10.2 Capital Asset Pricing Model (CAPM) & Beta

10.2 Capital Asset Pricing Model (CAPM) & Beta

While Markowitz's Modern Portfolio Theory established the mathematical benefits of diversification and the concept of the Efficient Frontier, it required complex mathematical inputs—specifically, hundreds of expected returns, variances, and thousands of pairwise covariances. In the mid-1960s, financial economists William Sharpe (1964), John Lintner (1965), and Jan Mossin (1966) independently developed the Capital Asset Pricing Model (CAPM). Building directly upon Markowitz's foundations, CAPM provides a parsimonious, practical framework for pricing risky financial assets and establishing the required rate of return for an investment based on its systematic market risk.

CAPM is one of the pillars of modern investment management and corporate finance. It establishes that because rational investors can costlessly eliminate unsystematic risk through diversification, the capital markets will only reward investors with an expected return premium for bearing systematic (market) risk.

---

The Risk-Free Asset and the Capital Allocation Line (CAL)

Markowitz's original portfolio framework considered only risky assets (such as equities, corporate bonds, and real estate). The key theoretical innovation of CAPM was the introduction of a Risk-Free Asset ($R_f$):

• Characteristics of the Risk-Free Asset: A theoretical security offering a guaranteed, certain nominal return over a specified investment horizon. It possesses zero variance of return ($\sigma_f = 0$) and zero covariance with every risky asset in the financial market ($\text{Cov}(i, f) = 0$).
• Empirical Proxy: In practical wealth management, the risk-free rate is approximated by short-term sovereign debt instruments backed by the full taxing authority and credit of a stable government—such as 3-month or 1-year United States Treasury Bills (T-bills) or United Kingdom Treasury Bills (gilts).

Capital Allocation Line (CAL)

When an investor combines the risk-free asset with a specific portfolio of risky assets ($P$), the resulting risk-return profile is linear. The line originating at the risk-free rate on the vertical axis and passing through risky portfolio $P$ is called the Capital Allocation Line (CAL):

$E(R_c) = R_f + \left( \frac{E(R_p) - R_f}{\sigma_p} \right) \sigma_c$

The slope of the CAL is the Sharpe Ratio of portfolio $P$:

$\text{Sharpe Ratio} = \frac{E(R_p) - R_f}{\sigma_p}$

The Sharpe Ratio measures the excess return generated per unit of total risk (standard deviation). An investor seeking to maximize risk-adjusted performance will rotate the CAL upward until it is precisely tangent to the Markowitz Efficient Frontier.

---

The Capital Market Line (CML) and the Market Portfolio

Under CAPM's core assumptions—that all investors possess identical economic horizons and identical, rational expectations (homogeneous expectations)—every investor will identify the exact same optimal tangency portfolio on the Markowitz Efficient Frontier. Because all investors hold risky assets in the identical proportions dictated by this tangency portfolio, the tangency portfolio must logically represent the aggregate of all available risky assets weighted by their market values. This unique asset basket is termed the Market Portfolio ($M$).

The Capital Market Line (CML)

The specific Capital Allocation Line connecting the risk-free rate ($R_f$) and passing through the Market Portfolio ($M$) is designated as the Capital Market Line (CML):

$E(R_p) = R_f + \left( \frac{E(R_m) - R_f}{\sigma_m} \right) \sigma_p$

Where:

• $E(R_p)$ = expected return of the combined efficient portfolio
• $R_f$ = risk-free rate of return
• $E(R_m)$ = expected return of the broad market portfolio
• $\sigma_m$ = standard deviation of the market portfolio
• $\sigma_p$ = standard deviation of the combined portfolio
• $[E(R_m) - R_f] / \sigma_m$ = the slope of the CML, representing the market price of risk

Lending and Borrowing on the CML

The CML provides investors with a simple two-stage decision process, known as the Tobin Separation Theorem:

1. Investment Decision (Technical): Identify the optimal Market Portfolio ($M$) of risky assets. This portfolio is identical for all rational market participants.
2. Financing Decision (Personal): Combine the Market Portfolio with the risk-free asset according to personal risk tolerance:
   • Lending Portfolios (Left of $M$, $0 \le w_m < 1$): A risk-averse investor allocates a portion of their capital to risk-free sovereign bills ($w_f > 0$) and the remainder to the market portfolio ($w_m < 1$).
   • The Pure Market Portfolio ($w_m = 1.0$): 100% allocation to risky market assets.
   • Borrowing / Leveraged Portfolios (Right of $M$, $w_m > 1$): An aggressive investor borrows funds at the risk-free rate ($w_f < 0$) to invest more than 100% of their net equity into the market portfolio, magnifying expected return and risk along the linear CML.

---

Transitioning from CML to the Security Market Line (SML)

A critical distinction in investment analysis is the boundary condition of the Capital Market Line:

• Limitation of the CML: The CML measures risk using total risk (standard deviation $\sigma$). Because the CML represents combinations of the risk-free asset and the fully diversified market portfolio, it contains zero unsystematic risk. Therefore, the CML is valid only for pricing fully diversified, efficient portfolios.
• Individual stocks (such as Apple, BP, or HSBC) and undiversified portfolios contain substantial unsystematic risk. If plotted on the CML graph, they will always fall far below and to the right of the CML, because the market does not compensate investors for their individual volatility.

To price individual securities and undiversified portfolios, we must evaluate only the systematic risk that the security adds to a diversified market portfolio. This universal pricing relationship is defined by the Security Market Line (SML).

---

The CAPM Formula and Beta ($\beta$)

The formal mathematical equation for the Capital Asset Pricing Model expresses the required rate of return of any individual asset $i$ as:

$E(R_i) = R_f + \beta_i \left[ E(R_m) - R_f \right]$

Where:

• $E(R_i)$ = the required or expected return on asset $i$
• $R_f$ = the risk-free rate of return
• $E(R_m)$ = the expected return on the overall market portfolio
• $[E(R_m) - R_f]$ = the Equity Risk Premium (ERP) or Market Risk Premium, representing the extra return investors demand for investing in the risky market portfolio instead of risk-free cash
• $\beta_i$ = the Beta coefficient of asset $i$

Mathematical Definition and Derivation of Beta

Beta measures the sensitivity or responsiveness of an individual security's returns to movements in the overall market portfolio. Mathematically, beta is the slope coefficient of the linear regression of asset $i$'s excess returns against the market's excess returns (the Characteristic Line):

$\beta_i = \frac{\text{Cov}(R_i, R_m)}{\sigma_m^2}$

Substituting the correlation relationship $\text{Cov}(R_i, R_m) = \rho_{i,m} \sigma_i \sigma_m$ into the numerator gives the alternative operational formulation:

$\beta_i = \rho_{i,m} \left( \frac{\sigma_i}{\sigma_m} \right)$

Where:

• $\rho_{i,m}$ = the correlation between asset $i$ and the market portfolio
• $\sigma_i$ = standard deviation of asset $i$
• $\sigma_m$ = standard deviation of the market portfolio

This formulation reveals that beta is driven by two distinct factors: the asset's relative volatility compared to the market ($\sigma_i / \sigma_m$), and the degree to which its returns move in sync with the market ($\rho_{i,m}$).

Practical Interpretation of Beta

Worked Calculation Example: Required Return under CAPM

Suppose a wealth manager is appraising an equity investment under the following parameters:

• Risk-free rate ($R_f$): 3.5%
• Expected market return ($E(R_m)$): 9.5%
• Stock A Beta (Defensive Utility): $\beta_A = 0.70$
• Stock B Beta (Aggressive Technology): $\beta_B = 1.40$

1. Calculate the Equity Risk Premium (ERP): $\text{ERP} = E(R_m) - R_f = 9.5\% - 3.5\% = 6.0\%$

2. Calculate Stock A's Required Return: $E(R_A) = 3.5\% + (0.70 \times 6.0\%) = 3.5\% + 4.2\% = 7.7\%$

3. Calculate Stock B's Required Return: $E(R_B) = 3.5\% + (1.40 \times 6.0\%) = 3.5\% + 8.4\% = 11.9\%$

---

Security Valuation, Mispricing, and Jensen's Alpha ($\alpha$)

The Security Market Line serves as the benchmark against which wealth managers determine whether an individual security is fairly priced, undervalued, or overvalued in the secondary market.

Jensen's Alpha ($\alpha$)

Alpha measures the difference between an asset's actual expected return (forecasted by fundamental analysis) and the required return mandated by the CAPM formula for its level of systematic risk:

$\alpha_i = E(R_i)_{\text{forecast}} - \left( R_f + \beta_i \left[ E(R_m) - R_f \right] \right)$

• Positive Alpha ($\alpha > 0$): Undervalued Security (Buy Signal) When an analyst forecasts that a stock will deliver 13.0%, but the CAPM required return for its beta is only 10.5%, the stock offers a positive alpha of $+2.5%$. The security plots above the Security Market Line. Because it offers excess risk-adjusted return, it is undervalued by the market. Buying demand will push its current market price up until its expected future return falls back onto the SML.

• Negative Alpha ($\alpha < 0$): Overvalued Security (Sell Signal) When a stock's forecasted return is 7.0%, but its systematic risk requires an 9.0% return under CAPM, the stock has a negative alpha of $-2.0%$. The security plots below the Security Market Line. The asset fails to compensate the investor adequately for its systematic risk. Rational investors will sell or short the security, driving its price down until its expected return rises back onto the SML.

• Zero Alpha ($\alpha = 0$): Fairly Priced Security (Hold) The asset plots directly on the Security Market Line. The expected return exactly matches the equilibrium compensation for its systematic risk.

---

Core Assumptions of CAPM and Real-World Limitations

While CAPM is mathematically elegant, it relies on several restrictive theoretical assumptions. Wealth managers must understand these assumptions to assess the model's limitations in practice:

1. Zero Transaction Costs and Taxes: There are no brokerage commissions, bid-ask spreads, stamp duties, or capital gains taxes.
2. Infinitely Divisible Assets: Investors can purchase fractional shares of any asset.
3. Homogeneous Expectations: All market participants share identical forecasts regarding expected returns, variances, and covariances for all securities over a single, common holding period.
4. Unrestricted Borrowing and Lending at the Risk-Free Rate: Investors can borrow or lend unlimited capital at the exact same risk-free rate ($R_f$). In reality, private investors borrow at rates substantially higher than sovereign governments.
5. Mean-Variance Rationality: Investors evaluate portfolios solely on expected return and variance, assuming asset returns follow normal distributions (ignoring fat tails, skewness, and kurtosis).
6. Information Efficiency: All market participants have immediate, costless access to all public information.

Practical Challenges in Wealth Management

• Beta Instability: Beta is not a static corporate constant. Historical betas change over time due to corporate restructuring, acquisitions, shifts in debt leverage, and evolving competitive landscapes.
• The Roll Critique (1977): Richard Roll pointed out that the theoretical Market Portfolio must include all risky assets in the global economy—including non-public equities, commercial real estate, art, human capital, and private infrastructure. Because such a total market portfolio is unobservable and cannot be indexed, standard market proxies (like the S&P 500 or FTSE All-Share) are incomplete, rendering true empirical tests of CAPM invalid.

---

Arbitrage Pricing Theory (APT)

Stephen Ross proposed Arbitrage Pricing Theory in 1976 as a deliberately weaker, more general alternative to CAPM. Where CAPM insists that exactly one factor — exposure to the market portfolio — explains expected return, APT holds that asset returns are generated by a linear combination of several systematic risk factors, and that arbitrage will force any security's expected return onto that relationship:

$E(R_i) = R_f + \beta_{i,1}\lambda_1 + \beta_{i,2}\lambda_2 + \dots + \beta_{i,n}\lambda_n$

where each $\beta_{i,k}$ is the security's sensitivity to factor $k$ and each $\lambda_k$ is the risk premium the market pays for bearing one unit of that factor.

The arbitrage argument. If a security's price implies a return above or below the factor relationship, an investor can construct a zero-net-investment, zero-factor-exposure portfolio — long the mispriced asset, short a replicating basket — and capture a riskless profit. Because such trades are self-financing and scalable, they are executed until the mispricing disappears. APT therefore needs no assumption about investor utility, no mean-variance optimisation, and no observable market portfolio, which neatly sidesteps the Roll critique.

Common factor sets. APT does not name its own factors; the modeller must specify them. Typical macroeconomic implementations use unexpected changes in industrial production, inflation, the term-structure slope, and the corporate default spread. Commercial equity risk models add fundamental style factors — size, value, momentum, quality, and low volatility — which is why APT is the theoretical ancestor of modern multi-factor and "smart beta" investing.

For a wealth manager, the practical payoff is diagnostic: a fund whose excess return disappears once size, value and momentum exposures are priced in is selling factor beta at an alpha fee, and should be replaced with a cheaper systematic vehicle.