The CAPM, developed by William Sharpe (1964), John Lintner (1965), and Fischer Black (1972), is the foundational asset pricing model. It predicts that the expected excess return on any security is proportional to its market beta:
Key assumptions
- Investors are mean-variance optimizers (Markowitz 1959)
- A single-period investment horizon
- Homogeneous expectations
- Frictionless markets (no taxes, no transaction costs)
- Unlimited borrowing and lending at the risk-free rate
Central prediction
The market portfolio of all invested wealth is mean-variance efficient. This implies:
- Expected returns are a positive linear function of market beta
- Market beta is sufficient to describe the cross-section of expected returns
Empirical challenges
Early tests (Black, Jensen, and Scholes 1972; Fama and MacBeth 1973) found a positive beta-return relation in pre-1969 data (t=2.57), supporting the model. BJS also found that the intercept exceeded the risk-free rate and the SML was flatter than predicted, consistent with Black’s (1972) zero-beta CAPM and foreshadowing the betting-against-beta literature. However:
- Fama and French (1992) showed the beta-return relation is flat in 1963-1990 data. The regression slope of returns on beta is 0.15%/month (t=0.46).
- Size and book-to-market capture the cross-sectional variation that CAPM attributes to beta.
- Anomalies including momentum, profitability, and investment patterns are unexplained by a single-beta model.
Legacy
Despite its empirical failures, the CAPM remains foundational:
- It established the framework of systematic vs. idiosyncratic risk
- Its market factor (RM-RF) is the first factor in all subsequent multi-factor models
- It motivates the question that all later models try to answer: what are the priced sources of risk?
The CAPM’s empirical failure led directly to the development of the Fama-French three-factor model, the Carhart four-factor model, and the Fama-French five-factor model. An alternative theoretical response was Ross’s Arbitrage Pricing Theory, which derived multi-factor pricing from no-arbitrage conditions without requiring CAPM’s restrictive assumptions. The Betting Against Beta literature (Frazzini and Pedersen 2014) explains the flat SML through leverage constraints rather than model misspecification.
Sources
- Capital Asset Prices: A Theory of Market Equilibrium under Conditions of Risk (File, DOI)
- The Cross-Section of Expected Stock Returns (File, DOI)
- Risk, Return, and Equilibrium: Empirical Tests (File, DOI)
- Betting Against Beta (File, DOI)
- The Arbitrage Theory of Capital Asset Pricing (File, DOI)
- The Capital Asset Pricing Model: Some Empirical Tests (File)