Message-ID: <12497745.1075856949679.JavaMail.evans@thyme> Date: Thu, 30 Dec 1999 08:31:00 -0800 (PST) From: vince.kaminski@enron.com To: vkaminski@aol.com Subject: Audit Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-From: Vince J Kaminski X-To: vkaminski@aol.com X-cc: X-bcc: X-Folder: \Vincent_Kaminski_Jun2001_8\Notes Folders\Sent X-Origin: Kaminski-V X-FileName: vkamins.nsf ---------------------- Forwarded by Vince J Kaminski/HOU/ECT on 12/30/99 04:31 PM --------------------------- VKaminski@aol.com on 10/03/99 04:25:26 PM To: Vince J Kaminski/HOU/ECT@ECT cc: Subject: Audit Thanks for your message regarding the Model Risk project. I think it's a very important issue and really critical to Enron. The model risk represents a very significant exposure to the company: a wrong model results in bad decisions and eventually leads to accounting errors and incorrect financial statements. Here are my comments. 1. Reasons of model risk. The main source of model risk in Enron is omitted in your list. I think it's due to the fact that the responsibility for the final model approval and validation is dispersed in the organization and there no single unit/person that is vested with the authority to audit, approve and monitor valuation models. The popular perception is that this is done by the Research Group, but this could not be further form the truth. Given the way the organization evolved over the last few years, we became a group of internal consultants, working on highly specialized, partial problems (sometimes, we work on a project from A to Z, but it's an exception to the rule). It's imperative that this is communicated to the senior management in a precise, non-ambiguous way. I think that this is the main source of potential problems and the first shortcoming of the modeling process to be corrected. 2. I would expand point 3 in your draft (inappropriate application of models) and add two lower level points: a) wrong inputs b) incorrect model used for a given transaction. TVA is an example of point a: a transaction in case of which the inputs have not been being updated for an extended period of time in the past. 3. Under Control Strength you can mention that several audits of the Research Group model by Arthur Andersen and Darrell Duffie have been initiated at our insistence.