
A Look at Altman's ZScore
By Cindy Moorhead
Moorhead Management Services
Question: What is Altman's zscore and how can I use it in my analysis
of customer financial statements? Answer: Altman's zscore is a statistical
ratio model developed by Edward I. Altman to predict the probability
of bankruptcy within two years. While some credit people tend to use
this as a magical formula that can predict bankruptcy, it really is
First, the modeler would identify some criteria, such as bankruptcy, for failing
firms. They then pick a sampling of firms who meet the criteria. They must
pick enough firms that meet the criteria for results to be considered statistically
valid.
Once they have identified firms having the desired criteria, they would find
a similar group of companies, with the only difference being these firms are
financially healthy.
Financial statements of these two types of companies would be entered into
a database. With the help of computer analysis, a determination would be made
of which financial ratios are consistently and significantly different for
a healthy and bankrupt company.
Last, a scoring system is developed to weight the importance of the different
ratios.
There are actually three different zscores that have been developed by Edward
Altman. The original zscore was developed in 1968. This formula was developed
for public manufacturing firms and eliminated all firms with assets less than
$1 million. This original model was not intended for small, nonmanufacturing,
or nonpublic companies, yet many credit managers today still use the original
zscore for all types of customers.
Altman later made two additional models (sometimes referred to as model "A" and
model "B") to the original zscore. In 1983, the model "A" zscore was developed
for use with private manufacturing companies. The weighting of the various
ratios is different for this model as well as the overall predictability scoring.
In addition, while the original score used the market value of equity to calculate
the equity to debt formula, model "A" used stockholder's equity on the balance
sheet.
Model "B" was developed for private general firms and included the service
sector. In this statistical model, the ratio of sales to total assets is not
used, the weighting on this model is different, and the scoring is, again,
different.
Although computerized statistical modeling would aid in determining the weighting
of each ratio, common sense helps us understand the purpose of each ratio used.
All three models use return on total assets, working capital to total assets,
retained earnings to total assets and the equity to debt ratio. In addition,
the original model and model "A" also used sales to total assets in the calculation.
Return on total assets is the ratio that has the highest weighting in each
of the three models. This would be the earnings before interest and taxes divided
by total assets. It is a measure of how efficiently a company operates before
financial and tax considerations are taken into account. It makes sense that,
in order to have longterm viability, a company must be able to efficiently
produce a profit. The higher the profit generated in relation to assets being
used, the stronger the company.
Working capital to total assets is another ratio used in the model. Working
capital is an indication of liquidity and this formula would measure this liquidity
in comparison to the size of all assets. Most firms file bankruptcy because
(for various reasons) they cannot pay their bills, therefore it makes sense
that some form of liquidity would be used in predicting bankruptcy.
Retained earnings to total assets is another formula used in the model. Longterm
profitability accumulates in retained earnings. Many companies who file bankruptcy
are new companies who have not yet had a period of time to accumulate profits.
Therefore, it would make sense that firms who have accumulated profits into
retained earnings over many years would be less likely to file bankruptcy.
Equity to debt is a formula in the model that would put a weight on the leverage
of a company. The higher the debt in proportion to the equity, the more riskier
a firm is considered.
The last ratio of sales to total assets is used only in the original zscore
and the model "A" scoring. It is a measure of how efficiently the total assets
are used to generate sales. Because this ratio varies greatly from industry
to industry, it was not included in the model "B" scoring.
The zscore models were developed using large companies. The original model
eliminated all companies with assets less than $1 million and the third model
used assets averaging approximately $100 million. Small firms may have very
different ratios than large companies. Therefore, none of the zscores may
be appropriate for small companies.
I have found in working with the zscore that it tends to be a quick look at
the likelihood of a company filing bankruptcy. However, if you have already
done good analysis of the financial statements and used your analysis to understand
what is going on with your customer, you probably have already come to the
conclusion as to whether or not your customer is on shaky financial ground.
Using the zscore, in my opinion, does not give you a magical answer. If your
customer is unprofitable, has negative retained earnings and is highly leveraged,
chances are the statistical model of the zscore will also show you that this
company is heading towards trouble.
Cindy Moorhead is owner of Moorhead Management Services. She specializes
in training credit departments to spot red flags in customer financial
trends. Her email is cindy@moorheadmgmt.com. Her website is www.moorheadmgmt.com.
Her email is cindy@moorheadmgmt.com
Reprinted by permission from Trade Vendor Quarterly
Blakeley & Blakeley
LLP Spring 01 



