On This Page
Key Takeaways
- Beta measures how much a security moves relative to the overall market, calculated as the covariance of its returns with the market divided by the variance of the market.
- Positive beta measures same-direction market sensitivity; negative beta measures opposite-direction sensitivity. Total volatility is the stock’s own measure, and the example’s 1.2 beta comes with volatility 1.5 times the market’s.
- Beta is the systematic-risk input in the Capital Asset Pricing Model, where expected return equals the risk-free rate plus beta times the market risk premium.
- Compare beta with market volatility and the regression’s residual variation to separate modeled market exposure from other return movements.
- Beta is an estimate from a chosen index, lookback window, and return frequency, so its value shifts with those inputs and over time.
What is beta?
Beta is a measure of how strongly a security’s returns move with a selected market benchmark. It equals the covariance of the security’s returns with the benchmark divided by the benchmark’s return variance. A beta of 1.2 means the fitted market component moves 1.2 percentage points for each one-point change in benchmark return.
The benchmark has beta 1 against itself. A different security at beta 1 can have its own intercept and residual returns. Negative beta represents opposite-direction market sensitivity.
How is beta calculated?
Beta is also the slope in a linear return regression with an intercept. The same value equals correlation times the security’s volatility divided by the benchmark’s volatility, measured on the same return series.
The example stock has 30% annualized return volatility, the market has 20%, and their correlation is 0.80. Beta is 0.80 × 30% ÷ 20% = 1.2.
Formula
Beta = covariance(asset return, benchmark return) ÷ variance(benchmark return) = correlation × asset standard deviation ÷ benchmark standard deviation; CAPM expected return = Rf + beta × [E(Rm) − Rf]- Covariance and variance are measured over a chosen return series, such as daily returns across the past two or three years.
- A benchmark has beta 1 against itself, and a different security with beta 1 can still have its own intercept and residual returns.
- The market risk premium is the expected market return minus the risk-free rate.
Worked Example
Estimate beta from matched daily stock and S&P 500 returns. Their annualized standard deviations are 30% and 20%, with correlation 0.80 under the same scaling convention. Use a 4% annual risk-free rate and 10% annual expected market return for the CAPM example.
- Beta: 0.8 × (0.30 / 0.20) = 1.2
- Market risk premium: 10% − 4% = 6%
- CAPM expected return: 4% + 1.2 × 6% = 11.2%
Beta is 1.2. The fitted market component changes by 1.2 percentage points for a one-point benchmark-return change. The stock’s total volatility is 30%, or 50% above the market’s 20%, while CAPM gives an 11.2% expected return from the stated inputs.
Beta vs volatility
The fitted market component changes 1.2 percentage points per one-point benchmark-return change. Total stock volatility is 50% above market volatility, a different comparison from beta being 20% above one.
For the matched one-market regression, the squared correlation is 0.64. The market component accounts for 64% of return variance, and the remaining residual standard deviation is 18% on the same annualized scale. Residual variation can include other common factors as well as company-specific movements.
Beta and expected return in CAPM
Under CAPM, expected return is the risk-free rate plus beta times the expected market premium. With 4% risk-free return and a 10% expected market return, the premium is 6% and the model output is 11.2%.
The 11.2% is the return CAPM requires for a 1.2 beta, and realized returns are measured against it after the period. In performance attribution, the fitted intercept and residuals describe the return left outside the factor relationship.
Portfolio beta and market exposure
For fixed portfolio weights on a common return basis, portfolio beta is the weighted sum of position betas. Include signed weights for short positions and the chosen treatment of cash and financing.
Compare that market exposure with other factors and the actual position payoff. A hedge targets a named exposure; assess the combined positions after adding it. Size a market hedge from beta: a 1.2-beta position needs short benchmark exposure of 1.2 times its value to neutralize the market component.
Limitations of beta
- Match the index, lookback window, return frequency and data adjustments before comparing two betas; each one changes the estimate.
- Re-estimate beta when the company, its leverage or market conditions change, and resize any hedge built on the old number.
- Hedge the residual separately. Beta prices linear co-movement with one benchmark, and in the example the market explains 64% of return variance, leaving the rest to other factors and company news.
- Check total volatility beside a low beta. A volatile stock with low correlation to the benchmark can have a small beta and still swing widely.
- Test sensitivity on large-move days separately. One slope for the whole window can misstate it, because the relationship during those moves can differ from the full-sample average.
Bottom line
Beta measures how strongly a security’s returns move with a chosen benchmark. Use it to size and hedge market exposure across a portfolio, and read it with total volatility, the residual, and the index, window and frequency that produced it.
Clear answers
Frequently Asked Questions
Sources and Further Reading
- CFA Institute: Portfolio Risk and Return, Part II
- CFA Institute: The Capital Asset Pricing Model, Theory and Evidence
How entries are sourced, calculated and updated: editorial methodology.