We calculate all the scores ourselves based on the Reuters data-set and the original Altman paper. You can use the popup to see the details of the calculation. Do feel free to contact us via the Green Support Messenger if you disagree with the calculation.
Altman generally works well as a risk indicator - in its initial test, the Altman Z-Score was found to be 72% accurate in predicting bankruptcy two years prior to the event - but no statistical indicator will ever get it right 100% of the time. That's why we provide the popups for the "red flags" so users can look into specific company factors that might lead someone to draw a different conclusion.
For that reason, it's also worth looking at the Altman score in combination with other health indicators like the Piotroski F-Score.
In summary, some of the issues with the Altman Z-Score are:
One other limitation of the Z-Score relates to some negative working capital companies. Negative working capital means that current liabilities exceed current assets. The reason it's used in the Altman calculation is that this fact is a strong indicator that the company could be in serious financial trouble. However, this is ambiguous, as it can also be a sign of managerial or business efficiency - for example, it might be a business with low inventory and accounts receivable (which means they operate on effectively a cash basis). We are not aware of any reliable adjustment factor that has been applied to weed out "good" negative working capital companies.
Although there is a more generic Z2 score, the original Z-Score was originally designed for "industrial companies". It is not designed for use with financial companies so we do not print these scores. This is because of the opacity of financial companies' financial statements and their frequent off-balance sheet items.
See this blog for further discussion of this point which suggests a better way of forecasting solvency risk for financials is by looking for: