A stock's beta: what it measures, and what it leaves out

When you compare two stocks, you often read their beta next to the P/E ratio or the yield. The figure appears on every stock page, but it is rarely explained: some see it as a risk score, others as a promise of performance. It is neither. This article explains exactly what a stock's beta measures, how it is calculated, how to read it without over-interpreting it, and when it can mislead.

Stock beta: a simple definition

Beta measures the sensitivity of a stock's movements to those of a reference market, usually a broad index such as the CAC 40, the S&P 500 or a world index. It answers a precise question: when the market rises or falls by 1%, by how much has the stock moved on average over the period observed?

It is read around the value 1:

Beta close to 1: the stock has moved roughly in line with the market.
Beta above 1: it has amplified market moves, upwards and downwards.
Beta below 1: it has dampened them. So-called defensive stocks often show a low beta.
Beta close to 0 or negative: the stock has moved independently of the market, or even in the opposite direction. This is rare for a single stock.

Beta therefore describes past behaviour relative to an index, not the strength of the company or its risk of bankruptcy.

How is beta calculated?

Beta is the ratio between the covariance of the stock's returns with the market's returns and the variance of the market's returns:

Beta = covariance (stock return, market return) ÷ variance (market return)

In practice, you compute regular returns (weekly or monthly) over a given period, often 2 to 5 years, then estimate the slope of the line linking the stock's returns to the index's returns. That slope is the beta.

Two choices made by whoever does the calculation change the result:

The reference index. The beta of a French stock is not the same when measured against the CAC 40 or against a world index.
The period and frequency. 5 years of monthly data and 2 years of weekly data give different betas for the same stock.

A beta shown without its index and period is therefore only an order of magnitude. Two sources can legitimately display different values.

Worked example: what a beta of 1.3 means

Suppose a stock has a beta of 1.3, measured against its index, and the market rises by 10% over a period. The statistical relationship observed in the past suggests a rise of about 13% for the stock (1.3 × 10%), all else being equal and ignoring the part specific to the stock.

If the market falls by 10%, the same relationship suggests a fall of about 13%. Beta therefore works in both directions.

Two precautions, which are the essence of reading it properly:

It is an observed average relationship, not a forecast. A stock can deviate widely from it over a given period because of its results, an announcement or its sector.
Beta says nothing about the expected return of an individual stock. A high beta means neither a "better" nor a "worse" stock.

Beta and risk: the most common confusion

Beta measures only part of the risk, the part that is linked to market movements. It is called systematic risk. It does not measure company-specific risk: a profit warning, a lawsuit, a dependence on one customer. That specific risk does not appear in beta and can be significant even for a low-beta stock.

That is why total volatility and historical maximum losses are also examined. See how to read volatility and drawdown and the Sharpe ratio.

The limits of beta

It is calculated on the past. A company that changes its business model, takes on debt or refocuses on another sector may behave differently without its historical beta reflecting it.
It depends on the quality of the fit. When the relationship between the stock and the index is weak (a low coefficient of determination), beta is unreliable because it describes the stock's behaviour poorly.
It is not stable. It varies with the period, index and frequency chosen.
It says little about small, thinly traded stocks, whose prices move in jerks.

Portfolio beta

A portfolio's beta is the average of its positions' betas, weighted by their size. A portfolio with a weighted beta of 1.2 has, in the past, moved about 1.2 times as much as the market: roughly +12% for a market at +10%, and roughly −12% for a market at −10%. This reading complements the analysis of allocation: two portfolios with the same number of holdings can have very different sensitivities to the market. To work on allocation, see how to diversify a stock portfolio.

Key takeaways

Beta measures the past sensitivity of a stock to an index, nothing else.
A beta above 1 amplifies the market in both directions; below 1, it dampens it.
It measures neither company-specific risk nor the quality of an investment.
It depends on the index, period and frequency: compare betas calculated in the same way.

In Earnnest, the beta of several stocks can be read side by side in the stock comparison and in the sector tables.

This article is for information and education only and does not constitute personalised investment advice. The figures in the examples are fictitious and illustrative. Past performance is not a reliable indicator of future performance.