TL;DR: The Graham Number is the most Benjamin Graham's rules let a defensive investor pay for a stock: the square root of 22.5 times earnings per share times book value per share. Below it, a stock passes his price discipline and earns deeper research. Above it, his answer is to walk away. It is a screen, not a valuation, and knowing the difference is the whole point.
What is the Graham Number?
Benjamin Graham, Warren Buffett's teacher and the author of The Intelligent Investor, spent his career on one problem: how do you stop yourself from overpaying? His answer was a set of price rules for the defensive investor, the person who wants sound results without making investing a second job.
The Graham Number compresses two of those rules into a single figure:
Graham Number = √(22.5 × EPS × BVPS)
EPS is earnings per share. BVPS is book value per share, the accounting value of what the company owns after subtracting what it owes, divided by the share count. Run the formula and you get a price. If the stock trades below that price, it passes Graham's discipline. If it trades above, it fails, no matter how good the story sounds.
Where does 22.5 come from?
It is not a magic constant. Graham set two ceilings: pay no more than 15 times earnings, and no more than 1.5 times book value. A stock at both ceilings at once is the most expensive stock his rules tolerate. Multiply the ceilings together, 15 × 1.5, and you get 22.5.
That construction explains how the number behaves. It punishes a high price-to-earnings ratio and a high price-to-book ratio at the same time, and it rewards the opposite combination. A profitable company with a solid balance sheet trading at modest multiples gets a Graham Number above its price. A market favorite priced for perfection gets one far below it.
How do you calculate it? A worked example
Take a company with earnings per share of $4.00 and book value per share of $25.00:
- 22.5 × $4.00 × $25.00 = $2,250
- √$2,250 ≈ $47.43
The ceiling price is $47.43. At a market price of $35, the stock passes the screen. At $60, it fails.
You can run your own numbers in our Graham Number calculator, which also shows how far the current price sits from the ceiling. Our step by step walkthrough, how to calculate the Graham Number, runs the same four steps by hand so you can check the calculator's work. Use a full-year or trailing-twelve-month EPS, not a single quarter. One unusual quarter, a big write-down or a one-time gain, will drag the result far from anything Graham intended.
Is the Graham Number the same as intrinsic value?
No, and confusing the two causes real damage. Intrinsic value is an estimate of what a business is worth based on the cash it will produce in the future. That is what a DCF model builds, assumption by assumption. The Graham Number looks backward, not forward. It uses today's reported earnings and today's balance sheet, and it answers a narrower question: at this price, am I overpaying by the standards of a man who survived 1929?
Used properly, the two tools stack. The Graham Number is the fast first pass that costs you ten seconds. The DCF is the slow second pass you run on whatever survives. Investors who skip the first pass spend their research hours on stocks a simple price rule would have eliminated. Investors who skip the second are screening, not investing.
When does the Graham Number mislead?
Graham wrote for an era of railroads, manufacturers, and utilities. His formula carries that world inside it, and it shows in four places:
- Asset-light businesses. A software company whose real assets are engineers and code carries a small book value, so its Graham Number comes out far below any price the market will ever offer. The formula does not hate the business. It simply cannot see it.
- Cyclical earnings. Use a steelmaker's peak-year EPS and the formula blesses the stock at the top of the cycle. Graham himself advised averaging earnings across several years to blunt exactly this.
- Negative numbers. Negative EPS or negative book value breaks the math, deliberately. The screen was built for established, profitable companies. A stock it cannot score is telling you to use a different framework, or none.
- Cheap for a reason. Some stocks trade below their Graham Number because the business is deteriorating and the earnings are about to follow the price down. Value traps pass screens. That is what the rest of your due diligence is for.
How does it fit into a real workflow?
Here is the honest division of labor. The Graham Number does not find you great companies. It enforces a price habit across every company you look at, including the ones you are excited about, which is when discipline matters most.
A workable routine: screen for candidates with the fundamentals you care about, run the Graham Number on each name before it goes on your watchlist, and let failures carry a note explaining why the price is justified anyway. Sometimes it will be, growth companies break Graham's rules for a living. But you will have made the exception consciously, with the burden of proof on the story, instead of drifting into it.
Inside FullPort, this calculation sits next to the full valuation toolkit, DCF, reverse DCF, Magic Formula, and the rest, applied to the stocks you actually follow. Start free and run it on your own watchlist: create your account. And when a name passes the screen, the discipline that matters next costs nothing either: paper trade the strategy first and let a month of live prices grade the thesis before real capital does.
Educational content, not financial advice. The Graham Number is a historical screening rule, not a recommendation to buy or sell any security.
