Portfolio Diversification: A Core Strategy

Unsystematic vs. Systematic Risk: The Two Faces of Risk

The first crucial step in understanding diversification is to differentiate between two fundamental types of risk. Not all risk is created equal, and not all risk can be managed in the same way.

1. Unsystematic Risk: Also known as specific or idiosyncratic risk, this pertains to a single company, industry, or security. Examples include a key product failing, a factory fire, a management scandal, or a poor earnings report. These events are "specific" to the asset. The key feature of unsystematic risk is that it can be significantly reduced or even eliminated through diversification. By holding a wide variety of assets, the impact of a single negative event becomes negligible for the portfolio as a whole.

2. Systematic Risk: Also known as market risk, this is risk that affects the entire market or a large segment of it. Examples include recessions, changes in interest rates, geopolitical conflicts, or inflation shocks. This type of risk cannot be avoided by adding more assets to your portfolio, because all assets are affected to some degree.

The image below provides a clear summary of the distinctions between these two types of risk.

The core idea of diversification is to eliminate the 'un-rewarded' unsystematic risk. The Capital Asset Pricing Model (CAPM), which you will have studied in CM1, posits that investors are not compensated for taking on unsystematic risk, precisely because it can be easily diversified away. The market only rewards investors for bearing systematic risk, which is unavoidable.

The Mathematics of Diversification: The Power of Correlation

How does adding more assets reduce risk? The answer lies in the statistical relationship between their returns, specifically their correlation.

Let's consider a simple portfolio with two assets, A and B. The variance of the portfolio's return (${\sigma }_p^2$), which is our measure of risk, is given by the formula:

Where:

• $w_A$ and $w_B$ are the weights of the assets in the portfolio.
• ${\sigma }_A^2$ and ${\sigma }_B^2$ are the variances of the individual assets.
• ${\rho }_{AB}$ is the correlation coefficient between the returns of asset A and asset B.

The crucial term is the last one. The correlation coefficient, ${\rho }_{AB}$, measures how the two assets move together.

• If ${\rho }_{AB}=+1$ (perfect positive correlation), the assets move in perfect lockstep. In this case, there is no reduction in risk from diversification. The portfolio's risk is simply the weighted average of the individual asset risks.
• If ${\rho }_{AB}<+1$, the assets do not move perfectly together. The portfolio's variance will be less than the weighted average of the individual variances. This reduction in risk is the diversification benefit.
• The lower the correlation (ideally, negative), the greater the diversification benefit. If ${\rho }_{AB}=-1$ (perfect negative correlation), it's theoretically possible to construct a portfolio with zero risk.

The following graph beautifully illustrates this principle. It shows how the proportional reduction in a portfolio's volatility changes as you add more assets, depending on the average correlation between them.

As you can see, the biggest risk reduction comes from the first few assets you add. As you add more and more, the marginal benefit decreases, and the total portfolio risk approaches the level of undiversifiable systematic risk.

From Diversification to the Efficient Frontier

The principle of diversification leads directly to one of the most important concepts in finance: the Efficient Frontier.

When we combine assets with different expected returns, risks (variances), and correlations, we create a universe of possible portfolios. Each point in this universe represents a portfolio with a specific risk and return combination.

The video below offers a simple and clear explanation of how diversification causes the risk-return trade-off to "bend," creating a set of superior portfolios.

This "bend" is the diversification benefit in action. The outer edge of this curve of possible portfolios is called the Efficient Frontier. It represents the set of portfolios that are "optimal" in the sense that they offer the highest possible expected return for any given level of risk. A rational investor would only ever choose a portfolio that lies on this frontier.

To see how this frontier is constructed from a set of individual assets, let's watch a more detailed walkthrough.

Diversification in Practice

For institutional investors like insurers and pension funds, diversification is not just a theoretical concept; it is a fundamental practice applied at multiple levels.

As the ActEd notes explain, diversification occurs both between and within asset classes:

• Between Asset Classes: A typical institutional portfolio will hold a mix of equities, bonds, property, cash, and alternative assets (like infrastructure or private equity). These broad categories have different risk-return profiles and react differently to economic events.
• Within Asset Classes:
  • Equities: Investing in different industries (technology, healthcare, energy), countries (UK, US, emerging markets), and company sizes (large-cap, small-cap).
  • Bonds: Holding a mix of government and corporate bonds, with varying maturities and credit qualities.
  • Property: Owning different types of property (office, retail, industrial, residential) in various geographic locations.

For an insurer, diversification takes on an even broader meaning. It includes managing the relationship between different types of risk across the entire enterprise.

As highlighted in the CRO Forum paper, an insurer's diversification strategies include:

• Pooling similar risks: The very basis of insurance, aggregating thousands of independent house or car insurance policies.
• Pooling dissimilar risks: Writing different lines of business (e.g., motor, property, liability) in different geographic markets.
• Combining opposite risks: This is a particularly insightful actuarial strategy. For example, an insurer might sell both life annuities (where the risk is people living longer than expected - longevity risk) and term life assurance (where the risk is people dying sooner than expected - mortality risk). These risks act as a natural hedge against each other.

The Limits of Diversification

While powerful, diversification is not a silver bullet. It's crucial to understand its limitations.

The most significant limitation is that correlations are not stable. During periods of extreme market stress, such as the 2008 financial crisis or the COVID-19 crash, correlations between seemingly different asset classes can suddenly increase and converge towards +1. In a crisis, everything tends to fall together. This means that diversification benefits can evaporate precisely when they are needed most.

Other challenges include:

• Diworsification: Adding more and more assets to a portfolio without a clear understanding of their underlying risk drivers can lead to excessive complexity and costs, without providing meaningful risk reduction. The portfolio may simply end up mimicking a market index but with higher fees.
• Quantification: Reliably estimating future correlations, especially in the "tails" of the distribution (i.e., during extreme events), is exceptionally difficult. This relies heavily on expert judgment and sophisticated modeling, which can itself be a source of risk.

Conclusion

In this lesson, we have unpacked the vital role of diversification in portfolio construction. We moved from the simple idea of not putting all your eggs in one basket to a more structured understanding of risk and return. We saw that the primary function of diversification is to eliminate firm-specific, or unsystematic, risk, for which investors are not rewarded. The residual market-wide, or systematic, risk is the risk that cannot be diversified away.

The "magic" of diversification is unlocked by combining assets with low or negative correlation. This allows for the construction of portfolios that have lower overall risk than the simple sum of their parts, a phenomenon that gives rise to the Efficient Frontier. However, we also noted the practical limitations, chiefly that the benefits of diversification can diminish during severe market crises when correlations spike.

In our next lesson, we will build directly on these concepts. We will explore Mean-Variance Optimization, the mathematical framework developed by Harry Markowitz that uses the principles of diversification to formally identify the optimal portfolio on the Efficient Frontier for a given investor's risk tolerance.

Key Takeaways & Cheatsheet