The Graham Number: A One-Formula Test for Cheap Stocks

Benjamin Graham — the man who taught Warren Buffett — compressed the entire question of "is this stock cheap?" into a single square-root formula. Run a beaten-down value name like Ford (F) through the Graham Number and you get a hard ceiling on what a careful investor should ever pay.

What Is the Graham Number?

It is a quick ceiling on what a defensive investor should pay for a stock. Benjamin Graham, the father of value investing, wanted a simple guardrail against overpaying, so he fused two of his rules into one number.

His two rules were blunt: never pay more than about 15 times earnings, and never pay more than about 1.5 times book value. Multiply those limits together — 15 × 1.5 — and you get 22.5, the magic constant buried inside the formula.

The Graham Number is not a price target; it is a "do not exceed" line drawn in the sand. If a stock trades below its Graham Number, it may be cheap on Graham's old-school terms. If it trades far above, a conservative investor is being asked to pay for optimism, not assets.

How Do You Calculate the Graham Number?

One formula, two inputs. The Graham Number equals the square root of (22.5 × earnings per share × book value per share).

Say a company earns about $6 per share and carries book value of roughly $100 per share. Multiply: 22.5 × 6 × 100 = 13,500. Take the square root, and the Graham Number lands near $116. That is the most Graham would tell you to pay.

The logic is elegant. By taking a square root of the earnings-times-book product, the formula quietly enforces both limits at once — it refuses to let a rich price on one measure hide behind a cheap price on the other. A stock can only clear the bar if it is reasonable on earnings and on assets together.

You need only two figures from the financial statements: trailing earnings per share and book value per share (total equity divided by shares outstanding). Both sit near the top of any stock's summary page.

What Does the Graham Number Say About Real Stocks?

It flags the classic value corners of the market — banks, automakers, and insurers — while dismissing almost every popular growth name. Here is an illustrative table built to teach the math, not to quote live prices.

Notice who shows up. Asset-heavy, lower-multiple businesses like General Motors (GM), Citigroup (C), Wells Fargo (WFC), and MetLife (MET) produce meaningful Graham Numbers because they carry real book value. A sprawling business like CVS Health (CVS), with a large equity base, screens the same way. To see whether any of them is actually below its ceiling, compare the number above to the live price on the stock's page.

Run a high-flying software or chip stock through the same formula and the Graham Number collapses to a fraction of the market price — which is exactly Graham telling you those companies are priced on their future, not their balance sheet. That is a feature, not a bug.

What Are the Biggest Graham Number Mistakes?

Three errors turn a useful screen into a trap.

First, using it on asset-light or unprofitable companies. A firm with tiny book value or negative earnings produces a meaningless or imaginary Graham Number. The formula needs positive EPS and positive book value to work at all.

Second, treating it as a buy signal. A stock trading below its Graham Number is a candidate for research, not an order to purchase. Cheapness by this measure often reflects a struggling business — the same value-trap risk that snares low-P/E hunters.

Third, ignoring debt and cash quality. Book value can be inflated by goodwill or stale asset marks. A Graham Number is only as trustworthy as the book value feeding it, so always glance at what those assets actually are before you lean on the output.

Pro Tips for Using the Graham Number

Use it as a first-pass filter, then dig. Graham himself ran it across dozens of names to build a shortlist, never to make a final call on one stock.

Pair it with a quality check. A company below its Graham Number that also passes a balance-sheet test — low debt, steady earnings — is far more interesting than a cheap-looking wreck. The forensic mindset behind our super investors profiles is the natural companion here.

And respect the era it came from. Graham built this for a 1930s–1970s market dominated by industrials and railroads, where book value meant factories and inventory. In a world run by software and brands, the Graham Number works best as a sanity check on old-economy stocks, not a universal ruler. For the broader toolkit, our guide to investment strategies puts it in context.

When Should You Ignore the Graham Number?

Whenever the value lives off the balance sheet. Three cases break it completely.

High-growth technology. Companies whose worth is in code, network effects, or brand equity carry little book value, so the formula badly understates them. Judging them by Graham Number guarantees you miss every great compounder.

Asset-light services. Consulting firms, exchanges, and rating agencies generate huge cash on thin equity — the Graham Number treats their light balance sheets as a weakness rather than the strength it usually is.

Cyclical peaks. An automaker like Ford (F) can post a flattering Graham Number at the top of an earnings cycle, right before profits roll over. Critics argue this is the metric's deepest flaw: it anchors on a single year of earnings that may not repeat. They are right, which is why the number is a starting point, never the finish line.

Ready to analyze these stocks yourself? Search any ticker on MainRatios to see valuations from 6 legendary investors - free.