A Look at the Rise of Stock-Bond Correlations

2022: When Diversification Failed

With stocks¹ returning -19.21% and bonds² returning -13.01%, 2022 was a rough year for most investors. The 60/40 portfolio, which is commonly thought of as the average investors’ portfolio, returned -16.73%; it’s second-worst year over the last three decades. Notably, 2022 was the only year where both stocks and bonds simultaneously had negative returns.

Figure 1 – Yearly Stock & Bond Performance

Shifting Correlations

Investors have historically relied upon the negative correlation between stocks and bonds to protect their portfolios when one of the asset classes has a drawdown, but many are beginning to question the assumption that stocks and bonds truly act as portfolio diversifiers to each other.

We can begin to investigate the evolution of stock-bond correlations by plotting the rolling, 1-year correlation between stocks and bonds over the past few decades:

Figure 2 – Rolling, 1-Year Stock & Bond Correlations

A few quick takeaways are immediately apparent:

• Stock and bond correlations are not stable over time.
• Shifts in correlation can be dramatic and swift.
• The past 2 years have seen a steady upward trend in stock and bond correlations.
• The current level (.20), while high relative to the past 2 decades, is not as high as the levels reached in prior peaks.

The current year also stands out for two reasons: it contains the highest ever inter-month, stock-bond correlation reading (.85 during the month of July) and 2023 has exhibited the most correlation volatility³ of any year in the dataset. (see appendix).

Figure 3 – 2023 Inter-Month Correlation

Correlation Drivers

The obvious next questions are: What is driving the correlation upward? Why is this regime shift so volatile? Will it remain elevated – perhaps even continue to trend upwards?

Numerous studies⁴ undertaken by researchers and industry practitioners have explored the fundamental drivers between stock-bond correlations. The primary takeaway from each study is that there are several recurrent macro factors which have historically helped explain changes in stock-bond correlations. We re-tested several of these stated macro factors ourselves and below summarize some of the most impactful variables for explaining the monthly stock-bond correlation reading by outputting the correlation of the inter-month stock-bond correlation (SBC) with several factors:

Table 1 – Macro Factors Relationship to Stock Bond Correlation (SBC)

Looking Ahead

The relationships above help show which factors explain correlation in the same period, but of more practical relevance is determining what we can expect going forward. Because those macro-variables are not known until after the fact, they cannot aid us in forming current predictions about the future, unless we concurrently formulate predictions about the relevant macro variables.

Fortunately, historical data has shown that near-term stock-bond correlations can be predicted reasonably well using a simple model: predict that next month’s correlation will be equal to the month that just ended. Stock-bond correlation tends to exhibit a high-degree of autocorrelation and stickiness so a sensible default option is to use the current value as our baseline forecast. The below scatterplot depicts a fitted regression equation which uses the current-month’s correlation as the only input to forecasting next months value:

Figure 4 – Basic Correlation Prediction Model

Taking above into account our view is that stock-bond correlations are likely to remain in elevated territory for roughly the next 12-months, but as inflation and inflation uncertainty continue to decline, the stock-bond correlation will drift back to lower levels.

Portfolio Takeaways

What are the consequences of this view?

While we believe diversification is a critical element to any investment strategy, 2022 was a good reminder that diversification doesn’t always work over the short-term. In today’s environment of elevated correlations, investors with stock-bond portfolios may have to assume more risk to maintain a consistent level of expected return. While it is always prudent to look to other regions and asset classes to provide added diversification, it is increasingly important in times like these. That said, not all alternatives – commodities, real estate, etc. – are created equally and should be evaluated based on their role within the portfolio.

It’s worth noting that the performance of the classic 60/40 portfolio in 2022 was well within the range of outcomes and that the long-term outlook for stocks and bonds has not radically changed. Investors should be careful not to extrapolate recent trends too far into the future and overreact with any dramatic portfolio changes.

Appendix

Table 2: Inter-Month Stock Bond Correlation Stats Summarized by Year

IMPORTANT NOTES

[1] Stock represented with the Rusell 3000 TR Index
[2] Bonds represented with the Bloomberg US Agg TR Index
[3] Volatility of correlation regimes is judged using the standard deviation of inter-month correlations
[4] A Changing Stock-Bond Correlation (aqr.com); The stock/bond correlation amid rising inflation: Increasing, but hardly a regime change (nl.vanguard); Empirical evidence on the stock-bond correlation

This material is for informational purposes and is intended to be used for educational and illustrative purposes only. It is not designed to cover every aspect of the relevant markets and is not intended to be used as a general guide to investing or as a source of any specific investment recommendation. It is not intended as an offer or solicitation for the purchase or sale of any financial instrument, investment product or service. This material does not constitute investment advice, nor is it a substitute for such professional advice or services, nor should it be used as a basis for any decision or action that may affect your business. Before making any decision or taking any action that may affect your business, you should consult a qualified professional adviser. In preparing this material we have relied upon data supplied to us by third parties. The information has been compiled from sources believed to be reliable, but no representation or warranty, express or implied, is made by Innealta Capital, LLC as to its accuracy, completeness or correctness. Innealta Capital, LLC does not guarantee that the information supplied is accurate, complete, or timely, or make any warranties with regard to the results obtained from its use. Innealta Capital, LLC has no obligations to update any such information. Past performance is not indicative of future results. All investments involve risk and unless otherwise stated, are not guaranteed. Be sure to consult with a tax professional before implementing any investment strategy.

Table 1 – Stock-bond correlation is the inter-month correlation between the Russell 3000 TR Index and the Bloomberg U.S. Aggregate Bond Total Return Index. Level of inflation refers to the trailing, year of year % change in the CPI, start date of 1/31/1989. Change in Real Interest Rates refers to the month over month gross change in the 10-year TIPs, start date of 2/28/2003. Unemployment refers to the U3 unemployment rate, start date of 1/31/1989. Equity volatility refers to the trailing 12-month, daily standard deviation of the Russell 3000 TR Index start date of 12/31/1989. Level of interest rates refers to the average level of the 10-year, US Treasury interest rate, start date of 1/31/1989. All variables aside from equity volatility were sourced from the St. Louis Federal Reserve (FRED)

Index Definitions: The Russell 3000 Value TR Index measures the performance of the largest 3,000 U.S. companies representing approximately 96% of the investable U.S. equity market, as of the most recent reconstitution. The Bloomberg U.S. Aggregate Bond TR Index is representative of the entire universe of taxable fixed-income investments. It includes issues of the U.S. Government and any agency thereof, corporate issues of investment grade quality (Baa/BBB or better), and mortgage-backed securities.

3423-INN-12/06/2023