What it means
Companies that fail tend to show a recognisable pattern before they do: shrinking working capital, accumulated losses eroding retained earnings, weak operating profit relative to assets, high liabilities relative to the value of the equity, and sluggish sales relative to the asset base. Altman's contribution was to identify which ratios best separated companies that failed from those that survived, and to weight them into a single score using statistical analysis of the two groups.
In the original study the model correctly classified about 95% of companies one year before failure and around 72% two years before. The five ratios each capture something different.
Working capital to total assets measures liquidity relative to size. Retained earnings to total assets measures cumulative profitability and, indirectly, age and leverage: young companies and those that have paid out or lost their earnings score low.
EBIT to total assets measures current operating productivity of the assets. Market value of equity to book value of total liabilities measures how far asset values can fall before liabilities exceed them, using the market's valuation.
Sales to total assets measures asset efficiency. The weights reflect each ratio's discriminating power in the original sample, with operating profitability carrying the most.
The model has known limits. It was built on US manufacturers in the 1960s, and its weights and cut-offs do not transfer perfectly to other sectors, countries or eras; Altman himself published a Z'-score for private companies (using book equity) and a Z''-score for non-manufacturers and emerging markets, with different weights and thresholds.
Financial companies are excluded because their balance sheets are structurally different. Accounting manipulation can flatter the inputs.
And a score is a probability, not a verdict: companies in the distress zone do survive, and companies in the safe zone do fail, particularly when failure comes from fraud or a sudden shock. Used with those cautions, the Z-score is a fast, objective first screen.
Lenders use it to flag borrowers for review, investors to avoid value traps, auditors as one input to going concern assessments, and suppliers to set credit terms. A falling score over several periods is more informative than any single reading.
In practice
Real-world examples.
Example
A credit insurer runs Z-scores on all 4,000 companies it insures each quarter and reviews any whose score has fallen below 1.81 or dropped by more than 0.5 in a year.
Example
A value investor screens out any stock with a Z-score below 1.81 to avoid companies whose low valuation reflects genuine distress risk.
Example
An auditor cites a client's Z'-score of 1.1, alongside its cash flow forecast and refinancing position, in documenting a going concern assessment.
Think of it
“The Altman Z-Score is like a credit score for companies. Higher scores mean lower risk of default; lower scores raise red flags.
Formula
Calculation
Original Z-score (public manufacturing companies):
Z = 1.2 x X1 + 1.4 x X2 + 3.3 x X3 + 0.6 x X4 + 1.0 x X5
where X1 = Working Capital / Total Assets; X2 = Retained Earnings / Total Assets; X3 = EBIT / Total Assets; X4 = Market Value of Equity / Total Liabilities; X5 = Sales / Total Assets
Zones: Z above 2.99 safe; 1.81 to 2.99 grey; below 1.81 distress.
Z'-score for private companies replaces X4 with Book Value of Equity / Total Liabilities and uses weights 0.717, 0.847, 3.107, 0.420 and 0.998, with zones above 2.9 safe and below 1.23 distress.
Worked example. A listed furniture manufacturer reports:
- Total assets: $400 million; current assets $160 million; current liabilities $110 million
- Retained earnings: $90 million
- EBIT: $24 million
- Market value of equity: $150 million; total liabilities: $250 million
- Sales: $520 million
X1 = ($160 million minus $110 million) / $400 million = 0.125
X2 = $90 million / $400 million = 0.225
X3 = $24 million / $400 million = 0.06
X4 = $150 million / $250 million = 0.60
X5 = $520 million / $400 million = 1.30
Z = 1.2 x 0.125 + 1.4 x 0.225 + 3.3 x 0.06 + 0.6 x 0.60 + 1.0 x 1.30
Z = 0.15 + 0.315 + 0.198 + 0.36 + 1.30 = 2.32
The company sits in the grey zone. Its sales efficiency and liquidity are reasonable, but modest operating profitability and a market value well below its liabilities pull the score down. A lender would want to see the trend: if the score was 2.9 two years ago, the direction matters more than the level.
Sensitivity: if the share price halved, X4 would fall to 0.30 and Z to 2.14. If EBIT fell to $10 million, X3 would fall to 0.025 and Z to 2.21. A combination of both would take the company to 2.03, still grey but approaching distress.Case study
Seen in the real world.
A bank's mid-market lending team had a policy of calculating the Z-score at every annual review but no policy on what to do with it. A retail chain client's score had declined from 3.1 to 2.4 to 1.7 over three years; the relationship manager noted it each year and renewed the facility each year, because the company was still profitable and its owners were long-standing clients. In the fourth year the chain entered administration owing the bank $11 million, of which it recovered $4 million.
The bank's post-mortem found that the Z-score's decline had been driven by exactly the factors the model was built to catch: retained earnings eroded by dividends, working capital shrinking as suppliers tightened terms, and leverage rising. The bank introduced a rule that any score below 1.81, or any fall of more than 0.6 in a year, triggers an independent credit review and a cash flow stress test before renewal. In the following two years the rule flagged nine borrowers; four were restructured early and none defaulted.
Watch out
Common mistakes.
- Applying the original Z-score to private companies, financial firms or service businesses without using the appropriate variant.
- Treating the score as a verdict. It is a probability estimate; use it to prioritise investigation, not to replace it.
- Reading a single score without the trend. A stable 2.2 and a 2.2 that was 3.5 last year are different situations.
Questions
People also ask.
What is a good Altman Z-score?
Above 2.99 on the original model is the safe zone; above 2.9 on the private-company version. Below 1.81 (or 1.23) signals significant distress risk.
Does the Z-score work outside manufacturing?
The original model was built on manufacturers. The Z''-score variant was developed for non-manufacturers and has been applied widely, but sector-specific models often perform better.
Can the Z-score predict fraud?
No. It reads reported figures, so a company that manipulates its accounts can show a healthy score until the manipulation is exposed.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%