How is correlation used in modern portfolio theory?

By Mark Holland / a couple of days ago

Modern portfolio theory (MPT) states that an investor can achieve diversification and reduce the risk of losses by reducing the correlation between the returns on the assets selected for the portfolio. The objective is to optimize the expected return against a certain level of risk.

Key takeaways

• Followers of MPT look for zero or close to zero correlation in the price movements of the various assets in a portfolio.
• That is, they look for assets that respond to macroeconomic trends in clearly different patterns.
• The ideal selection of assets will have the highest possible return for the desired level of risk.

Modern portfolio theorist recommends that an investor measure the correlation coefficients between the returns of various assets to strategically select those that are least likely to lose value at the same time. That means determining to what extent asset prices tend to move in the same direction in response to macroeconomic trends.

Perfect correlation

MPT is a mathematical system for selecting investments that, in combination, will provide the best returns for a given level of risk.

The theory looks for the best correlation between the expected return and the expected volatility of various potential investments. The economist Harry Markowitz, who introduced modern portfolio theory in 1952, titled the optimal risk-reward ratio as the efficient frontier.

A portfolio is known as “Markowitz-efficient” if its asset selection is designed to return the maximum possible returns without any increase in risk.

If the correlation is zero, the two assets do not have a predictive relationship.

In MPT, the efficient frontier is where the investor will find the combination of assets that offers the highest possible return for a chosen level of risk. These assets demonstrate the optimal correlation between risk and return.

The correlation scale

Correlation is measured on a scale from -1.0 to +1.0:

• If two assets have an expected return correlation of 1.0, that means they are perfectly correlated. If one earns 5%, the other earns 5%. If one falls 10%, so does the other.
• A perfectly negative correlation (-1.0) implies that the gain on one asset is proportionally equal to the loss on the other asset.
• A zero correlation indicates that the two assets do not have a predictive relationship.

MPT emphasizes that investors should look for an asset pool that is uncorrelated (almost zero) to limit risk. In practical terms, that pretty much guarantees a diversified portfolio.

Criticisms of the perfect correlation theory

One of the main criticisms of Markowitz’s theory lies in its assumption that the correlation between assets is fixed and predictable. In the real world, the systematic relationships between different assets do not remain constant.

That means MPT becomes less useful in times of uncertainty, which is exactly when investors need the most protection against volatility.

Others claim that the variables used to measure correlation coefficients are themselves flawed and the actual risk level of an asset can be miscalculated. The expected values ​​are mathematical expressions of the implicit covariance of future returns and not historical measures of actual returns.

www.investopedia.com