Chapter 15: Introduction to Modern Portfolio Theory
15.1 Assumptions of Modern Portfolio Theory and Investor Risk Archetypes
The Markowitz Revolution
Introduced by Harry Markowitz in his seminal 1952 paper Portfolio Selection, Modern Portfolio Theory (MPT) revolutionized investment finance. Prior to Markowitz, asset allocation focused almost exclusively on evaluating the individual risk and return characteristics of isolated securities. Markowitz proved mathematically that an asset's risk should not be assessed in isolation, but by how it contributes to the overall risk and return of a diversified portfolio through its covariances with other portfolio assets.
+---------------------------------------------------------------------------------------------------+
| CORE ASSUMPTIONS OF MODERN PORTFOLIO THEORY (MPT) |
+-----------------------------------+---------------------------------------------------------------+
| Single-Period Investment Horizon | - All investors plan for an identical single holding period |
| | (e.g., 1 year) and make decisions at time t = 0 |
+-----------------------------------+---------------------------------------------------------------+
| Mean-Variance Optimization | - Investors evaluate portfolios based solely on two parameters|
| | Expected Return E(R) and Variance of Return (sigma^2) |
+-----------------------------------+---------------------------------------------------------------+
| Strict Risk Aversion | - Given identical expected returns, investors prefer lower |
| | variance; for higher risk, they demand higher return |
+-----------------------------------+---------------------------------------------------------------+
| Frictionless Capital Markets | - Zero transaction costs, zero taxes, perfect asset |
| | divisibility, and no restrictions on short-selling |
+-----------------------------------+---------------------------------------------------------------+
| Homogeneous Expectations | - All market participants agree on the probability |
| | distributions, expected returns, variances, and covariances |
+-----------------------------------+---------------------------------------------------------------+
Investor Risk Archetypes
Modern portfolio economics defines three primary investor risk orientations governed by the mathematical curvature of their utility functions $U(W)$:
+-----------------------+-------------------------------+-------------------------------------------+
| Risk Archetype | Marginal Utility of Wealth | Mathematical Utility Function Curvature |
+-----------------------+-------------------------------+-------------------------------------------+
| Risk-Averse Investor | Diminishing Marginal Utility | Strictly Concave (U''(W) < 0) |
| | (Prefers certain wealth over | Rejects a fair gamble with E(payoff) = 0 |
| | risky gamble of equal mean) | Requires a positive Risk Premium |
+-----------------------+-------------------------------+-------------------------------------------+
| Risk-Neutral Investor | Constant Marginal Utility | Linear (U''(W) = 0) |
| | (Indifferent between certainty| Evaluates choices solely on expected value|
| | and gamble of equal mean) | Zero risk premium demanded |
+-----------------------+-------------------------------+-------------------------------------------+
| Risk-Seeking Investor | Increasing Marginal Utility | Strictly Convex (U''(W) > 0) |
| | (Prefers gamble over certain | Willingly accepts unfair gambles to attain|
| | wealth of identical mean) | extreme potential payoffs |
+-----------------------+-------------------------------+-------------------------------------------+
MPT assumes that institutional and private wealth investors are strictly risk-averse.
15.2 Mathematical Formulations: Expected Return, Security Variance, and Two-Asset Portfolio Variance
Expected Rate of Return and Variance of Individual Securities
For an individual security $i$ evaluated across $S$ possible economic states of nature, where each state $s$ has an objective probability $p_s$:
- Expected Rate of Return $E(R_i)$: $$E(R_i) = \sum_{s=1}^S p_s R_{is}$$ Where $\sum_{s=1}^S p_s = 1.0$.
- Variance of Return (Ex-Ante Risk) $\sigma_i^2$: $$\sigma_i^2 = \sum_{s=1}^S p_s \left[ R_{is} - E(R_i) \right]^2$$ The standard deviation $\sigma_i = \sqrt{\sigma_i^2}$ quantifies the total dispersion of returns around the expected mean.
Portfolio Expected Return
The expected rate of return of a portfolio consisting of $N$ assets is the weighted arithmetic average of the expected returns of the individual securities: $$E(R_p) = \sum_{i=1}^N w_i E(R_i)$$ Where $w_i$ represents the proportion of total portfolio capital invested in security $i$, subject to $\sum_{i=1}^N w_i = 1.0$.
Two-Asset Portfolio Variance Formulation
For a portfolio comprising two risky assets (Asset 1 and Asset 2) with weights $w_1$ and $w_2 = (1 - w_1)$:
$$\sigma_p^2 = w_1^2 \sigma_1^2 + w_2^2 \sigma_2^2 + 2 w_1 w_2 \text{Cov}(R_1, R_2)$$
Since covariance can be expressed in terms of the correlation coefficient $\rho_{12}$: $$\text{Cov}(R_1, R_2) = \rho_{12} \sigma_1 \sigma_2$$ The two-asset portfolio variance equation becomes: $$\sigma_p^2 = w_1^2 \sigma_1^2 + w_2^2 \sigma_2^2 + 2 w_1 w_2 \rho_{12} \sigma_1 \sigma_2$$ $$\sigma_p = \sqrt{w_1^2 \sigma_1^2 + w_2^2 \sigma_2^2 + 2 w_1 w_2 \rho_{12} \sigma_1 \sigma_2}$$
+---------------------------------------------------------------------------------------------------+
| Mathematical Proof of Diversification: |
| If Asset 1 has sigma_1 = 20%, Asset 2 has sigma_2 = 12%, weights w_1 = 0.50, w_2 = 0.50: |
| - If rho = +1.0 (Perfect Positive): sigma_p = 0.5(20%) + 0.5(12%) = 16.0% (Zero Risk Reduction) |
| - If rho = 0.0 (Uncorrelated): sigma_p = sqrt(0.25*400 + 0.25*144 + 0) = sqrt(136) = 11.66% |
| - If rho = -1.0 (Perfect Negative): sigma_p = |0.5(20%) - 0.5(12%)| = 4.0% (Dramatic Reduction!) |
+---------------------------------------------------------------------------------------------------+
15.3 Correlation Dynamics, Global Minimum Variance Portfolio (GMVP), and Markowitz Efficient Frontier
The Correlation Spectrum ($\rho \in [-1.0, +1.0]$)
The degree of portfolio risk reduction achieved through diversification depends upon the correlation coefficient $\rho_{ij}$ between asset pairs:
Correlation rho = +1.0 : Perfectly Positive. Straight line between assets in risk-return space. Zero diversification.
Correlation rho = 0.0 : Uncorrelated. Significant curved bow to the left. Substantial risk reduction.
Correlation rho = -1.0 : Perfectly Negative. Triangle folding back to the vertical return axis. Perfect risk elimination!
The Global Minimum Variance Portfolio (GMVP)
For two risky assets, there exists a unique combination of weights that minimizes total portfolio variance, known as the Global Minimum Variance Portfolio (GMVP):
$$w_1^* = \frac{\sigma_2^2 - \rho_{12} \sigma_1 \sigma_2}{\sigma_1^2 + \sigma_2^2 - 2 \rho_{12} \sigma_1 \sigma_2}; \quad w_2^* = 1 - w_1^*$$
Whenever $\rho_{12} < \frac{\sigma_1}{\sigma_2}$ (assuming $\sigma_1 < \sigma_2$), the GMVP achieves a standard deviation strictly lower than the standard deviation of the least risky individual asset!
The Markowitz Efficient Frontier
When the mean-variance optimization algorithm is applied across an entire universe of $N$ securities:
Expected Return E(R)
^
| / Efficient Frontier (Dominant Portfolios)
| /
| +-------------+
| /
| GMVP
| \
| +-------------+ Inefficient / Dominated Segment
| \
+-------------------------------------------> Risk / Standard Deviation (sigma)
- Attainable Set (Opportunity Set): The convex region containing all feasible portfolios that can be constructed from the universe of $N$ securities.
- The Efficient Frontier: The upper boundary of the attainable set, spanning from the Global Minimum Variance Portfolio (GMVP) upwards to the maximum-return asset.
- Definition of Mean-Variance Dominance: A portfolio on the Efficient Frontier dominates all other portfolios because it satisfies one of two conditions:
- It offers the maximum expected return for a given level of risk ($\sigma$).
- It offers the minimum risk for a specified target expected return. Portfolios lying below the efficient frontier are suboptimal, as superior returns can be achieved at identical or lower risk levels.