Oracle auditing for FAS 48 Oracle Auditing Tips by Donald Burleson

 

For related notes, see my notes on FAS 48 and income statements for publicly held corporations.

 

---

When auditing income statements it is easy to verify the actual revenue dollars received, but more challenging to verify the method used to accurately forecast returns.

Auditing Oracle Financials requires an intimate knowledge of Oracle at several levels:

• The database layer - An auditor must understand the Oracle schema tables where the relevant financial data is stored, and be able to track changes over time using tools such as Oracle LogMiner.
   

• The applications layer - The auditor must be familiar with the functional inputs within all modules of Oracle Financials (e.g. AR, AP. GL), and well as external modules that feed data to the Financials component (e.g. INV, MFG).  To effectively track the evolution of financial statements, the auditor must know the specific screen names and their interfaces to the underlying database structure.
   

• The functional layer - It is not uncommon to have external systems which utilize data from Oracle Financials, and the auditor must be able to track-backwards from the audited financial statements, reconstituting the evolution of the document over time.

A complete audit of income and "expected returns" would need to cover all of these areas.

Auditing for estimated returns

Pursuant to FAS 48, SAB 104 and FAS 5, corporations must use the historical sales database to create "reasonable" estimates of future returns using valid statistical methods.

To audit the figures for estimated returns, we would need to identify these data factors:

• Historical sales and returns - We must identify the specific Oracle eBusiness Suite modules that contain historical shipment data (the INV module).  Identify the database tables that contain this data and all online screens that allow changes to this information.
   
• Income received and refunds issued - The Oracle AR module records actual sales and the AP module records refunds for returned goods.  An auditor would need to identify the database tables and all online screens related to this data.

Once we have identified the sources of the data used to forecast estimated returns, we need to examine these functional business methods:

• Returns estimation method - We identify the specific Oracle Financials module that is used for estimating returns from sales, identifying the database tables and all online screens and output reports.
   
• Financial reporting method - The Oracle Financials output modules for producing income statements would be examined for the actual data sources.

 

This article notes some issues with accounting for estimated returns:

"SAB 101, along with SAB 104, further precludes such immediate recognition of revenue under the following conditions:

- Lack of ?visibility,? or the inability to determine, or observe, the levels of inventory in a distribution channel and the current level of sales to end users;

- Significant increases in, or excess levels of, inventory in a distribution channel (sometimes referred to as channel stuffing);

- Expected introductions of new products that may result in the technological obsolescence of, and larger than expected returns of, current products;

- The significance of a particular distributor to the firm's business, sales, and marketing; and

- The newness of a product.

This approach is fraught with danger for a company that has a high proportion of sell-through revenue. In 2000, Lucent Technologies restated its earnings, decreasing revenues by $679 million primarily because of a $452 million reversal of previously booked sales to distributors that were subsequently returned
(see www.aicpa.org/download/antifraud/120.ppt).

Similarly, Bristol-Myers Squibb had to restate its earnings by $1.5 billion primarily due to improper booking of sales through distributors (Glass Lewis & Co., in its Revenue Recognition Trend Report dated May 4, 2004, available at www.glasslewis.com/solutions/trends.php).

In the opinion of FASB and the SEC, these circumstances make reasonable estimate of returns difficult, making Option 1 the preferred choice for only a limited set of companies. Also, this method is not available for companies which have other clauses in the sales contracts that make the sales price not fixed or determinable."