The criteria Warren Buffett printed, placed in this market
Buffett set out how he judged a business mostly in words. Five shareholder letters and Berkshire’s acquisition criteria give the six criteria gathered here, each printed beside the letter it was read in. four of them carry a level this product can count the US market against tonight, and every one of those levels is ours rather than his.
Nothing on this page is about the man. No holdings, no filings of his, no portfolio: those are a different product and this one does not build them. What is here is a set of questions to put to a company’s accounts, and an honest account of how far our arithmetic sits from his words.
Owner earnings yieldours: our number or our measure≥ 8%A tenth to a quarter of the 4,472 ranked companies with an owner earnings yield meet it.
Buffett defined owner earnings as reported earnings plus depreciation, depletion, amortisation and certain other non-cash charges, less the average annual capital spending a business needs to keep its competitive position and unit volume. He called that last term a guess, and still preferred the figure to reported earnings for valuation.
8% sits between the 75th percentile, 4.86%, and the 90th, 10.49%.
Who is counted
Companies with filed statements, one each; exact zeros left out as missing data; banks, insurers, mortgage REITs, industry "commodity contracts brokers & dealers", industry "financial - capital markets" and industry "investment - banking & investment services" left out (555); winsorised.
What we changed
The appendix defines a measure and sets no yield; 8 percent is the level the product's owner earnings screen runs. Filings rarely separate maintenance capital spending from spending to grow, so the product subtracts all of it, and a company investing heavily to grow reads lower here than on his measure.
Return on equityours: our number or our measure≥ 15%A tenth to a quarter of the 4,581 ranked companies with a return on equity meet it.
Buffett named a high rate of earnings on equity capital, achieved without undue debt or accounting tricks, as the primary test of a management's economic performance, rather than consistent gains in earnings per share. Berkshire's acquisition criteria sought good returns on equity with little or no debt.
15% sits between the 75th percentile, 12.80%, and the 90th, 25.43%.
Who is counted
Companies with filed statements, one each; exact zeros left out as missing data; extremes capped at the 1st and 99th percentiles.
Return on equity benchmarks ›What we changed
Buffett gave this test no number. Fifteen appears in the letters, but not as his test: in 1986 as the return on equity Berkshire hoped to average, and in 1987 as a Fortune study's yearly floor, beside its ten-year average above 20 percent. His 1979 measure was operating earnings over beginning equity, with securities at cost. The product reads one filed year of net income over shareholders' equity, or a vendor's trailing figure.
Debt to equityours: our number or our measure≤ 0.5A quarter to a half of the 2,321 ranked companies with a debt to equity ratio meet it.
Berkshire's acquisition criteria sought businesses employing little or no debt.
0.5 sits between the 25th percentile, 0.16, and the 50th, 0.57.
Who is counted
Companies with filed statements, one each; exact zeros kept (a debt-free balance sheet really is 0); banks left out (323); extremes capped at the 1st and 99th percentiles.
Debt to equity benchmarks ›What we changed
Berkshire wrote these criteria for acquiring whole businesses, not for choosing stocks. Little or no debt is a phrase, not a ratio. The product reads total debt over shareholders' equity, so a level of 0.5 allows debt up to half the equity. The 1987 letter described the debt of a Fortune study's leaders against their capacity to pay interest, which is a different measure again.
Return on tangible assetsours: our number or our measure≥ 20%A twentieth to a tenth of the 5,024 ranked companies with a return on tangible assets meet it.
Grading Berkshire's own manufacturing, service and retailing businesses by after-tax earnings on net tangible assets, measured without the effect of debt, Buffett called 25 percent to far more than 100 percent terrific and 12 to 20 percent good.
20% sits between the 90th percentile, 14.47%, and the 95th, 20.58%.
Who is counted
Companies with filed statements, one each; exact zeros left out as missing data; extremes capped at the 1st and 99th percentiles.
What we changed
His grades describe Berkshire's own operating businesses, on net tangible assets with no benefit from debt. The product divides net income by total assets less goodwill and intangibles, with no liabilities subtracted, a larger base that likely reads lower for the same company; so 20 percent is its own level, and his 25 is not drawn on a base he did not use.
Not measured, and why
Demonstrated consistent earning power. A record, not a level; no cohort of streaks is published.
Berkshire's acquisition criteria sought demonstrated consistent earning power, and said future projections and turnarounds were of no interest. Berkshire wrote these criteria for acquiring whole businesses, not for choosing stocks. A record rather than a level, and the criteria give no span of years and no allowance for a bad one.
An economic franchise. Three conditions about customers and regulation. No field holds any of them.
Buffett described an economic franchise as a product or service that is needed or desired, that customers think has no close substitute, and that is not subject to price regulation, and said it shows in the ability to price aggressively and earn high returns on capital.
GET /api/public/distribution/{key}. Cohorts built Oct 16, 2026.
Prices of 16 Oct 2026.
Why every number here says Ours
The thirty-four Berkshire texts we hold were searched for a level stated as a test for other companies, and there is none. Fifteen percent does appear twice: in the letter for 1986 as the return on equity Berkshire hoped to average itself, and in the letter for 1987 as the yearly floor in a Fortune study he cited. Neither is a bar he set for anybody else, so neither is printed as his. The levels above are the ones our own screens run after these criteria, and each row opens onto what we changed to run them.
A criterion we cannot measure is not scored. It reads Not measured and the discipline carrying it is Incomplete, which is neither met nor failed. That is the same rule the whole catalogue runs on and it is why two of these six sit under the fold rather than in the ledger.
What this page cannot tell you
Meeting all of these tonight is not a reason to own anything. A share of the market clearing a line is a fact about the market, not about a company, and the part of his method that mattered most, whether a business will still be able to price its product in ten years, appears in no filing and in no field here. Two of the six criteria on this page cannot be measured at all, and they are named rather than quietly dropped.
Put your own companies through these criteria in the audit, free. What is sold is what comes after: you set the line yourself, and a return on equity that slips below the one YOU chose is the guardrail, not the one Berkshire wrote.
Every guardrail you have saved is re-tested every night against newly filed accounts, and you get one email the first night one breaks, naming the company, the reading and the filing it came from. What to do about it is yours. It comes with Professional.
The check runs every night at 23:30 UTC. A filing is picked up the night after it is published.
The audit needs no account. A new account starts with 30 days of Professional, no card.
Meeting a test is not a reason to buy. Every count above is read from the public endpoints when the page is built, never typed into it: the whole table is at the company catalogue and every formula is at methodology.
ValueScreener is not affiliated with or endorsed by Warren Buffett. The criteria above are quoted from published Berkshire Hathaway documents and every level run against them is ours. The same tests, with the rest of the roster, are at Gurus.