Markowitz Modern Portfolio Theory: Expected Return, Covariance, Portfolio Variance, and Diversification

1. Conceptual Foundations: Harry Markowitz's Mean-Variance Revolution

Prior to 1952, investment analysis was dominated by intuitive, rules-of-thumb stock selection. Investors evaluated individual securities in isolation, seeking firms with the highest expected returns and sound balance sheets. In March 1952, a 25-year-old doctoral candidate at the University of Chicago, Harry M. Markowitz, published his landmark paper "Portfolio Selection" in The Journal of Finance, fundamentally transforming the discipline and earning him the 1990 Nobel Prize in Economic Sciences.

Markowitz's modern portfolio theory (MPT) is anchored in foundational behavioral and mathematical axioms:

• The Holistic Portfolio Perspective: Markowitz proved that an asset should never be judged in isolation; it must be evaluated strictly on its contribution to the risk and return of the investor's overall portfolio. A highly volatile stock can actually lower total portfolio risk if its returns move inversely to the rest of the portfolio.
• Mean-Variance Investor Rationality: Investors are assumed to be rational and risk-averse. When making investment choices, investors consider only two statistical parameters: the Mean (Expected Return), which is desirable, and the Variance (Standard Deviation), which is undesirable.
• The Dominance Principle: Given two portfolios with identical expected returns, rational investors will strictly prefer the one with lower variance. Given two portfolios with identical variance, rational investors will strictly prefer the one with higher expected return.
• The Essence of Diversification: Markowitz formalized the ancient adage "don't put all your eggs in one basket." However, his profound mathematical insight was that true diversification depends not on the number of stocks held, but on the correlation and covariance among them.
• Frictionless Capital Markets: The theoretical framework assumes zero transaction costs, zero personal taxes, infinite asset divisibility, and long-only portfolio constraints where asset weights sum to 1.0.

2. Portfolio Expected Return and the Budget Constraint

The expected return on a multi-asset portfolio is mathematically straightforward: it is the linear weighted average of the expected returns of the individual constituent securities.

• Two-Asset Expected Return Formula:
  $$\boxed{ E(R_p) = w_A E(R_A) + w_B E(R_B) }$$
  where \( w_A \) and \( w_B \) represent the proportion of total wealth invested in Security A and Security B.
• General N-Asset Expected Return:
  $$\boxed{ E(R_p) = \sum_{i=1}^{N} \left[ w_i \times E(R_i) \right] }$$
• The Budget Constraint: All available capital is fully deployed across the portfolio, meaning the sum of all asset weights must equal 1.0:
  $$\boxed{ \sum_{i=1}^{N} w_i = w_1 + w_2 + \dots + w_N = 1.0 \quad \text{or} \quad w_B = 1 - w_A }$$
• Long-Only Constraint: In standard introductory portfolio theory, short selling is prohibited, restricting asset weights to positive values: \( 0 \le w_i \le 1.0 \).

3. Covariance, Correlation, and the Mathematics of Portfolio Variance

While portfolio return is a simple linear weighted sum, portfolio risk is decidedly non-linear. The variance of a portfolio depends not only on individual asset variances, but heavily on the pairwise covariances between all pairs of assets.

• Two-Asset Portfolio Variance Formulation:
  $$\boxed{ \sigma_p^2 = w_A^2 \sigma_A^2 + w_B^2 \sigma_B^2 + 2 w_A w_B \text{Cov}(A,B) }$$
  or substituting ( ext{Cov}(A,B) = ho_{AB} sigma_A sigma_B ):
  $$\boxed{ \sigma_p^2 = w_A^2 \sigma_A^2 + w_B^2 \sigma_B^2 + 2 w_A w_B \rho_{AB} \sigma_A \sigma_B }$$
• Covariance (( ext{Cov}_{AB} )): A statistical measure of the degree to which two random variables move together:
  $$\boxed{ \text{Cov}(A,B) = \sum_{i=1}^{m} P_i \left[ R_{Ai} - E(R_A) \right] \left[ R_{Bi} - E(R_B) \right] }$$
  Positive covariance indicates assets move in the same direction; negative covariance indicates they move in opposite directions.
• Correlation Coefficient (( ho_{AB} )): Standardizes covariance between -1.0 and +1.0:
  $$\boxed{ ho_{AB} = \frac{ \text{Cov}(A,B)}{\sigma_A \times \sigma_B} }$$
• The Correlation Spectrum Dynamics:
  • Perfect Positive Correlation (( ho = +1.0 )): Portfolio standard deviation is a simple weighted average: ( sigma_p = w_A sigma_A + w_B sigma_B ). Zero diversification benefit; risk is not reduced!
  • Uncorrelated Assets (( ho = 0.0 )): The cross-product term drops to zero: ( sigma_p^2 = w_A^2 sigma_A^2 + w_B^2 sigma_B^2 ). Substantial risk reduction occurs; portfolio risk is strictly lower than the weighted average.
  • Perfect Negative Correlation (( ho = -1.0 )): Complete risk elimination is mathematically possible! By setting weights properly, portfolio variance can be driven to exactly ZERO.
• The Minimum Variance Portfolio (MVP): For a two-asset portfolio, calculus determines the exact weight in Asset A that minimizes portfolio variance:
  $$\boxed{ w_A^* = \frac{\sigma_B^2 - \text{Cov}(A,B)}{\sigma_A^2 + \sigma_B^2 - 2 \text{Cov}(A,B)} }$$

4. Mathematical Formulation & Worked Numerical: The Diversification Effect Under Variable Correlation

The following analytical framework details Markowitz portfolio formulations, accompanied by a step-by-step worked numerical problem evaluating how varying correlation coefficients alter portfolio risk.

6. Corporate Case Vignette: Institutional Multi-Asset Portfolio Allocation

7. AKTU University Exam Practice Drill