Skip to content

New to Omni?·Get 20% off your first purchase with code WELCOME20 at checkout.See plans

FRM® beta: why can a volatile asset have low market risk?

Beta measures co-movement with a market benchmark, not total volatility. Work through covariance, correlation and portfolio beta without mixing their units.

ConceptOmni Curriculum Team6 min read

Two unlabelled cords trace related curves across white paper beside a ruler on a navy desk.
On this page

Beta measures the sensitivity of an asset’s return to a chosen market return, not the asset’s total volatility. It is the covariance of the two return series divided by the market return variance. An asset can fluctuate sharply yet have a low beta when those fluctuations do not consistently coincide with market moves. That distinction is the central calculation and interpretation a new FRM candidate needs.

Read the denominator before interpreting the numerator

Covariance records whether two return series tend to be above or below their own averages together. Positive covariance indicates that their deviations commonly have the same sign; negative covariance indicates opposite signs. Its raw size depends on the scale of both series. Dividing by market variance changes that joint movement into a slope: the estimated change in the asset return associated with a one-unit change in the market return in a simple regression. The CFA Institute performance-measurement guidance states this covariance-over-benchmark-variance definition explicitly.

βi=Cov⁡(Ri,Rm)Var⁡(Rm)\beta_i=\frac{\operatorname{Cov}(R_i,R_m)}{\operatorname{Var}(R_m)}
Beta of asset i: covariance of asset and market returns divided by the variance of market returns.

The denominator is variance, not standard deviation. If the market standard deviation is supplied, square it before dividing. Keep the units consistent: returns written as decimals must be used as decimals throughout the covariance calculation, and returns written as percentage points require covariance and variance in percentage-point-squared units. Dividing a decimal covariance by a percentage-point variance can create an answer that looks precise but is off by a large factor.

Beta is benchmark-relative. A number computed against a broad equity index does not automatically describe exposure to a bond index or to every possible notion of “the market”. The same asset can have different estimated betas when the benchmark, sample period, return frequency or regression method changes. In an exam problem, use the market proxy specified by the question and do not silently swap in another. In real analysis, label the benchmark and estimation window beside the beta estimate.

A worked example with consistent units

Use invented annual return statistics for a hypothetical asset and a market proxy. The asset return standard deviation is 30%, the market standard deviation is 20%, and their correlation is 0.40. These figures are deliberately simple and are not observations from any security. First compute covariance as correlation multiplied by both standard deviations: 0.40 times 0.30 times 0.20 equals 0.024. Next compute market variance: 0.20 squared equals 0.040. The beta is therefore 0.024 divided by 0.040, or 0.60.

βi=ρi,mσiσm\beta_i=\rho_{i,m}\frac{\sigma_i}{\sigma_m}
Equivalent form: correlation between asset and market returns times asset volatility divided by market volatility.

The correlation form gives the same result directly: 0.40 times 0.30 divided by 0.20 equals 0.60. The asset has more standalone volatility than the market, yet its beta is below one because much of its movement is not shared with that market proxy. This does not make the asset safe or predictable. It says only that its average linear sensitivity to that benchmark, under the assumed statistics, is lower than the benchmark’s own sensitivity to itself.

To see why correlation matters, keep both volatilities unchanged but change the invented correlation to 0.80. Beta becomes 0.80 times 1.5, or 1.20. The asset’s total standard deviation has not moved at all. The market-linked component has doubled because more of its variation now lines up with market variation. A candidate who equates beta with standalone volatility cannot explain this change; the covariance formula can.

What beta does and does not predict

A beta of 0.60 is a slope fitted to the assumed relationship. It is tempting to say that a one percent market rise guarantees a 0.60 percent asset rise, but the formula does not justify that claim. Actual returns include an intercept and a residual; the latter can be large for a single asset. Beta summarises average linear co-movement over a defined sample or model. It neither bounds losses nor describes every individual market day. A market decline could coincide with an asset rise, or with a much larger fall than its beta alone suggests.

Within the capital asset pricing model, beta represents systematic risk, the portion connected to the market factor. CFA Institute’s portfolio risk reading distinguishes that from asset-specific risk that can be diversified. This model interpretation is why beta enters the CAPM expected-return relation. The present question, however, is about calculating and reading beta correctly, not using it to decide whether an observed return meets a required-return line. That latter decision starts only after the beta input is understood.

The market portfolio has beta one against itself because its covariance with itself equals its variance. A genuinely risk-free return has zero covariance with variable market returns and therefore beta zero in the model. A negative beta indicates an inverse average relationship with the specified market proxy, but it does not promise a positive return whenever the market falls. These are reference points for interpretation, not categories that tell a reader which holding to choose.

Portfolio beta follows the weights

A portfolio return is a weighted sum of component returns, so its covariance with a given market is the same weighted sum of the component covariances. Dividing that sum by the same market variance gives a weighted average of component betas. This is a useful shortcut when the question asks for a portfolio’s market sensitivity rather than its total variance. It does not mean that portfolio standard deviation is a weighted average of component standard deviations; total variance depends on the correlations among the holdings.

βp=∑iwiβi\beta_p=\sum_i w_i\beta_i
Portfolio beta is the sum of each position weight multiplied by its beta against the same market proxy.

For another invented check, suppose a portfolio is half in an asset with beta 0.60, one quarter in an asset with beta 1.20, and one quarter in a position with beta zero against the same market. The weighted contributions are 0.30, 0.30 and zero, giving portfolio beta 0.60. The position with zero beta may still fluctuate for reasons unrelated to that market. The calculation also assumes the weights are current and that all component betas were estimated on a comparable basis.

This weighted-average result provides a useful error check. If all portfolio weights are positive and add to one, and every component beta lies between zero and one, the portfolio beta must also lie in that interval. A calculation outside it signals a weight, sign or arithmetic error. If short positions or borrowing are present, that simple interval check no longer applies, because a negative or greater-than-one weight can pull the total outside the range of the component betas. Read the portfolio description before using the shortcut mechanically. It is an identity for the specified weights and benchmark, not a statement that the portfolio is diversified or that its future loss is limited.

A short exam checklist

When the input is covariance and market standard deviation, square the market standard deviation. When the input is correlation and two standard deviations, multiply correlation by the ratio of asset to market standard deviation. Check that the sign of beta matches the sign of correlation because standard deviations are non-negative. Check that the answer is dimensionless. A result described as “percent beta” is a warning that the units have not been handled carefully.

Then interpret the number narrowly: identify the benchmark, distinguish relative market sensitivity from total asset volatility, and describe the estimate as an average relationship rather than an exact next-period outcome. If a question gives several assets, keep the same benchmark across them before comparing betas. GARP places foundations of risk management and quantitative analysis in FRM Part I, as its exam FAQ explains. The FRM programme page gives the broader route, while the how-it-works page can help organise practice around the two equivalent calculations.

Sources

  1. Investment Performance Measurement | CFA Institute · rpc.cfainstitute.org · Retrieved
  2. Portfolio Risk and Return Part II | CFA Institute · cfainstitute.org · Retrieved
  3. FRM Exam FAQs | GARP · garp.org · Retrieved

Explore FRM study resources

Educational content, not investment advice. This article is written to help candidates prepare for professional exams. It is not a recommendation to buy, sell or hold any security, and it does not take your personal circumstances into account.

Omni Finance Academy is an independent exam-preparation provider and is not affiliated with or endorsed by CFA Institute, GARP, CAIA Association or CFP Board. CFA®, FRM®, CAIA® and CFP® are trademarks of their respective owners. Full disclaimer.