Correlations of financial asset returns - stock-bond correlation changes over longer periods

A couple of months ago the research paper on "Empirical evidence on the stock-bond correlation" (July 2023) was published which  shows that there are shifts in the stock-bond correlation over the long term and the notion that the stock-bond correlation is negative is something which we only saw in recent decades.

The correlation coefficient - shown in above graph -  measures the strength and direction of the relationship between two variables - in this case the monthly returns of equities and government bonds with 10-year maturity. If there is a perfect positive linear relationship between two financial assets, the correlation will be 1.0.  0 means no correlation, the variables are independent and do not affect each other.  If there is a perfect negative linear relationship between the two holdings in the period, the correlation coefficient is -1.0. 

Correlation coefficients play a crucial role in portfolio diversification. The key idea is to combine assets that have low or negative correlations to reduce risk withou necessarily sacrificing returns.

The typical portfolio structure for passive investors includes government bonds as a hedge against the volatility of riskier assets like equities. However, if returns on bonds and equities become positively correlated, bonds will no longer serve as an effective hedge.

References:

• Visualising asset price correlations with NetworkX
• Correlation and autocorrelation 
• Which bond types provide the most diversification for stock investors (Morningstar)?
• Morningstar 2023 diversification landscape report - how did diversified portfolios hold up in 2022?
• Does it pay to diversify by style?
• Negative correlations, positive allocations (Pimco - Nov 2024) - the inverse correlation between bonds and stocks has returned, broading potential for risk-adjusted returns in multi-asset portfolio
• The correlation of equity and bond returns