The problem with CAPM

The Capital Asset Pricing Model (CAPM) is one of the most widely used and taught frameworks in finance, offering a simplified approach to estimate the cost of equity. However, its application in biotech and pharma is deeply flawed. CAPM relies on assumptions of market efficiency, normal distributions, and systematic risk as the sole driver of returns. These assumptions clash with the unique characteristics of biotech and pharma, where value creation hinges on unpredictable, binary outcomes, and firm-specific risks dominate. Moreover, CAPM’s relevance is confined to public markets, further limiting its utility in a sector where private funding and non-traded assets play critical roles.

The CAPM Formula

At its core, CAPM calculates the expected return on equity as:

$$
r_e = r_f + \beta \cdot (r_m - r_f)
$$

Where:

• re: Expected return on equity
• rf​: Risk-free rate
• rm​: Expected return of the market
• β: Sensitivity of the stock to systematic market risk

This model assumes that the risk of an asset can be distilled into a single variable, β, representing its correlation with the overall market. While this approach has theoretical appeal, it fails to capture the complexities of biotech and pharma.

Problems with CAPM Components

1. Beta (β) and the Illusion of Systematic Risk

Beta is the cornerstone of CAPM, representing a stock's systematic risk relative to the broader market. However, beta is inherently flawed in biotech and pharma:

• Unstable and Non-Stationary:
  Beta is backward-looking, calculated using historical data. In biotech, where valuations hinge on unpredictable, binary events (e.g., trial outcomes, FDA decisions), past performance provides little insight into future risk. Beta evolves over time as firms transition through development phases, from high-risk R&D to lower-risk commercialization. CAPM fails to account for this dynamic risk profile.
• Non-Linear Risk Profiles:
  Biotech outcomes are highly non-linear; the value of a drug candidate increases exponentially with clinical success. Beta, based on linear regression, cannot capture this convexity, underestimating upside potential and downside risk.
• Idiosyncratic Risk Misclassification:
  Many biotech risks, such as specific trial failures or patent disputes, are idiosyncratic (firm-specific) rather than systematic (market-wide). CAPM ignores these risks, despite their outsized impact on equity valuation.

2. Market Risk Premium (rm−rf) and Biotech’s Unique Risk Landscape

CAPM assumes a uniform market risk premium (MRP) across industries. This assumption falters in biotech:

• Excess Risk Premium:
  Investors demand higher returns to compensate for biotech’s extreme uncertainties, including long R&D cycles, high attrition rates, and regulatory hurdles. A generic MRP cannot capture these elevated risk expectations.
• Fat-Tailed Distributions:
  Biotech returns often exhibit heavy tails, where extreme gains or losses dominate the distribution. CAPM’s reliance on Gaussian assumptions underestimates the likelihood of such outcomes, skewing MRP calculations.

3. Risk-Free Rate (rf​) and the Misalignment of Time Horizons

Biotech projects often span decades, far exceeding the maturity of commonly used risk-free benchmarks, such as 10-year government bonds. This mismatch introduces errors:

• Term Structure of Risk:
  Longer time horizons amplify uncertainty, yet CAPM assumes the risk-free rate remains constant across all projects.
• Global Disparities:
  Biotech firms operating across multiple geographies face diverse regulatory environments and currency risks, making a single rfr_frf​ inadequate.

Ignoring Tail Risks and Extreme Outcomes

CAPM operates under the assumption of normally distributed returns, where risks are captured by the first two moments (mean and variance). However, biotech’s fat-tailed distributions defy these assumptions. For tail risk modeling, the probability density function can be expressed as:

$$
f(x; x_m, \alpha) = \frac{\alpha \cdot x_m^\alpha}{x^{\alpha+1}}
$$

Where:

• xm​: Minimum return
• α: Shape parameter (tail heaviness)

This leads to:

• Underestimation of Catastrophic Risks:
  Trial failures or regulatory rejections can wipe out significant value, events poorly reflected in CAPM’s symmetric risk framework.
• Overlooking Rare, High-Impact Successes:
  Blockbuster drugs that generate billions in revenue disproportionately impact firm valuation. CAPM’s Gaussian assumptions fail to account for these outliers.

Conclusion

The Capital Asset Pricing Model, while foundational in finance, is poorly equipped to handle the unique challenges of biotech and pharma. Its reliance on systematic risk, Gaussian assumptions, and market efficiency clashes with an industry dominated by idiosyncratic risks, fat-tailed distributions, and non-linear outcomes. Moreover, CAPM’s focus on public markets renders it irrelevant for the private funding mechanisms that drive much of biotech’s innovation. While it persists due to its simplicity and historical legacy, CAPM is a blunt tool for an industry defined by complexity, uncertainty, and extreme outcomes.