Benjamin Graham Seven Criteria for Stock Selection

A stock can look cheap because its share price has fallen. That is not the same as being cheap relative to a durable business. The Benjamin Graham seven criteria were designed to separate established, financially sound companies from businesses whose apparent bargain price merely reflects real risk.

Graham wrote these tests for the defensive investor: the investor who wants a repeatable process, broad diversification, and protection from permanent capital loss rather than the excitement of predicting the next market move. They are deliberately demanding. A company does not receive credit for a compelling narrative, a popular product, or a favorable analyst target.

What the Benjamin Graham Seven Criteria Measure

The seven criteria appear in Graham's framework for selecting common stocks for the defensive investor. Together, they examine a company's scale, liquidity, debt burden, profit history, dividend record, earnings growth, and valuation. The logic is straightforward: pay a moderate price only after establishing that the underlying business has survived adversity and carries a conservative financial structure.

The original numerical thresholds were written in a different economic era. A dollar-sales requirement that made sense in the early 1970s does not translate directly to a modern market with far larger companies and different accounting conventions. The principles, however, remain useful. Investors should update the scale of the tests, not discard their purpose.

1. Adequate enterprise size

Graham wanted defensive investors to own businesses with meaningful operating scale. In his final editions, he used minimum sales for industrial companies and minimum total assets for utilities. Size was not a proxy for superior returns. It was a rough safeguard against the fragility, narrow customer bases, and financing dependence often found in smaller enterprises.

For a US large-cap investor, this test is usually less about meeting a fixed nominal threshold and more about asking whether the company has durable commercial relevance. Revenue scale, market position, customer concentration, and access to capital matter. A large company can still be weak, but a small or highly concentrated one deserves a larger margin of safety.

2. A sufficiently strong financial condition

For industrial companies, Graham's standard required current assets of at least twice current liabilities. He also wanted long-term debt no greater than net current assets, meaning current assets minus current liabilities. This is a strict balance-sheet test built around liquidity.

The ratio cannot be applied mechanically across every sector. Banks, insurers, real estate investment trusts, utilities, and asset-light software firms operate with different balance-sheet structures. Still, the underlying question does not change: can the company meet obligations without relying on an accommodating credit market, asset sales, or new equity issuance?

Read the balance sheet alongside the notes. Total debt alone can conceal lease obligations, pension deficits, supplier-financing programs, or large near-term maturities. Free cash flow can make a lower current ratio acceptable for some mature businesses. It does not make refinancing risk disappear.

3. Positive earnings in each of the past 10 years

Graham required no deficit in earnings over the previous decade. This criterion favors businesses that have demonstrated economic resilience across at least one difficult period. A single profitable year says little. Ten years of positive earnings provide evidence that a company can endure changing demand, cost pressure, and recessions.

The quality of those earnings matters as much as the reported total. Investors should distinguish recurring operating profit from gains on asset sales, tax benefits, acquisition-related accounting, or reversals of prior write-downs. A company may technically pass the test while its economic earnings are less stable than the income statement suggests.

For financial companies, credit losses and reserve releases deserve particular scrutiny. For acquisitive companies, compare net income with cash generation and track whether per-share results improved after accounting for stock issuance.

4. Uninterrupted dividends for at least 20 years

A long, unbroken dividend record was Graham's evidence that a business had repeatedly produced distributable cash and treated outside shareholders as genuine owners. The test also imposed discipline on management. A company that can fund dividends through multiple cycles has usually avoided some of the more extreme forms of financial strain.

But dividends are not universally superior to retained earnings. A company with exceptional reinvestment opportunities may create more value by retaining capital, while a high dividend paid with borrowed money is not a sign of strength. The useful modern interpretation is to examine the history of capital allocation: dividends, repurchases, debt reduction, acquisitions, and dilution.

A dividend record should also be measured per share. Total dividends can rise while share issuance spreads the economic benefit across a growing shareholder base.

5. At least one-third earnings-per-share growth over 10 years

Graham required aggregate growth of at least one-third in per-share earnings over the preceding decade, using three-year averages at the beginning and end of the period. The use of averages was intentional. It reduces the influence of a temporary peak or trough.

Per-share earnings are central here. A company can increase total profit while issuing shares so rapidly that each shareholder owns a smaller claim on the result. Buybacks can improve per-share figures too, but investors should determine whether those repurchases were funded from genuine excess cash or from incremental debt.

This is not a demand for rapid growth. Graham sought modest, proven progress. That restraint is valuable when markets are pricing companies as if recent growth will continue indefinitely. A slow but dependable compounder purchased at a sensible price can be more defensible than a fast grower purchased on assumptions.

6. A moderate price relative to average earnings

Graham set a maximum price-to-earnings ratio of 15 based on average earnings from the prior three years. Averaging earnings reduces the chance of valuing a cyclical company at the top of its profit cycle, when its trailing P/E can look deceptively low.

The number 15 is not a permanent law. Interest rates, inflation, business quality, and the durability of earnings all affect what a reasonable multiple may be. Yet the discipline remains clear: valuation should begin with normalized earnings, not a management-adjusted forecast several years into the future.

For cyclicals, three years may not cover a full cycle. Commodity producers, homebuilders, and industrial businesses may require a longer history and a mid-cycle estimate. For companies with substantial stock-based compensation, use a per-share measure that recognizes the cost of dilution rather than accepting an adjusted earnings figure at face value.

7. A moderate price relative to assets

Graham also required a price-to-book ratio no higher than 1.5. He allowed flexibility between the earnings and asset tests through a combined rule: the P/E ratio multiplied by the price-to-book ratio should not exceed 22.5. A stock at 15 times earnings needed to trade at no more than 1.5 times book value; a lower P/E could justify a somewhat higher price-to-book ratio.

Book value remains more informative for banks, insurers, and asset-intensive businesses than for companies whose value rests on software, brands, research, or customer networks. Intangibles, goodwill, and aggressive acquisition accounting can make reported book value a poor measure of liquidation value. In those cases, tangible book value, returns on invested capital, and sustainable free cash flow may provide better evidence.

The principle is not that every quality business must trade near book value. It is that investors should not pay a premium to earnings and assets at the same time without unusually strong, well-supported reasons.

Applying Graham's Tests Without Pretending It Is 1972

The mistake is not adapting Graham's thresholds. The mistake is adapting them until every favored stock qualifies. A disciplined screen should preserve the direction of the original safeguards: meaningful scale, conservative financing, long records of profitability, shareholder-friendly capital allocation, per-share progress, and a price that leaves room for error.

Start with reported figures from annual filings rather than summary websites. Reconcile multi-year earnings with restatements, discontinued operations, and changes in share count. Check whether debt has moved into footnotes or off-balance-sheet arrangements. Then compare management's public claims about resilience, growth, and capital returns with the numbers formally disclosed to shareholders.

This is where a filing-first process earns its keep. Hety evaluates long-term financial records, valuation ranges, Graham-style quality tests, and management statements against corporate disclosures so investors can see both the result and the evidence behind it. A screen is a starting point, not a substitute for judgment.

Where the Criteria Need Judgment

A company that fails one criterion is not automatically uninvestable. A well-capitalized software business may carry little working capital yet produce recurring cash flow. A regulated utility may use more leverage than Graham's industrial standard permits while still having stable, contract-like revenue. A high-return consumer business may trade above book value for decades.

Conversely, a company that passes every mechanical test can still be a poor investment if its industry is deteriorating, its accounting is aggressive, or management is allocating capital badly. Screens identify candidates. They do not establish intrinsic value.

The most useful question is not, “Does this stock pass?” It is, “What would have to be true for this business to fail the spirit of the test?” That question directs attention to evidence: debt maturities, customer churn, competitive pressure, pension obligations, dilution, acquisition accounting, and the gap between executive language and filed facts.

Graham's criteria remain valuable because they force patience before commitment. Let a stock earn its place in the portfolio through a record that can be verified, then insist on a price that recognizes uncertainty. Past performance is not predictive, and this material is not investment advice, but a process built on primary evidence is a better starting point than a headline.