Capital Asset Pricing Model: Beta, Market Risk Premium, and Expected Return

What is the Capital Asset Pricing Model?

The Capital Asset Pricing Model, usually abbreviated CAPM, is an equilibrium model that relates an asset's expected return to its exposure to systematic market risk.

The standard equation is:

1E(Ri) = Rf + βi[E(Rm) - Rf]

where:

1E(Ri) = expected return of asset i
2Rf    = risk-free rate
3βi    = beta of asset i relative to the market
4E(Rm) = expected return of the market portfolio

The term:

1E(Rm) - Rf

is the Market Risk Premium.

CAPM therefore says that an asset's expected return equals the risk-free rate plus compensation for the amount of market risk measured by Beta.

CAPM prices systematic risk, not total volatility

One of CAPM's central ideas is that investors can diversify away company-specific risk.

A stock can be volatile because of:

1market-wide forces
2+
3company-specific forces

CAPM separates those into Systematic Risk and Nonsystematic Risk.

Systematic risk cannot be eliminated merely by holding more securities because it affects broad markets or common economic factors.

Nonsystematic risk is local to a company, industry, project, or other narrow exposure and can be reduced through diversification.

Under CAPM, investors are compensated for bearing systematic risk, not for bearing avoidable nonsystematic risk.

That is why beta, rather than total standard deviation, appears in the expected-return equation.

A simple CAPM example

Suppose:

1Risk-free rate      = 4%
2Market risk premium = 6%
3Stock beta          = 1.3

Then CAPM expected return is:

1E(Ri) = 4% + 1.3 × 6%
2      = 11.8%

The model is not saying the stock will earn exactly 11.8% next year.

It is saying that under the selected CAPM inputs, an asset with beta 1.3 would require an expected return of 11.8% in equilibrium.

The realized return could be much higher or lower.

What beta contributes to the model

Beta measures an asset's sensitivity to movements in the selected market benchmark.

Conceptually:

1βi = Cov(Ri, Rm) / Var(Rm)

A beta of:

11.0 -> market-like systematic sensitivity
21.5 -> greater sensitivity to market movements
30.5 -> lower sensitivity to market movements

A negative beta would indicate modeled movement in the opposite direction from the market, though stable negative betas are uncommon for ordinary equities.

Beta is estimated from data and depends on the benchmark, return frequency, sample window, and statistical method.

There is no universal timeless beta for a security.

The market risk premium is also an estimate

CAPM needs an expected market excess return:

1Market Risk Premium = E(Rm) - Rf

That quantity is not directly observable before the future occurs.

Analysts may estimate it from historical returns, surveys, valuation models, or other forward-looking methods.

Different methods can produce materially different values.

Because both beta and the market risk premium are estimated, a CAPM expected return can look precise while resting on uncertain inputs.

A result such as 9.73% should not be interpreted as known to two decimal places merely because a spreadsheet can calculate it that way.

CAPM and the Security Market Line

The Security Market Line is the graphical representation of the CAPM equation.

It plots:

1horizontal axis -> beta
2vertical axis   -> expected return

The intercept is the risk-free rate.

The slope is the market risk premium.

A beta of zero maps to the risk-free rate in the basic model, and a beta of one maps to the expected market return.

The SML applies to individual securities and portfolios under CAPM because it relates expected return to systematic risk.

CAPM and the Capital Market Line are not the same

The Capital Market Line also comes from CAPM, but it uses a different risk measure and applies to a different set of portfolios.

The CML plots:

1expected return versus total volatility

for efficient combinations of the risk-free asset and market portfolio.

The SML plots:

1expected return versus beta

for any asset or portfolio under CAPM.

An individual stock with large company-specific volatility may sit on the SML while not lying on the CML.

Confusing the two lines can lead to the mistaken belief that total volatility is always the priced risk in CAPM.

The market portfolio is theoretical

CAPM assumes investors collectively hold a market portfolio containing all risky assets in market-value proportions.

In practice, analysts use proxies such as broad equity indexes.

That simplification matters.

A US large-cap index, total-market index, and global index can produce different betas and market risk premiums.

The true theoretical market portfolio is not directly observable and would be broader than a single equity index.

This is one reason empirical CAPM testing depends heavily on benchmark choice.

CAPM is an equilibrium model

CAPM is not primarily a short-term forecasting rule.

It describes a relationship that should hold in equilibrium under its assumptions.

If investors can diversify and all hold the same efficient risky portfolio, only non-diversifiable market risk should command an expected return premium.

The model therefore links price, expected return, and beta.

If an asset's price falls while its expected future cash flows are unchanged, its expected return rises. In equilibrium, prices adjust so expected returns line up with systematic risk.

This theoretical logic is different from saying beta mechanically causes realized returns.

Core CAPM assumptions

Textbook CAPM relies on strong simplifying assumptions.

Common versions assume:

• investors are mean-variance optimizers;
• investors share homogeneous expectations about returns, volatility, and correlations;
• there is a common investment horizon;
• markets are frictionless, with no taxes or transaction costs;
• securities are divisible and tradable;
• investors can borrow and lend at the risk-free rate; and
• investors can diversify broadly across risky assets.

Real markets violate many of these assumptions.

That does not make CAPM useless. It means the model is a benchmark, not an objective law of nature.

CAPM expected return is not realized return

Suppose CAPM assigns a 10% expected return to a stock.

The stock can still realize:

1+40%
2-25%
3+3%

in a particular year.

Expected return is a statistical or equilibrium concept about a distribution of possible outcomes, not a promised one-period result.

Likewise, a stock that earns more than its CAPM expected return during one period has not automatically demonstrated persistent skill or mispricing.

Random outcomes, changing beta, changing risk premiums, unexpected information, and model error can all matter.

CAPM and alpha

Alpha is often defined relative to a benchmark model.

A simple CAPM-style alpha compares realized or expected return with the return implied by beta exposure.

Conceptually:

1Alpha = Actual or modeled return - CAPM benchmark return

A positive historical alpha does not by itself prove manager skill.

The estimate depends on the sample, benchmark, risk-free rate, beta specification, fees, and whether other systematic factors were omitted.

A strategy may appear to generate CAPM alpha simply because CAPM fails to capture another persistent factor exposure.

CAPM versus factor models

CAPM is effectively a one-factor asset-pricing model centered on market beta.

A Factor Model can include multiple sources of systematic variation or expected return.

Examples of factor families include:

1market
2value
3growth
4size
5momentum
6quality
7interest rates
8inflation

CFA Institute notes that multifactor models are widely used because they can provide more granular risk and return attribution than a single market factor.

More factors do not automatically make a model true. They introduce additional choices about factor definition, estimation, covariance, premiums, and stability.

CAPM in capital budgeting

CAPM is also used outside portfolio selection.

A company or analyst may estimate a project's cost of equity by combining:

1risk-free rate
2+
3project or comparable-company beta × market risk premium

That expected return can help form a discount rate for valuation or capital budgeting.

This application requires careful treatment of leverage, business risk, country exposure, and whether the chosen beta is appropriate for the project.

A mechanically copied public-company beta is not necessarily the correct discount rate for every project.

Historical beta can be unstable

Beta estimates change through time because business mix, financial leverage, investor behavior, and market regimes change.

A company can become more cyclical after an acquisition or less leveraged after paying down debt.

A short sample can be dominated by one crisis.

A long sample can include an outdated business model.

Regression beta also contains estimation noise.

Analysts sometimes adjust betas toward one or use industry averages, but those methods introduce additional assumptions.

The risk-free rate is not a trivial input

CAPM often uses a government security yield as a risk-free proxy.

The appropriate maturity and currency should be consistent with the analysis.

Using a three-month bill rate for a long-duration valuation can create a mismatch.

Using a US dollar Treasury yield for cash flows fundamentally exposed to another currency can also be problematic.

The risk-free rate is a model input that should be defined, not a universal constant.

What CAPM cannot tell you

CAPM does not guarantee future returns.

It does not identify a stock's true intrinsic value by itself.

It does not make beta stable.

It does not make the market risk premium directly observable.

It does not prove that one broad equity index is the true market portfolio.

It does not capture every systematic source of risk.

It does not eliminate estimation error or model risk.

CAPM remains valuable because it provides a simple, coherent benchmark for thinking about diversification, systematic risk, beta, expected return, and the opportunity cost of capital.

Grizzly Bulls' Models can provide broader systematic-research context, while the Macroeconomic Conditions Index can frame changing market regimes. Neither route publishes a canonical live CAPM expected return, current market risk premium, or security-specific fair-value recommendation.

Sources and further reading

• CFA Institute: Portfolio Risk and Return: Part II, 2026 curriculum
• CFA Institute: Using Multifactor Models, 2026 curriculum
• CFA Institute Research & Policy Center: The Equity Risk Premium

Explore more topics in the Financial Research Encyclopedia.