How to Calculate Portfolio Beta
To calculate portfolio beta, multiply each holding's beta by its share of the portfolio and add the results. The figure estimates how the portfolio has moved relative to a benchmark, and it says nothing about the risks that are specific to your holdings.
Portfolio beta is the weighted average of your holdings’ betas. The arithmetic is one multiplication per holding and one addition. The interpretation is narrower than the figure’s popularity suggests.
Before you start
Each holding’s beta against a stated benchmark, since the figure is benchmark-specific. The same holding has different betas against different indices.
The weight of each holding as a share of the total portfolio. Current market value, not what you paid.
An understanding that beta measures one relationship and ignores everything else. It is market sensitivity, not risk in general.
The steps
1. Fix the benchmark first
Every beta figure is relative to an index. Mixing betas computed against different benchmarks produces a weighted average of incompatible numbers.
2. Compute each holding’s weight
Current market value of the holding divided by the portfolio total. The weights sum to one, which is a useful check that nothing has been missed.
3. Multiply each beta by its weight
A holding at 30% of the portfolio with a beta of 1.4 contributes 0.42. Do that for every holding including cash, which contributes zero.
4. Add the contributions
The total is the portfolio beta. Above one means it has historically moved more than the benchmark; below one, less.
5. Read it as an estimate of market sensitivity only
It says nothing about concentration, liquidity, or anything specific to your holdings. A beta of 1.0 is consistent with a portfolio far riskier than the index.
6. Recompute it when weights drift
Weights change as prices move, so the portfolio beta changes without you doing anything. Quarterly is usually enough unless the portfolio is concentrated.
7. Check it against a correlation grid
Beta tells you sensitivity to one index. A correlation grid across your holdings tells you whether they are the same bet, which beta cannot show.
How to tell it worked
1 benchmark was used for every beta in the calculation.
The weights sum to 1, which confirms nothing was omitted.
Cash was included at a beta of 0, rather than left out.
And the figure was recomputed within the last 90 days.
What beta does not measure
Concentration. Five holdings in one sector can average to a beta near the market’s while carrying a risk profile nothing like it.
Anything specific to the holding. A company-level event, a liquidity problem, a regulatory change — none of it appears in a measure of co-movement with an index.
Why it changes on its own
Weights drift with prices. A holding that doubles becomes a larger share of the portfolio, so its beta contributes more, and the portfolio figure rises without a single transaction.
And the individual betas are rolling estimates. They are computed over a window of history, so each one moves as that window advances and older periods drop out.
Which makes it a monitoring figure rather than a fixed property. Computed once and treated as permanent, it describes a portfolio you had at some point in the past — which on this site’s shared series, where 95% of bars sat below a prior peak, is a portfolio whose weights have already moved.
A worked example
Four holdings and some cash. Holding A is 30% of the portfolio with a beta of 1.4. B is 25% at 0.9. C is 20% at 1.1. D is 15% at 0.6. Cash is the remaining 10% at 0.
Multiply each pair. 0.42, then 0.225, then 0.22, then 0.09, then 0. Five numbers, one per line.
Add them: 0.955. The portfolio has historically moved slightly less than the benchmark, which is mostly the cash position doing the work rather than any of the holdings being defensive.
Now change one thing. If holding A doubles in price while everything else stays flat, its weight rises to roughly 46% and the portfolio beta moves to about 1.09 — a meaningful change produced entirely by a price move, with no decision taken by anybody.
What to do with the number
Compare it against your intended market exposure. If you wanted something defensive and the figure is 1.3, that is a specific, actionable mismatch.
Use it to size a hedge if you are hedging the index. The beta tells you roughly how much index exposure the portfolio carries, which is the quantity a broad hedge would need to offset.
And use it as a drift alarm. A portfolio beta that has moved substantially without any trading is telling you the weights have changed, which is usually the first sign that a rebalance is due.
The original data
Of the 24,971 unique videos in research/search-study-corpus.jsonl, 4 mention beta in the title, at a
median of just 313 views across 4 channels. Correlation appears in 5 at 1,162 and diversification in
3 at 487. The counts come from site/corpus_count.py and site/rank_investing.py.
4 videos at a 313 median — the smallest audience per video of anything measured on this site. The portfolio-risk group as a whole has twelve videos across 25,000, and beta is the least watched of them.
The answer to the question on that chart is that 0.95 describes sensitivity to one index and nothing else. Check the correlation grid before concluding anything about risk — five holdings that all move together can produce that figure while carrying concentration the beta cannot see.
When it fails
The failure is treating it as a risk summary, and it produces false comfort in a concentrated portfolio. Six holdings, all in related businesses, all with betas near one. The weighted average comes out at 1.0, which reads as market-like risk. What the figure has measured is how the group moves with an index; what it has not measured is that the six are effectively one position, which is the risk that actually determines the worst month.
The second failure is mixing benchmarks. The average is then meaningless.
A third is omitting cash. It has a beta of zero and a weight.
A fourth is using purchase weights. The current values are what matters.
A fifth is computing it once. Weights drift without any trading.
And a sixth is reading it as total risk. It is one relationship with one index.
Related
Beta covers the measurement and how it is estimated. Correlation is the figure that shows what beta cannot. And Sharpe ratio is the other common portfolio statistic and what it adds.
It is a useful figure and a narrow one. A portfolio of five technology holdings can show a beta close to one and be far riskier than the market, because the concentration risk it carries is not the kind beta measures at all.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.