Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Picture an investor in the 1940s hunched over a set of financial statements, calculator in hand, hunting for the handful of stocks the crowd had ignored. Odds are that investor had absorbed the teachings of Benjamin Graham, the man most people credit as the founder of value investing. Graham distrusted enthusiasm. He built simple, almost stubbornly conservative rules to keep ordinary investors from overpaying, and one of those rules has outlived nearly every market regime since: the Graham Number.
The formula is more than seventy years old. The interesting question is not whether it is “correct” — it plainly was not designed for the world we invest in now — but why a back-of-the-envelope calculation from the Truman era keeps showing up in screeners, valuation tools, and the mental models of disciplined investors. The answer says more about market psychology than about arithmetic.
What the number actually measures
The Graham Number is not a fair-value estimate in the way a discounted-cash-flow model is. It is a ceiling. It answers a narrower, more defensive question: what is the most a cautious investor should be willing to pay for a stock before the price itself becomes the risk? Graham, who lived through the Great Depression and watched investors chase hope while ignoring fundamentals, wanted a line in the sand. Cross it, and you are paying for optimism rather than substance.
The mechanics are trivial. You take earnings per share, multiply by book value per share, multiply that by 22.5, and take the square root. The output is a per-share price.
The only genuinely interesting figure is the 22.5. It is not arbitrary, and it is not magic. It is two of Graham’s separate rules multiplied together: a maximum price-to-earnings ratio of 15 and a maximum price-to-book ratio of 1.5. Fifteen times one-and-a-half is 22.5. The square root simply collapses earnings power and asset backing into a single number you can hold up against the live quote. The idea, refined across editions of Graham’s 1949 classic The Intelligent Investor, was to give a non-professional a fast, conservative anchor — a way to flag a stock as worth a closer look, not a verdict.
That distinction matters and gets lost constantly. The number is a filter, not a thesis.
Why a 1949 formula refuses to die
Here is the contrarian read. The Graham Number persists not because the multiples are right — they almost certainly are not, for most of today’s market — but because it encodes a behavioral discipline that markets punish people for abandoning. Graham’s deeper point was that the margin of safety is always a function of the price you pay, and that the cheapest path to risk reduction is simply refusing to overpay. The formula is a mechanical enforcement of that refusal.
Every bull market produces a fresh cohort of investors convinced that valuation is a relic. The Graham Number is irritating to them precisely because it keeps saying no. In frothy markets it disqualifies almost everything, which feels like a malfunction and is actually the signal working as intended. Warren Graham’s most famous student built an empire on the underlying instinct even as he outgrew the rigid formulas; the principle survived the math.
Where the formula breaks — and why that is the more useful insight
The honest case against the Graham Number is strong, and a serious investor should sit with it rather than wave it away.
The formula leans entirely on book value, and book value has been quietly eroding as a measure of corporate worth for decades. Modern balance sheets do a progressively worse job of capturing what a company is actually worth, because the most valuable assets in an intangible-heavy economy — brands, software, research pipelines, network effects — are rarely recorded as assets at all. When a company expenses its R&D and brand-building rather than capitalizing it, reported book value sits far below economic reality, and any price-to-book screen will scream “overvalued” at businesses that are nothing of the sort.
The distortion is not subtle. Researchers have shown that adding intangibles back to book value can completely reverse the apparent failure of value strategies over the past decade, which tells you the problem was the measuring stick, not the philosophy. Run the Graham Number on an asset-light compounder and it will reject a genuinely good business. That is not the formula being wrong about value; it is the formula being blind to a category of value its inputs cannot see.
There is a second, quieter failure mode. Both earnings per share and book value are accounting outputs, and accounting outputs bend to one-time items, write-downs, and discretionary choices. A single non-recurring gain can inflate the number; an impairment charge can gut it. Feed the formula unadjusted figures and it will confidently price a distortion.
And then there is the oldest trap in value investing: the value trap. A stock trading below its Graham Number may be cheap because the market is wrong, or cheap because the market is right and the business is deteriorating. The number cannot tell the difference. It has nothing to say about competitive position, management, industry structure, or the direction of travel. Treated as a final answer rather than a first question, it walks investors straight into stocks that are cheap for excellent reasons and stay that way.
How to actually use it
The defensible way to deploy the Graham Number is narrow and deliberate. It works best where its assumptions hold: mature, profitable, asset-heavy businesses with stable earnings and real tangible book value — utilities, consumer staples, industrials, financials. In those corners of the market, book value still means something close to what Graham intended.
Use it as the first sieve, not the analysis. Screen for stocks trading below the number, then do the work the formula cannot: assess the business, normalize the earnings, check the balance sheet for the intangible distortions above, and cross-reference against other lenses entirely. Pair it with a real intrinsic-value estimate so you are comparing a conservative ceiling against a forward-looking valuation rather than trusting either alone. The number’s job is to narrow the field and impose discipline on what you are willing to pay. Everything after that is judgment.
The enduring value of a Depression-era square root, in the end, is not the price it spits out. It is the question it forces: am I buying substance, or am I buying hope? Markets reward people who keep asking it and eventually humble the ones who decide it no longer applies.
