# Altman Z-Score The Altman Z-score is used to estimate the risk of bankruptcy within about two years. **The question.** How close is this company to financial distress, judged from its published accounts? Also known as: Often just the Z-score. The later variants are written Z-prime and Z-double-prime. ## When to use it - A lender, supplier, or acquirer wants a quick screen of many counterparties before detailed credit work, to decide who gets looked at first. - You join a turnaround and need to check whether the ratios were already weakening in the last few years of accounts. - A client's supplier base includes thinly capitalized firms, and procurement wants to know which ones to review first. ## When to avoid it - The company is a bank, insurer, or other financial firm. Balance-sheet models of this kind are not meant for them, because their balance sheets are opaque and they carry off-balance-sheet items. - The company is not a public manufacturer and you only have the original coefficients. Use the variant built for its type, or another approach. - A liquidity decision is days away. Published accounts lag, and a 13-week cash forecast tells you more than a ratio does. - You need a prediction for a young or asset-light firm. The model was built on manufacturers, and the retained earnings ratio penalizes any company that has not had time to accumulate profit. ## Overview The Altman Z-score combines five ratios from a company's accounts into one number and places it in a safe, grey, or distress zone. It supports a screening choice: which counterparties, customers, or targets need a closer credit look first, and which can wait. The original weights were fitted to separate bankrupt from surviving manufacturers up to about two years before failure. Which version you use matters as much as the arithmetic. The 1968 coefficients and cutoffs suit public manufacturers; private firms and non-manufacturers need the variants. Read the trend and which ratio moved, not the decimal, and treat a low score as a prompt to ask for a cash forecast. ## What it is not - A probability of failure. The score sits in one of three zones, and the zones are cutoffs from a 1960s sample, not calibrated default rates. - A verdict. A company in the grey zone is neither healthy nor failing, and some companies that score low survive. - A single formula. The original, Z-prime, and Z-double-prime have different coefficients and different cutoffs, and mixing them gives a number that means nothing. ## The steps 1. **Choose the version.** Use the original model for public manufacturers, Z-prime for private firms, and Z-double-prime for non-manufacturers and emerging markets. Write down the coefficients and cutoffs that belong to the version you chose, and keep them with the result. 2. **Collect consistent inputs.** Pull three to five years of statements. You need total assets, working capital, retained earnings, EBIT, sales, total liabilities, and for the original model the market value of equity at the balance sheet date. Use the same definitions for every company and year. 3. **Compute the ratios and the score.** Calculate each ratio as a decimal, multiply by its coefficient, and sum. Keep the contribution of each term in a separate column, because the contributions show which ratio is driving a change. 4. **Read the zone and the direction.** Note the zone and how the score has moved over the years. A grey score that is falling fast is a different case from a grey score that is rising. Look at which terms moved before you comment on the total. 5. **Test against forward-looking evidence.** For every flagged company, check what the accounts cannot show. That means the cash forecast, debt maturities, covenant headroom, order book, and lender behavior. The output is a ranked watch list, with the follow-up checks named. ## The formula Z = 1.2 × X1 + 1.4 × X2 + 3.3 × X3 + 0.6 × X4 + 1.0 × X5 A weighted sum of five ratios. In the original model for public manufacturers, a score above about 2.99 is the safe zone, a score between 1.81 and 2.99 is the grey zone, and a score below 1.81 is the distress zone. The coefficients and cutoffs belong to that version only. - X1: working capital ÷ total assets. Short-term liquidity relative to the size of the company. Enter X1 to X4 as decimals, such as 0.20 for 20 percent, when using these coefficients. - X2: retained earnings ÷ total assets. Accumulated profit, which also reflects the age of the company. - X3: EBIT ÷ total assets. The earning power of the assets before interest and tax. - X4: market value of equity ÷ total liabilities. How far the equity cushion can shrink before liabilities exceed assets. - X5: sales ÷ total assets. Asset turnover, in times. ## In practice Procurement at a client flags a stamping supplier whose lender has gone quiet, and the supplier is publicly traded. Last year: total assets 500, working capital 70, retained earnings 90, EBIT 40, market value of equity 170, total liabilities 340, sales 650. That gives X1 = 0.14, X2 = 0.18, X3 = 0.08, X4 = 0.50, X5 = 1.30, and Z = 0.168 + 0.252 + 0.264 + 0.300 + 1.300 = 2.28. This year: total assets 500, working capital 50, retained earnings 80, EBIT 25, market value of equity 100, total liabilities 350, sales 600. X1 = 0.10, X2 = 0.16, X3 = 0.05, X4 = 0.286, X5 = 1.20, and Z = 0.120 + 0.224 + 0.165 + 0.171 + 1.200 = 1.88. Both years are in the grey zone, so the screen does not say the supplier will fail. It says the direction is bad, and 1.88 is only 0.07 above the distress cutoff. The market value of equity accounts for about 0.13 of the 0.40 decline, and falling sales and EBIT together for about 0.20. The market's view is doing a lot of the work, so the team asks for a cash forecast and the covenant position before it changes any supply terms. ## What you leave with - A score for each company, with the version and cutoffs used stated beside it. - The contribution of each ratio, and the trend over several years. - A ranked watch list for follow-up. - A list of forward-looking checks to run on each flagged company, such as cash forecast, covenant headroom, and debt maturities. ## Failure points ### The wrong version The original coefficients are applied to a private company or a services firm, and the original cutoffs are applied to a variant score. The number looks precise and means little. ### Reading the grey zone as an answer The grey zone is the range where the model does not separate the groups well. Decisions get made on the second decimal, when the honest reading is that the screen is inconclusive. ### Lagging inputs Statements are months old by the time they are filed. A company already in a liquidity squeeze can still score in the grey zone, because the ratios describe the last reporting date. ### Market price in the denominator X4 uses the market value of equity. A sector sell-off lowers the score of a sound company, and a private company has no price to use at all. Use the book-value variant for those, and do not compare its score with public ones. ### Structural differences read as distress A young company with little retained earnings, or a company in an industry with a different asset base, scores low for structural reasons. Compare against peers of the same type and age before you flag it. The failures above are usually [Commercial Acumen](https://usefulconsultingframeworks.com/your-edge/#commercial): the work was not tied to money, for the client and/or for you. ## Origin Edward Altman, then an assistant professor of finance at New York University, published the model in the Journal of Finance in September 1968. He applied multiple discriminant analysis to an initial sample of 66 manufacturing firms, half of them bankrupt and half not, and chose the combination of five ratios that separated the two groups best. The original model is for publicly traded manufacturers. Altman later published variants. Z-prime replaces the market value of equity with book value, for private firms. Z-double-prime drops the sales to assets ratio and is estimated for non-manufacturers and emerging markets. Each variant has its own coefficients and its own zone cutoffs. ## Further reading - [Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy](https://onlinelibrary.wiley.com/doi/10.1111/j.1540-6261.1968.tb00843.x). Edward Altman, Journal of Finance, September 1968. The original paper, with the sample, the ratio selection, and the zones. Paywalled at the publisher, so check library access. - [Altman Z-score](https://en.wikipedia.org/wiki/Altman_Z-score). Wikipedia's summary of the original formula, the variants, and the main criticisms, including later academic models that use market-based data. Free. ## Related frameworks - [DuPont Analysis](https://usefulconsultingframeworks.com/frameworks/dupont-analysis/): Also works from the published ratios, but explains the level of return instead of screening for survival. Pick it when the company is healthy and the question is how well it performs. - [Kraljic Matrix](https://usefulconsultingframeworks.com/frameworks/kraljic/): Sets a purchasing posture from supply risk. A Z-score is one input to the supply-risk side when the supplier is a bottleneck or strategic one. Last updated October 19, 2026.