FNCE 3030 - Investment and Portfolio Management: CAPM in Practice

Course Information

  • Course: FNCE 3030

  • Topic: Investment and Portfolio Management

  • Focus: Capital Asset Pricing Model (CAPM) in Practice

  • Semester: Spring 2025

  • Instructor: ABIS

  • Institution: University of Colorado Boulder

Roadmap for CAPM in Practice

  1. Using CAPM to Identify Mispricing

    • Explore how CAPM can be applied to identify mispriced stocks.

  2. Multi-Factor Models of Stock Returns

    • Fundamental (value) and size strategies

    • Long-only and long-short portfolios

    • Multi-factor modeling:

      • Fama-French 3-factor model

      • Technical (momentum) strategies

      • Fama-French-Carhart 4-factor model

    • Critical assessment of the models and their implications.

    • Provide examples and case studies related to the content.

Understanding CAPM

  • CAPM predicts that the market portfolio should have the highest Sharpe ratio. All securities should lie on the Security Market Line (SML):

    • Equation: \mui - Rf = \betai(\mum - R_f)

    • Rearranging gives: \mui = Rf + \betai(\mum - R_f)

Key Implications of CAPM:
  1. Market beta ((\beta)) is positively related to expected return ((\mu)).

  2. Beta is the only determinant of expected return; other factors like size, growth, etc., are irrelevant.

    • Asserted Equation: \mui - Rf = \alphai + \betai(\mum - Rf)

  3. Stocks with a non-zero CAPM alpha can help improve the market's Sharpe ratio.

  4. Anomalies are identified as stocks that deviate from model predictions. Specifically, CAPM anomalies are stocks that exhibit non-zero alphas.

Testing CAPM and Identifying Mispricing

  • Alpha vs. Anomalies: Both alpha and anomalies represent portfolio strategies yielding returns that differ systematically from CAPM predictions.

    • Testing CAPM traditionally involves running regressions to check for significant alpha, which can be problematic due to statistical errors and regression assumptions.

    • A more robust testing method involves identifying additional variables or portfolios beyond the market that predict stock returns.

Multi-Factor Models of Stock Returns

  • Fama-French Model (1992) identifies characteristics beyond beta predicting stock returns:

    • Size: Market Capitalization (ME)

    • Definition: ME = Market\,Equity = Number\,of\,Shares \times Share\,Price

    • Value: Book-to-Market Ratio (BM)

    • Definition: BM = \frac{BE}{ME} where BE = Book Equity.

    • MV should ALWAYS be greater than MV

    • if BV is MUCH less than MV we call it a “growth” company

    • if BV is a little less than MV we call it a “value” company

    • Value firms are much easier to value

  • Findings:

    • Small firms and value firms consistently outperform larger and growth firms when controlling for CAPM beta.

    • Value is a particularly strong indicator of performance.

Definitions of Stock Types:
  • Value Stocks: Stocks undervalued by the market, usually sell at low prices relative to earnings or book value and provide above-average dividends.

  • Growth Stocks: Shares of companies with high growth prospects, typically trading at high prices relative to earnings or book value and paying little to no dividends.

Constructing Factor Portfolios
  • Fama and French build portfolios on size and book-to-market in their 1992-1993 studies.

  • Portfolios are formed each July and held until June the following year.

  • Analysis is based on performance of these portfolios (5x5 methodology).

Value Effect Analysis

  • The cumulative gains from investments between 1926 to 2022 demonstrate significant differences in returns across portfolio strategies.

  • Examples:

    • Market (9.9% annual return) vs Small Value (16.0% annual return).

    • These statistics underscore the importance of incorporating value and size in investment strategies.

Long-Only and Long-Short Portfolios

  • Long-Only Portfolios: Invest entirely in assets, weighted to sum to one. Example calculation of return:

    • Given investments in shares yielding specific profits, calculate gross and net returns.

  • Long-Short Portfolios: Involves both long positions and shorts, where weights can sum to zero. Formulas include:

    • R{LS} = R{Long} - R_{Short}

    • Calculate alphas and betas for both long and short positions:

    • \alpha{LS} = \alpha{Long} - \alpha_{Short}

    • \beta{LS} = \beta{Long} - \beta_{Short}

Example of a Long-Short Portfolio
  • Given risk-free rate at 1% and expected betas and returns for companies:

    • Calculation of expected return for a long-short portfolio consists of using excess returns and individual stock calculations, ultimately leading to an overall expected return based on both stocks' performances including alphas.

Building Multi-Factor Portfolios Using Value & Size

  • Construction involves strategic positioning across various assets based on historical performance.

  • Implement a diversified strategy, including high book-to-price ratio stocks long and low book-to-price ratio stocks short.

  • Portfolio returns take into account market exposure and specific investment strategies (value vs. growth).

Movement Towards Four-Factor Models

  • The inclusion of momentum (UMD) is important in extending the Fama-French model, leading to a four-factor approach:

    • UMD targets the potential upward and downward movement of assets based on prior performances.

  • Jegadeesh and Titman (1993) developed momentum investment strategies through systematic ranking and shorting practices based on stock performance.

Cumulative Returns in Multi-Factor Models

  • Chart data indicating cumulative returns from different strategic portfolios from July 1926 to September 2022 highlights the substantial returns from multi-factor models including momentum elements.

  • Momentum (UMD) demonstrated a clear positive impact on Sharpe ratio improvements, raising it from 0.525 to 0.955.

Final Thoughts on Multi-Factor Models

  • Future efficacy of these strategies remains in question due to potential crowding effects diminishing returns as more investors enter these trades.

  • Recent analyses suggest that while fundamental strategies are well-known, their performance has decreased post-2008.

  • Continued evolution and adaptation of models remain essential to stay ahead in investment strategy effectiveness and maximizing portfolio returns.