How correlation impacts diversification in investing

Key takeaways:

• Correlation helps explain how assets move in relation to one another, which can show whether diversification may reduce reliance on a single investment or risk driver.
• Measuring correlation through formulas, portfolio correlation concepts and correlation matrices can support diversification analysis, but these figures are based on historical relationships.
• In portfolio risk management, low or negative correlations may help manage volatility, while systematic risk cannot be diversified away and market stress can change asset relationships.
• Practical portfolio building may use correlation insights through Modern Portfolio Theory, dynamic rebalancing, minimum variance strategies and broader asset class diversification.
• Correlation has limitations because relationships can change over time, past data may not predict future behaviour, and liquidity, costs, valuation and time horizon also matter.

Investing involves balancing growth opportunities with the inherent risks of financial markets.

Diversification is one widely used way to reduce reliance on any single investment by combining assets with different characteristics. Correlation helps explain how those assets may move in relation to one another.

Understanding the relationship between correlation and diversification can help investors assess how a portfolio may behave under different market conditions.

What is correlation?

Correlation is a statistical measure that describes how two variables move in relation to each other. In the context of investments, it evaluates the relationship between the price movements of different assets within a portfolio. Correlation is expressed on a scale from -1 to +1:

• +1.0. Perfect positive correlation. The assets move in the same direction at the same rate.
• 0. No correlation. The two assets show no linear relationship in their movements.
• -1.0. Perfect negative correlation. The assets move in completely opposite directions.

These relationships are useful when assessing diversification. For example:

• Positively correlated assets. Equities from similar industries may move together, which can increase exposure to the same gains and losses.
• Negatively correlated assets. Stocks and certain bonds have sometimes moved in opposite directions, which can help reduce volatility—but correlations can change.
• Zero or weakly correlated assets. Combining equities with assets such as commodities or real estate may introduce less-related return drivers and reduce some concentration risks.

Understanding correlation can help investors assess how an asset mix may affect risk and return. This may support diversification decisions, although it cannot prevent losses, and diversification benefits may vary across conditions.

Measuring correlation: Tools and formulas

Correlation is useful for evaluating the relationships between assets and can support portfolio diversification analysis. Various tools and methods help measure correlation and provide quantitative context for portfolio decisions.

Correlation formula

The correlation formula quantifies the linear relationship between two variables, such as the returns of two assets. It is commonly expressed as:

$\frac{\mathrm{COV}(X,Y)}{{\sigma }_X*{\sigma }_Y}$

Where:

• ρ₍X,Y₎ = correlation coefficient between variables X and Y
• Cov(X, Y) = covariance between variables X and Y.
• σX = standard deviation of X.
• σY = standard deviation of Y.

This formula generates a value between -1.0 and +1.0:

• +1.0: Perfect positive linear relationship; the assets move in the same direction.
• -1.0: Perfect negative linear relationship; the assets move in opposite directions.
• 0.0: No linear relationship.

Portfolio correlation concept

In portfolio management, correlations between assets are one input in diversification analysis. Individual asset correlations, asset weights and volatility all influence overall portfolio risk. Pairwise correlations can help estimate diversification effects, but they do not determine future diversification benefits on their own.

Correlation matrix

A correlation matrix is a table displaying correlation coefficients between multiple asset pairs. This tool helps identify relationships across a broad range of assets. For instance, low or negative correlations between equities and bonds may indicate potential diversification effects, depending on the period and asset mix.

What's important to remember is that correlations are not static. Economic events, market cycles, or changes in monetary policy can alter relationships between assets. Reviewing correlation data periodically may help you test whether your diversification assumptions still reflect current market conditions.

Correlation in portfolio risk management

Correlation can play a role in portfolio risk analysis by showing how assets have moved in relation to each other and how those relationships may affect volatility. This can help investors assess whether a portfolio is overly exposed to similar risk drivers.

Portfolio risk can be divided into two categories: systematic risk and unsystematic risk. 

• Systematic risk (market risk). This type of risk impacts all assets due to macroeconomic factors such as global recessions or geopolitical instability. It cannot be diversified away.
• Unsystematic risk. This risk is specific to individual sectors or companies. Diversification across uncorrelated or negatively correlated assets can reduce its impact.

The relationship between correlation and portfolio standard deviation is important for understanding overall risk. Standard deviation measures the variability of portfolio returns, and correlation helps explain how different assets contribute to that variability. When assets with low or negative correlations are combined, portfolio volatility may fall, depending on asset weights, volatility and market conditions.

For instance, equities and high-quality bonds have sometimes shown weak or negative correlations. In some downturns, bonds may help offset equity losses, but this relationship is not stable across all periods.