Altman Z-Score: assessing a company's bankruptcy risk

The Altman Z-Score estimates a company's risk of default from five ratios drawn from its balance sheet, its income statement and its market value. In the original version, a score above 2.99 puts the company in the safe zone, a score below 1.81 in the distress zone, and anything in between in the grey zone. It measures risk, not quality or price: its main use is spotting fragile situations.

Where does the Z-Score come from?

Edward Altman, a professor at New York University, published it in 1968 in the Journal of Finance. He compared 33 US manufacturers that went bankrupt between 1946 and 1965 with 33 similar companies that stayed solvent, then looked for the combination of ratios that best separated the two groups (a discriminant analysis).

On that sample the model correctly classified 95% of the companies one year before bankruptcy, and about 72% two years before. Those figures belong to the original sample; on other periods and markets accuracy is lower, which is why variants were developed later.

The Altman Z-Score formula

Z = 1.2 × X1 + 1.4 × X2 + 3.3 × X3 + 0.6 × X4 + 1.0 × X5

The heaviest weight goes to X3: a company that no longer earns anything from its assets is in danger whatever its size. X4 is the only market-based ratio, which makes the score sensitive to share-price moves.

A worked example: a fictitious company

Take Company B, a fictitious manufacturer (in € millions): total assets 2,000, working capital 300, retained earnings 500, operating income 160, market capitalisation 1,800, total liabilities 1,200, revenue 2,400.

At 2.89, Company B sits in the grey zone, just below the 2.99 line. Now suppose its share price falls 40% and nothing else changes: market capitalisation drops to 1,080, X4 becomes 0.90 and the Z-Score 2.53. The score has worsened while the business is exactly the same. That is both the strength and the weakness of X4: it captures what the market expects, and its excesses too.

How to read the result

A score in the distress zone does not announce a bankruptcy: it flags a financial profile that resembles companies that defaulted. Conversely, the safe zone offers no protection from a sudden shock (fraud, loss of a major customer, a liquidity crunch).

Z-Score variants

The 1968 formula was built for listed manufacturers. Altman later proposed others:

Dropping X5 in Z'' matters: asset turnover varies enormously across sectors (a retailer turns its assets far faster than an infrastructure operator), which distorts comparisons.

What are the limits?

• Banks and insurers do not fit. Their leverage is structural — it is their business. Applied as is, the formula puts almost all of them in distress.
• Cash-rich companies with low retained earnings (after large buybacks or past losses) can show a weak X2 without being in danger.
• The score depends on the share price through X4, so it amplifies market moves.
• The data lags: balance sheet and income statement come from the last published year.

What our universe shows

On 7 November 2026, across the 481 non-financial companies in the universe InvestIQ follows, 60% were in the safe zone, 19% in the grey zone and 21% in the distress zone on the original formula. One in five in distress among large listed companies is a lot: it is the limits above at work, notably heavy buybacks that shrink retained earnings and equity, and coefficients built for manufacturers applied to service companies. Hence the value of reading the score alongside cash, debt and the Piotroski F-Score, and of going back to the reported accounts (company financials). The figures are updated monthly on the stock market statistics page.

Source: Altman, E. (1968), "Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy", The Journal of Finance, vol. 23, no. 4, pp. 589–609 (doi:10.1111/j.1540-6261.1968.tb00843.x).

How InvestIQ uses the Z-Score

In InvestIQ's engine the Altman Z-Score makes up 35% of the "quality" group, alongside the Beneish M-Score, a screen for earnings manipulation, and capital efficiency. For banks, insurers and real-estate trusts the engine uses adapted measures instead of the standard grid. The details are in the methodology, and where the data comes from in data sources.

The Z-Score reads well alongside the Piotroski F-Score: the first asks "could this company default?", the second "are its accounts improving?".

In short

The Altman Z-Score condenses five ratios into one number that places a company against the profile of companies that failed. It is a useful safeguard at the start of an analysis, provided the right variant is used for the right kind of company and it is kept away from the financial sector.

Educational guidance only, not investment advice.