Predicting financial distress
Altman Z-Score
The Altman Z-Score is a bankruptcy prediction model developed by Professor Edward Altman at NYU Stern in 1968. We use the Z''-Score variant, which is adapted for non-manufacturing companies and works across industries — except financial companies.
What Is the Z-Score?
The original Altman Z-Score uses five financial ratios to predict the probability that a company will go bankrupt within two years. It was developed using discriminant analysis on a sample of 66 firms — 33 that went bankrupt and 33 that did not. The model has been revised multiple times; we use the Z''-Score (1993 revision), which drops the Sales/Total Assets ratio to make it applicable across industries, not just manufacturing.
Why Z''-Score Instead of Z-Score?
The original Z-Score includes X5 (Sales / Total Assets), which varies dramatically by industry — a retailer might have X5 = 3.0 while a software company has X5 = 0.5. The Z''-Score removes this variable and re-weights the remaining four, making it fair to compare across sectors. This is the variant recommended by Altman himself for general use.
The 4 Variables
Measures short-term liquidity. A company with negative working capital has more current liabilities than current assets — it may struggle to meet near-term obligations. Consistently negative X1 is a distress signal.
Measures cumulative profitability relative to size. Young companies or those with large accumulated losses will score low. Companies with negative equity (from buybacks or losses) can have deeply negative X2.
Measures operating efficiency — how much the company earns from its assets before interest and tax. This is the most powerful predictor of bankruptcy in the model.
Measures solvency — how much the company's assets exceed its liabilities. A ratio below 1.0 means liabilities exceed equity. Companies with negative book equity (e.g., from aggressive buybacks) will have negative X4.
The Formula
Z'' = 6.56×X1 + 3.26×X2 + 6.72×X3 + 1.05×X4Note the heavy weighting on X1 (6.56) and X3 (6.72). Working capital and operating profitability are the strongest predictors of distress in this model.
Thresholds
The company has a strong financial position. Bankruptcy risk within 2 years is very low.
Uncertain territory. The company shows some stress signals but is not in imminent danger. Warrants monitoring.
The company's financial profile matches historically bankrupt firms. Does NOT guarantee bankruptcy — many distressed companies restructure successfully.
Not Applicable to Financial Companies
Banks, insurance companies, REITs, and other financial institutions have fundamentally different balance sheets. Their "working capital" and "total liabilities" have entirely different meanings — a bank's deposits are liabilities, but they're also the raw material of its business. We skip the Z-Score for financial companies and note "N/A — Financial" in our reports.
How We Use Z-Score
- *We calculate the Z''-Score for every non-financial stock we grade, using the most recent 10-K data.
- *A Z''-Score in the distress zone (< 1.10) is a significant flag that contributes to the overall grade.
- *We always show the raw Z''-Score and each X variable in our reports.
- *The Z-Score is used alongside (not instead of) the M-Score and F-Score — it measures distress, not manipulation.
- *Companies with negative book equity from share buybacks (like Apple, Starbucks) may score low on X4 without being distressed. We note these structural explanations in our reports.
Limitations
- *Not applicable to financial companies (banks, insurance, REITs).
- *Companies with negative book equity from buybacks (not losses) will appear distressed when they may be financially healthy.
- *The model was built on 1960s bankruptcy data. Modern companies have different capital structures and access to capital markets.
- *Z-Score does not account for off-balance-sheet liabilities, operating leases (pre-IFRS 16), or contingent liabilities.
- *A safe Z-Score does not mean the stock is a good investment — it only means bankruptcy risk is low.