The material copied below had been published on a web site maintained by the New York Society of Security Analysts ("NYSSA"), and was accessible from a link on a summary page for the Forum Program.

 This report was presented at a February 28, 2001 meeting of the Amazon Forum addressing "Foundations of Differing Analyses."

A Credit Managers Perspective of Amazon

The Amazon Forum at the New York Society of Security Analysts on 2/28/01

Presented by Lance Ettus

Member of the Corporate Governance Committee

The current controversy over Amazon and its future prospects is primarily over predictions of how suppliers will make credit decisions.  We have therefore prepared a credit analysis of Amazon.com using a standard financial model provided by the Credit Research Foundation ("CRF") of the National Association of Credit Managers ("NACM"), available from the CRF web site at:

<http://www.crfonline.org/publications/freecfsa/cfsa-index.html>

The purpose of this is not to make a judgement on the future of Amazon, but rather to see how a credit manager would view the situation.  To do our credit manager's analysis, we completed the CRF model's spreadsheet using Amazon’s financial numbers and industry averages for what is known as the "Altman Z test."  We first took actual numbers from 1999 and 2000, put them in and got our results.  We then repeated the test to get a score for the future as predicted for the 2001 fiscal year in the model recently published by Merrill Lynch's Henry Blodget, generally recognized as an Amazon bull.  We then used a third model, which used Mr. Blodget’s pro forma EBIT numbers along with his projections.

The results were relatively unfavorable Altman Z scores, for the actual periods as well as for the future based on Mr. Blodget’s projections.

And without Blodget’s forecast.

And with the pro forma numbers

And here is an explanation of Altman Z.

Scale   

Probability of Failure     Altman Z-score                       

Unlikely                                    Z-Score =  > 3.0                                             

Undeterminable                        Z-Score = 1.81  to  2.99                                             

Very High                                 Z-Score < 1.81                                               

(1) Altman's Z-Score is a bankruptcy prediction model developed by Dr. Edward I. Altman of New York University in the mid-1960's.  The model assigns a coefficient to four financial ratios and sums the resulting numerical values to arrive at the Z-Score.                                                             

The score is then compared to the above scale as an analytical tool to predict the probability of a firm going bankrupt.                                                               

While a low score does not guarantee bankruptcy, the model has been proven accurate in many situations in the past and can be used to help evaluate the overall financial position and trends of a firm. 

This is not to say that Amazon will go bankrupt, but rather to show how a credit manager would view the company.  It should be noted that there are indeed many possible problems with the Altman Z score.

·                    Some people think Altman Z scores are best used for capital intensive industrial-type companies, and may not be suitable for Amazon even though it is a capital intensive business.

·                    The industry used as a standard of comparison could influence the results.  We used the catalog industry for an industry comparison, but could have used any of several alternatives such as direct sales, department stores, etc.

·                    There are now more complex credit scoring systems than Altman Z, and some of these more complex credit scoring systems could yield better results for Amazon.

·                    Some of the numbers we used were rough estimates, since we had to make some minor inferences from our data in order to transfer it into the format of the spreadsheet.

Finally, we note that there are significant differences between the job of a credit manager and the job of a security analyst.  The credit manager is trying to minimize risk of losses while the security analyst is looking for excess returns.  This is likely to make a credit manager more strict with his analysis.