A group of rectangular objects
INVESTMENT ACTIONS

Rethinking Fixed Income Asset Allocation: A Systematic Approach

Explore ways to unlock a more dynamic allocation framework which can offer versatility for both negative and positive stock-bond correlation regimes.

Overview

The post-pandemic shifts in U.S. interest rates and inflation have ushered in a regime largely unfamiliar to most investors. This new reality means both discretionary and systematic investors need to revisit asset allocation frameworks that have been built and relied upon in a low interest rate, low inflation environment.

The role of the stock-bond correlation (“SBC”)

Changes in the U.S. macroeconomy post-pandemic have ushered in a regime largely unfamiliar to most investors. For over twenty years, price of stocks and bonds typically moved in opposite directions making bonds a reliable diversifier in portfolios. This inverse relationship began to change in 2021 with stock-bond correlation (“SBC”) turning positive. What followed in 2022 was the fastest and most aggressive monetary tightening since the Volcker shock resulting in rare simultaneous downturn in stocks and bonds leading to historically poor performance (Figure 1).

Figure 1: Annual U.S. equity and bond returns

The return of bonds are based on the annual return of 10-year U.S. Treasury Bond. Stocks are represented by the total return of S&P 500 Index from 1957 onwards. Prior to 1957, the returns are based on price changes in S&P Composite Index. Source: BlackRock, with data from LSEG Datastream as of May 31, 2024. The figures shown relate to past performance. Past performance is not a reliable indicator of current or future results. Index performance returns do not reflect any management fees, transaction costs or expenses. Indices are unmanaged and one cannot invest directly in an index.

The outlook for consumption, and by extension the broader economy, is now negatively related to increases in inflation, leading to a change in the relationship between stocks and bonds. 1 Based on our analysis of bond and equity returns, we believe the post-pandemic era highlights a return to a negative consumption/inflation relationship contributing to positive stock-bond correlations – a stark departure from the negative correlations of the previous two decades. 2

Moving forward, the crucial insight for portfolio management is that a dependable negative stock-bond correlation, which has historically allowed bonds to diversify portfolios effectively, may be unlikely to reemerge without a resurgence of a positive link between consumption and inflation.

Traditional asset allocation approaches

In a systematic framework, “tilting” and “timing” correspond to strategic and tactical asset allocation. “Tilting” is slow-moving and seeks to capture long-term risk premia. It generally contributes positively to portfolios during low volatility periods and expansion phases of the economic cycle. “Timing” is fast and complements tilt by seeking to generate positive returns in periods of higher volatility or economic transitions.

Combining tilt with timing in asset allocation can result in portfolios with a more consistent return profile. Both strategies are crucial for navigating the full spectrum of economic cycles and market volatility, corresponding to both transitions into and out of economic recessions and/or financial crises. Timing can offset periods of underperformance when tilts are struggling. Conversely, in periods where volatility is muted, and risk premia are stable, tilts can add value.

A systematic approach to asset allocation in positive SBC regimes

Asset allocation frameworks, that have been developed and tested over the last two decades, a period defined by negative SBC, must adapt to the current positive but unstable SBC regime. Grappling with the task of developing insights for a relatively unfamiliar positive SBC regime, investors can look back to the early 60s to include periods of high inflation and positive SBC in research. This approach may enable the testing and validation of new insights that are robust and can navigate risk on (off) and risk parity on (off) regimes. We define a “Risk On/ Risk Off” (R2) factor and a “Risk Parity On/ Risk Parity Off” (P2) factor.

In negative SBC regimes, investors can seek to capture the maximal return of R2 factor through a long position in risky assets and a corresponding short position in safe assets with asset weights set such that each asset contributes the same amount of risk. The direction of these exposures means the factor exhibits the maximum gain (loss) during risk on (off) periods.

In positive SBC regimes, investors can seek to capture the maximal return of P2 factor through a long position in risky assets and a corresponding long position in safe assets with asset weights set such that each asset contributes the same amount of risk. The direction of these exposures means the factor exhibits the maximum gain (loss) during risk parity on (off) periods. 

By expanding our research to cover data from the 1960s and tapping into sophisticated machine learning techniques to analyze complex data, we’ve developed a framework that captures SBC regime shifts without directly estimating the SBC parameter, aligning with the risk on/off and risk parity on/off factors.

Back-testing performance from July 31, 2006–March 31, 2024, we examine the average performance of timing R2 and P2 factor across various regimes by constructing a two-asset portfolio made up of 5 Year High Yield CDX as the risky asset and the 10-year US treasury Note Future as the safe asset. Below is the performance of each factor across return quintiles for a buy and hold R2 and P2 portfolio. On average the returns are positive across all quintiles and most notably concentrated at the tails of return distribution. This clustering of performance in the tails indicates a defensive return profile and outperformance during periods of extreme market conditions.

Figure 2: Portfolio timing returns in R2 and P2 factor return quintiles

For tilt, we use a blend trend-following and carry/vol based approach, balancing the defensiveness of trend following with risk seeking and aiming for higher return profile of carry/vol.

We look at correlation between tilt and timing to ensure their additivity. Figure 3 displays the rolling 252-day correlation. The average correlation over performance window is -18%. The low or slightly negative correlation between tilt and timing suggests that together, they can enhance risk-adjusted returns and generate a more consistent return profile across SBC regimes.

Figure 3: Correlation between tilt and timing

Notes: 252-day correlation between tilt and timing returns measured over period July 31, 2006–March 31, 2024. Correlation statistics from a stylized back-test for a hypothetical two-asset tilt portfolio containing U.S. 10-year Treasury Note Future and 5-year U.S. High Yield CDX rebalanced daily. Asset returns based on TY1 Comdty and Markit CDX.NA.HY 5-year Excess Return Index. Performance measured over period July 31. 2006–March 31, 2024. Correlation statistics from a stylized back-test for a hypothetical two-asset tilt portfolio containing U.S. 10-year Treasury Note Future and 5-year U.S. High Yield CDX rebalanced daily. Asset returns based on TY1 Comdty and Markit CDX.NA.HY 5-year Excess Return Index. Performance measured over period July 31, 2006–March 31, 2024. Risk and return statistics are reported annualized using monthly data. Growth and inflation regimes defined using difference between 3-month and 6-month average for U.S. Manufacturing PMI and headline CPI YOY respectively. Interest rate volatility regimes defined using ICE BofA MOVE Index. Risk regimes defined using Chicago Board Options Exchange Volatility (VIX) Index. Risk parity regime returns calculated for a hypothetical inverse volatility weighted portfolio of S&P 500 Index and Bloomberg Barclays U.S. Treasury Index. Performance measured over period July 31, 2006–March 31, 2024. Source: BlackRock, with data from Bloomberg. For illustrative purposes only. The figures shown relate to past performance. Past performance is not a reliable indicator of current or future results.

This presentation contains back-tested data for the indices listed above. Unless otherwise noted, returns do not reflect any management fees, transaction costs or expenses. Indexes are unmanaged and one cannot invest directly in an index. Hypothetical data results are based on criteria applied retroactively with the benefit of hindsight and knowledge of factors that may have positively affected its performance and cannot account for risk factors that may affect actual performance. The back-tested past performance returns are shown for illustrative purposes only and are not meant to be representative of actual performance returns of any account, portfolio or strategy. The back-tested performance period is from July 31, 2006–March 31, 2024.

The securities or asset classes in the back-tested portfolios were selected with the full benefit of hindsight, after their performance returns over the period shown was known. It is not likely that similar results could be achieved in the future. Back-tested performance returns have certain limitations. Unlike actual performance returns, they do not reflect actual trading, liquidity constraints, fees and other costs. Back-tested performance returns are indicative of a hypothetical portfolio rebalanced daily. No representation is being made that any account, portfolio or strategy will or is likely to achieve results similar to those shown.

Optimization methods balance the return forecasts from these characteristics against the risk of deviating from a given benchmark. The systematic asset allocation framework laid out in this research is versatile and can be useful for top-level multi-asset allocation seeking to harvest broad risk premia and for systematic global macro strategies that exploit short-term opportunities. It adapts to market conditions by emphasizing tilts in stable markets and focuses on timing for enhanced alpha potential and defensiveness in periods when outcomes are driven by tails of asset return distribution.

Figure 4 and the table below show how a combination of tilt and timing would have yielded better risk adjusted performance with an information ratio of 1.23 vs. 0.81 for tilt and 0.86 for timing. The correlation statistics highlight the additivity from combining a defensive timing overlay to a traditional risk on tilt as measured by lower correlation to broad market betas. The Systematic framework laid out in this research seeks to deliver a portfolio with a more upside participation and helps mitigate losses in down markets across SBC regimes.

Figure 4: Cumulative simulated returns of tilting + timing

Performance statistics for blended tilt +timing portfolio 

Performance statistics for blended tilt +timing portfolio

Notes: Summary of stylized backtest showing cumulative performance of hypothetical performance statistics for tilt, timing and tilt + timing portfolio in Figure 8 and in Table 2. Correlation statistics calculated using data for Bloomberg US Treasury Total Return Index, Bloomberg US Corporate High Yield Total Return Index and S&P 500 Total Return Index. The figures shown relate to simulated past performance. Past performance is not a reliable indicator of current or future results. Index performance returns do not reflect any management fees, transaction costs or expenses. Indices are unmanaged and one cannot invest directly in an index.

Stylized backtest for a hypothetical two-asset tilt portfolio containing US 10-Year Treasury Note Future and 5 Year US High Yield CDX rebalanced daily. Performance measured over period July 31, 2006 - March 31, 2024. Risk and return statistics are reported annualized using monthly data. Growth and inflation regimes defined using difference between 3-month and 6-month average for US Manufacturing PMI and headline CPI YOY respectively. Interest rate volatility regimes defined using ICE BofA MOVE Index. Risk regimes defined using Chicago Board Options Exchange Volatility (VIX) Index. Risk Parity regime returns calculated for a hypothetical inverse volatility weighted portfolio of S&P500 Index and Bloomberg Barclays US Treasury Index. Performance measured over period July 31, 2006 - March 31, 2024. Source: BlackRock, with data from Bloomberg. For illustrative purposes only. The figures shown relate to past performance. Past performance is not a reliable indicator of current or future results.

This presentation contains back-tested data for the indices listed above. Unless otherwise noted, returns do not reflect any management fees, transaction costs or expenses. Indexes are unmanaged and one cannot invest directly in an index. Hypothetical data results are based on criteria applied retroactively with the benefit of hindsight and knowledge of factors that may have positively affected its performance, and cannot account for risk factors that may affect actual performance. The back-tested past performance returns are shown for illustrative purposes only and are not meant to be representative of actual performance returns of any account, portfolio or strategy. The back-tested performance period is from July 31, 2006 – March 31, 2024.

The securities or asset classes in the back-tested portfolios were selected with the full benefit of hindsight, after their performance returns over the period shown was known. It is not likely that similar results could be achieved in the future. Back-tested performance returns have certain limitations. Unlike actual performance returns, they do not reflect actual trading, liquidity constraints, fees and other costs. Back-tested performance returns are indicative of a hypothetical portfolio rebalanced daily. No representation is being made that any account, portfolio or strategy will or is likely to achieve results similar to those shown.

Conclusion

The post-pandemic shifts in interest rates and inflation have significantly changed market dynamics. This new reality means both discretionary and systematic investors should reconsider asset allocation approaches that have been developed and tested over the last two decades. Given the current positive but unstable SBC, its crucial for those that manage assets to be more deliberate about navigating “risk” and “risk parity” regimes. By examining historical periods of high inflation and positive SBC, we can unlock valuable insights to allow for more dynamic allocation for both negative and positive SBC regimes. A systematic allocation approach, guided by economic sensibility and powered by these data-driven insights is uniquely positioned in seeking to deliver a more consistent portfolio outcomes in a world of unpredictability.

Authors

Raffaele Savi
Global Head of BlackRock Systematic and Co-Chief Investment Officer and Co-Head of BlackRock’s Systematic Equities
Tom Parker, CFA
Chief Investment Officer of Systematic Fixed Income
Jeffrey Rosenberg, CFA
Senior Portfolio Manager for Systematic Fixed Income
Ignacio Blanch
Head of Research and Innovation for Systematic Fixed Income
Jasmita Mohan, FRM
Portfolio Manager for Systematic Fixed Income
Nikolaos Prezas
Quantitative Researcher for Systematic Fixed Income