Street Global Advisors, use in the implementation and management of Systematic Active Fixed Income strategies.
Explore the State Street Global Advisors thought leadership series on Systematic Active Fixed Income investing to learn more about this innovative investment approach and its benefits.