Skip to content
Menu
Finabase Learn

Fundamentals · Ratios

Altman Z-score: formula, zones and what it cannot tell you

Build the 1968 Altman Z-score from five balance-sheet ratios, work it through on Intel, and see why a distress-zone score is a flag and not a forecast.

On this page
  1. The idea
  2. The formula
  3. Seven companies
  4. Where it misleads
  5. Compared with its neighbours
  6. Where to find it in Finabase
  7. Frequently asked questions
  8. Summary
  9. Check your understanding

The Altman Z-score combines five financial ratios into one number that places a company in a safe, grey or distress zone. Edward Altman published it in 1968 as a way to separate manufacturers that later went bankrupt from those that did not, using only figures from the annual report and the stock market.

It is still printed on stock data pages more than fifty years later because it is quick: one number that summarizes liquidity, accumulated profit, current profit, the market's view and how hard the assets work. That convenience is also the danger. The score was fitted to a small group of mid-century factories, and it is now applied to retailers, software companies and banks it was never built for.

This page builds the original score term by term for Intel, sets it beside six other large companies, and spends as long on the limits as on the formula.

The idea

A lender wants to know whether a borrower will be able to pay. No single ratio answers that. A company can have plenty of cash and no profit, or good profit and a wall of debt. Altman's idea was to let the data choose which ratios matter and how much weight each deserves.

He took 66 listed manufacturers: 33 that had filed for bankruptcy between 1946 and 1965 and 33 that had not. He then used a statistical method called discriminant analysis to find the weighted sum of ratios that best separated the two groups using their last statements before the failure. Five ratios survived, and the weighted sum is the Z-score.

Each ratio answers a different question:

  • X1, working capital ÷ total assets. Does the company have short-term assets left after covering short-term bills? See Working capital and the cash conversion cycle, with Apple’s numbers.
  • X2, retained earnings ÷ total assets. How much of the asset base was paid for by profit the company kept over its whole life? Young or loss-making companies score low.
  • X3, EBIT ÷ total assets. How much operating profit do the assets earn before interest and tax? It has the largest weight per unit.
  • X4, market value of equity ÷ total liabilities. How far could the share price fall before the company's market value is less than what it owes?
  • X5, sales ÷ total assets. How much revenue does each dollar of assets produce? This is asset turnover.

The formula

Enter each X as a plain ratio (0.152, not 15.2). Working capital is current assets minus current liabilities. EBIT is earnings before interest and taxes; this page uses operating income as reported. The market value of equity is the share price times the shares outstanding. Four of the five inputs come from the fiscal year end, while the market value can be taken on any day, so say which day you used.

Intel FY2025: the five weighted terms of a Z-score of 2.67 1.2 × working capital ÷ assets 0.18 1.4 × retained earnings ÷ assets 0.32 3.3 × EBIT ÷ assets −0.03 0.6 × market value ÷ liabilities 1.94 1.0 × sales ÷ assets 0.25 Z-score 2.67 (grey zone) Balance sheet and income statement from Intel’s Form 10-K for 2025-12-27; market value at the 2026-11-17 close.
Intel’s Z-score of 2.67, term by term. Balance sheet and income statement from the Form 10-K for the year ended December 27, 2025; market value at the November 17, 2026 close. The market-value term is almost three-quarters of the total.

The chart shows where the score comes from. The market-value term is 1.94 of 2.67. The four terms built from the financial statements add up to about 0.72. Intel had an operating loss for the year, so the term with the heaviest weight contributed nothing, and its sales were only a quarter of its assets because chip factories are expensive.

Now hold the statements fixed and halve the share price. X4's term falls from 1.94 to 0.97 and the score drops to about 1.7, into the distress zone, with not one dollar of debt added. That is arithmetic on this example, not a forecast, and it shows how much of a Z-score can be the stock market's opinion fed back as an input.

Seven companies

The same formula on six more companies, each using its latest Form 10-K and the closing price of November 17, 2026. Fiscal years end in December 2025 except Walmart (January 31, 2026) and Procter & Gamble (June 30, 2026).

Company X2 X3 X4 Z-score
Walmart 0.368 0.105 5.687 6.68 (safe)
Procter & Gamble 1.074 0.156 5.528 5.90 (safe)
Caterpillar 0.664 0.113 3.437 4.24 (safe)
Intel 0.232 −0.010 3.239 2.67 (grey)
Boeing 0.103 0.025 0.791 1.38 (distress)
Ford 0.078 −0.032 0.178 0.79 (distress)
Kraft Heinz −0.057 −0.057 0.743 0.50 (distress)

X2 is retained earnings ÷ assets, X3 is EBIT ÷ assets and X4 is market value ÷ liabilities.

Altman Z-score, seven companies (below 1.81 distress, above 2.99 safe) Procter & Gamble FY2026 5.90 (safe zone) Caterpillar FY2025 4.24 (safe zone) Walmart FY2026 6.68 (safe zone) Intel FY2025 2.67 (grey zone) Kraft Heinz FY2025 0.50 (distress zone) Boeing FY2025 1.38 (distress zone) Ford FY2025 0.79 (distress zone) Original 1968 formula; latest Form 10-K of each company and market value at the 2026-11-17 close.
Altman Z-score on the original 1968 formula, from each company’s latest Form 10-K and its market value at the November 17, 2026 close. Sources: SEC EDGAR; prices from Finabase.

What moves the scores is visible in the terms:

  • Walmart has the highest score, and 2.51 of its 6.68 comes from X5. A retailer sells $2.51 for every dollar of assets; no factory does that. Its working capital is negative ($22,595 million short), which the model reads as a weakness, but a retailer that collects cash at the till before paying suppliers runs that way by design. The model is being used outside its sample here, and the result flatters.
  • Procter & Gamble also has negative working capital and scores 5.90. Its retained earnings are larger than its total assets (X2 of 1.074), which can happen when a company has earned a great deal and spent much of it buying back shares, a deduction recorded elsewhere in equity.
  • Caterpillar is the closest to the companies in Altman's sample: a maker of heavy machines. All five of its terms are positive and its score is 4.24.
  • Boeing made an operating profit of $4,281 million on $168,235 million of assets, so X3 is small. Its liabilities of $162,778 million are larger than its market value of $128.7 billion, so X4 is below 1.
  • Ford had an operating loss of $9,169 million and total liabilities of $253,180 million against a market value of $45.1 billion. A carmaker that also lends to its customers carries the debt of that lending business in its liabilities, which pushes X4 down in a way the 1968 sample of plain manufacturers never had to deal with.
  • Kraft Heinz has the lowest score, 0.50. Its operating loss of $4,669 million for fiscal 2025 is operating income as reported, after a goodwill impairment charge, and its retained earnings are an accumulated deficit of $4,629 million. Both X2 and X3 are negative.

Three of seven large, well-known companies sit in the distress zone on this formula. That says as much about the formula's fit to modern balance sheets as it does about the companies.

Where it misleads

  • A small, old sample. The weights were fitted to 66 manufacturers whose failures happened between 1946 and 1965. Accounting rules, capital markets and the typical balance sheet have all changed since. The weights have not.
  • Not for banks, insurers or asset-light firms. A bank's liabilities are its raw material (deposits), so X4 is always tiny and X1 has no meaning. A software company has few assets, so every ratio with assets underneath is inflated. Altman later published variants for private companies and for non-manufacturers; the original should not be stretched to cover them.
  • X4 moves with the share price. A falling stock lowers the score and a rising one raises it. The score can improve in a bull market while the business gets worse, as the Intel example shows in reverse.
  • Buybacks shrink the equity terms. Share repurchases reduce shareholders' equity, and in many companies' accounts they reduce retained earnings directly. A company with a strong brand that has returned most of its profit can show low or negative retained earnings and score poorly on X2 while having no trouble paying its debts. See Stock buybacks: how share repurchases work and what they do.
  • One-off write-downs sink X3. An impairment of goodwill is a non-cash charge that says a past acquisition is worth less than its book value. It reduces operating income for one year, and the term with the 3.3 weight takes the hit. It reduces total assets and retained earnings as well.
  • Negative working capital is not always weakness. Retailers and consumer-goods makers with fast-selling products and patient suppliers run it on purpose.
  • Mixed dates. The statements are from the fiscal year end and the market value from a later day. After a large move in profit or in the share price, the parts no longer describe the same moment.
  • A flag, not a prediction. The zones describe how groups of companies behaved in the original study. They say nothing certain about one company. Large companies in the distress zone often have access to credit lines, asset sales and new equity that the ratios do not see. Read the debt maturity schedule and the cash flow statement in the 10-K before drawing any conclusion.

Compared with its neighbours

Measure What it uses What it tells you
Altman Z-score Five ratios, fixed weights A zone: safe, grey or distress
Piotroski F-score: the nine tests, worked through on Nike Nine yes-or-no tests on two years Whether the fundamentals are improving
Interest coverage ratio: can earnings pay the interest bill? Operating profit ÷ interest expense Whether profit covers the interest bill
Net debt to EBITDA: years of earnings to repay debt, and its limits Debt after cash ÷ EBITDA Years of earnings needed to repay debt
Current ratio and quick ratio: measuring short-term liquidity Current assets ÷ current liabilities Whether short-term bills are covered

The Z-score is the only one of these that mixes in the share price. The others can be computed for a private company from its statements alone.

Where to find it in Finabase

Open the Valuation tab for Intel and scroll to the Financial health block. It shows Altman Z-Score with its zone and the line "below 1.81 distress, above 2.99 safe", next to the Piotroski F-Score.

That figure is the data provider's, built on the provider's own inputs and dates. It will not match a fiscal-year calculation like this page's exactly. On staging on November 18, 2026 the tab showed 6.46 for Walmart against the 6.68 computed here. The zone is usually the same; the decimals are not.

To rebuild the score yourself, the balance sheet has Current Assets, Current Liabilities, Accumulated Retained Earnings (Deficit), Total Assets and Total Liabilities. The income statement has Total Revenue, Operating Income and Earning Before Interest & Taxes (EBIT). For the market value, multiply the share price by the share count; the income statement lists Weighted Average Shares.

Frequently asked questions

What is a good Altman Z-score?

On the original model, a score above 2.99 is in the safe zone, 1.81 to 2.99 is the grey zone and below 1.81 is the distress zone. Those cut-offs apply to listed manufacturers. For other kinds of company the same number means less.

Does a Z-score below 1.81 mean a company will go bankrupt?

No. It means the company's ratios resemble those of the failed manufacturers in the original sample. Many companies stay in the distress zone for years without defaulting, because of resources the five ratios do not measure.

Can the Altman Z-score be used for banks?

No. Banks and insurers were not in the sample, and their balance sheets make several of the ratios meaningless. A bank is judged on capital ratios, credit losses and funding instead.

Why do two websites show different Z-scores for the same company?

Because they use different inputs. The date of the market value, trailing-twelve-month or fiscal-year profit, which line is treated as EBIT and how total liabilities are counted all change the result by a few tenths.

Why does the share price affect the Z-score?

Because X4 is the market value of equity divided by total liabilities. Altman included it so that the market's assessment would enter the score, which also means the score rises and falls with the stock.

Summary

The Altman Z-score adds five weighted ratios (working capital, retained earnings, EBIT and sales, each over total assets, and market value over liabilities) and reads the total against two cut-offs, 1.81 and 2.99. Intel's fiscal 2025 statements and its November 17, 2026 market value give 2.67, in the grey zone, with 1.94 of that from the market-value term alone. Across seven large companies the scores run from Walmart's 6.68 to Kraft Heinz's 0.50, and the reasons include fast asset turnover, buybacks, a goodwill write-down and a lending business, none of which the 1968 sample of manufacturers contained. Use the score as a prompt to open the balance sheet, and never as a verdict on a named company.

Check your understanding

5 questions on this lesson. Each answer comes with the reason, and nothing is scored.

  1. An invented manufacturer has X1 = 0.10, X2 = 0.20, X3 = 0.10, X4 = 1.00 and X5 = 1.00. What is its Z-score on the original formula, and which zone is it in?

    1. 2.40, safe zone
    2. 2.33, grey zone
    3. 1.40, distress zone
    Show the answer

    B. 2.33, grey zone 1.2 × 0.10 + 1.4 × 0.20 + 3.3 × 0.10 + 0.6 × 1.00 + 1.0 × 1.00 = 0.12 + 0.28 + 0.33 + 0.60 + 1.00 = 2.33. That is between 1.81 and 2.99, the grey zone.

  2. A company’s statements do not change, but its share price falls by half. What happens to its Z-score?

    1. It falls, because X4 is the market value of equity divided by total liabilities
    2. Nothing, because the score uses only the balance sheet
    3. It rises, because the shares are now cheaper
    Show the answer

    A. It falls, because X4 is the market value of equity divided by total liabilities X4 uses the market value of the shares, so its term halves. In the Intel example that alone would move the score from 2.67 to about 1.7.

  3. Why is the original Z-score a poor tool for a bank?

    1. Banks do not publish balance sheets
    2. Banks always have negative retained earnings
    3. The model was fitted to manufacturers, and a bank’s deposits make its liabilities huge against its equity by design
    Show the answer

    C. The model was fitted to manufacturers, and a bank’s deposits make its liabilities huge against its equity by design The 1968 sample held only manufacturers. A bank funds itself with deposits, so market value over liabilities is always small and working capital has no meaning, whatever the bank’s health.

  4. A consumer-brand company has spent most of its past profit on buybacks and records a goodwill impairment this year. Which terms of its Z-score are pushed down?

    1. Only X5, sales ÷ assets
    2. X2 (retained earnings) and X3 (EBIT), even if it can pay its debts comfortably
    3. None, because buybacks and impairments are non-operating
    Show the answer

    B. X2 (retained earnings) and X3 (EBIT), even if it can pay its debts comfortably Buybacks can reduce retained earnings, and an impairment reduces operating income for the year. Both lower the score without telling you whether the company can meet its obligations.

  5. A large company scores 1.4, in the distress zone. What is the most accurate reading?

    1. It will file for bankruptcy within two years
    2. Its shares are undervalued
    3. Its ratios resemble those of failed manufacturers in a 1968 study, which is a reason to read its debt maturities and cash flow, not a prediction
    Show the answer

    C. Its ratios resemble those of failed manufacturers in a 1968 study, which is a reason to read its debt maturities and cash flow, not a prediction The zones describe groups in the original sample. For one company the score is a statistical flag; credit lines, asset sales and access to new capital are outside the five ratios.

Sources and further reading

  1. Edward I. Altman, “Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy,” Journal of Finance, vol. 23, no. 4 (1968), pp. 589–609
  2. Intel Corporation — Form 10-K for the fiscal year ended December 27, 2025, filed January 23, 2026 (SEC EDGAR)
  3. The Procter & Gamble Company — Form 10-K for the fiscal year ended June 30, 2026, filed August 4, 2026 (SEC EDGAR)
  4. Caterpillar Inc. — Form 10-K for the fiscal year ended December 31, 2025, filed February 13, 2026 (SEC EDGAR)
  5. Walmart Inc. — Form 10-K for the fiscal year ended January 31, 2026, filed March 13, 2026 (SEC EDGAR)
  6. The Kraft Heinz Company — Form 10-K for the fiscal year ended December 27, 2025, filed February 12, 2026 (SEC EDGAR)
  7. The Boeing Company — Form 10-K for the fiscal year ended December 31, 2025, filed January 30, 2026 (SEC EDGAR)
  8. Ford Motor Company — Form 10-K for the fiscal year ended December 31, 2025, filed February 11, 2026 (SEC EDGAR)
  9. SEC EDGAR — XBRL company facts (AssetsCurrent, LiabilitiesCurrent, RetainedEarningsAccumulatedDeficit, OperatingIncomeLoss, Assets, revenue) for Intel and the six other companies, read November 18, 2026
  10. Finabase chart data — closing prices of November 17, 2026; scores computed by Finabase