Benjamin Graham is the father of value investing. He taught Warren Buffett at Columbia, wrote the two books that built the entire discipline ("Security Analysis" in 1934 with David Dodd, and "The Intelligent Investor" in 1949), and ran the Graham-Newman partnership at roughly 20 percent a year for two decades. Almost everything Buffett, Munger, and modern value investors do traces back to him.
Graham's whole philosophy compresses into one sentence: buy a business for meaningfully less than it is worth, and let the gap between price and value protect you. He called that gap the margin of safety. This post lays out his actual rules, the formulas he used, and how to screen for Graham-style stocks in 2026.
Price is not value
Graham's most famous teaching device is Mr. Market. Imagine a business partner who shows up every day and quotes you a price to buy your shares or sell you his. Some days he is euphoric and names a silly-high price. Other days he is depressed and offers to sell you his stake for far less than it is worth. Mr. Market is there to serve you, not to guide you. Your job is to transact only when his mood hands you a bargain, and to ignore him the rest of the time.
The lesson: the market price of a stock and the underlying value of the business are two different numbers. Investing is the discipline of estimating the second and buying only when the first is well below it. For the mechanics of estimating value, see what is intrinsic value.
Margin of safety: the three most important words in investing
Graham argued that no estimate of value is precise, so you should demand a discount large enough to absorb your own errors. If you think a business is worth 100 dollars a share, you do not buy it at 95. You buy it at 60 or 70, so that even if you are wrong by 20 percent, you still do not lose money.
That buffer is the margin of safety, and it is the single idea that separates investing from speculation. It is also why Graham insisted on quality of the balance sheet, not just cheapness of the price: a discount only protects you if the business survives long enough for value to be recognised.
Graham's checklist for the defensive investor
In "The Intelligent Investor," Graham gave a concrete screen for the everyday "defensive" investor who does not want to do deep analysis. The criteria:
1. Adequate size
Avoid tiny companies. Graham wanted businesses large enough to have survived a full economic cycle. In today's terms, think mid-cap and up.
2. Strong financial condition
Current assets at least twice current liabilities (a current ratio of 2 or more), and long-term debt below net current assets. A fortress balance sheet is what lets a cheap stock wait out a bad year.
3. Earnings stability
Positive earnings in each of the last 10 years. No boom-and-bust.
4. Dividend record
Uninterrupted dividends for at least 20 years. A sign of durable cash generation and shareholder discipline.
5. Earnings growth
At least a third growth in per-share earnings over the last 10 years (using 3-year averages at the start and end to smooth the noise).
6. A moderate price
A P/E ratio no higher than 15 on average earnings, and a price-to-book no higher than 1.5.
7. The combined rule
Graham allowed a trade-off between the two valuation limits: P/E times P/B should not exceed 22.5. That single number is the origin of the famous Graham number.
The Graham number
Combining the P/E limit of 15 and the P/B limit of 1.5 gives Graham's shortcut for the maximum "defensive" price to pay:
Graham Number = square root of (22.5 x EPS x Book Value per Share)
If a company earns 4 dollars per share and has a book value of 30 dollars per share, the Graham number is the square root of (22.5 x 4 x 30) = the square root of 2,700 = about 52 dollars. Above that, Graham would say you are paying too much for a defensive holding.
Net-nets: Graham's deepest-value screen
For his own enterprising money, Graham went further. He bought "net-nets": companies trading below their net current asset value, meaning you were effectively getting the entire business, factories and brand and future profits, for free.
NCAV per share = (Current Assets - Total Liabilities) / Shares Outstanding
Graham's buy zone = price below two-thirds of NCAV per share
Buy a basket of these, he argued, and diversification does the rest: some go to zero, but the group, bought at a deep discount to liquidation value, wins over time.
Does Graham still work in 2026?
Two honest answers.
Net-nets are rare now. In a picked-over, algorithm-scanned large-cap market, a business trading below liquidation value almost always has a real problem. They still appear in small caps, in unloved international markets, and in sectors the crowd has written off, but you have to hunt.
The principles, though, are timeless. Demand a margin of safety. Refuse to overpay for growth. Insist on a balance sheet that can survive a bad year. Diversify enough that any single mistake cannot ruin you. Those rules protected investors in 1934, in 2008, and they will protect you in 2026. What changes is where the bargains hide, not whether the method works.
How to screen for Graham stocks on invest-like
The Graham framework is one of the seven investor lenses every stock is scored against on invest-like. Instead of computing the Graham number and NCAV by hand across thousands of names, you get a Graham-fit score on any ticker in seconds. See how the Graham fit score works, then run a name you are curious about, for example Berkshire Hathaway, and read its full framework breakdown.
The higher-conviction signal is the seven-framework consensus screen: stocks that pass Graham, Buffett, Lynch, Greenblatt and the rest at the same time. That screen has beaten the S&P 500 by a wide margin over five years, and every pick is logged in public on the track record.
The one mistake to avoid
The value trap. A stock is not a bargain just because it is cheap. Graham diversified precisely because deep-value names can be cheap for a reason: a dying business, a broken balance sheet, a management team destroying capital. The fix is exactly what Graham taught: pair cheapness with quality. A cheap price on a durable business is a bargain. A cheap price on a melting ice cube is a bill you have not received yet.
If you want the next step from Graham's rules, read our breakdown of Peter Lynch's tenbagger criteria, which adds growth to Graham's value discipline, or how Michael Burry hunts deep value, which takes Graham's margin of safety to its contrarian extreme.