Discussion of "A Model of errors and irregularities as a general framework for risk-based audit planning";

              Discussion of "A Model of Errors and Irregularities as a General Framework for Risk-Based Audit Planning"
Timothy B. Bell
Director, Assurance Services, KPMG Peat Marwick
Francis and Grimlund (henceforth F&G) develop a simple taxonomy of financial statement transactions and evaluate classes of transactions in terms of their susceptibility to error, misappropriation, and fraudulent misrepresentation. The paper provides insightful discussions about the ways errors and irregularities can happen and "what to audit and why." I begin my discussion by providing a brief summary of the risk-based audit planning model presented in the paper. I then compare the F&G model to the approach outlined in Houghton and Fogarty [1991] and attempt to reconcile any apparent differences between these two approaches. Remaining remarks focus on what I believe are key audit planning issues not discussed in the paper, and current trends within the auditing profession that indicate a movement away from a transactions orientation for audit planning and toward a more holistic business orientation.
The F&G Risk-Based Audit Planning Model
In order to better understand the susceptibility of "accounting populations" to errors and irregularities, F&G subdivide transactions into distinct groups based on the timing of the recording of the underlying economic events, and the circumstances that trigger the recording of transactions. An accounting transaction is categorized by F&G as "incomplete" if, at the end of the accounting period, an account balance related to the transaction will remain on the books, and one can reasonably expect at least one additional accounting transaction will be recorded in any future accounting period related to the same underlying event.1 All other accounting transactions are defined as "completed." For example, a credit sale (purchase) is an incomplete transaction prior to collection (payment) of the related receivable (payable). The transaction is completed when cash is collected or paid. F&G's audit planning model is based, in part, on the assumption that incomplete and completed transactions have different propensities for errors and irregularities.
The recording of accounting transactions is sometimes — but not always — triggered by the occurrence of external events. F&G subdivide completed and incomplete transactions based on the triggering event — external event was, or was not, the trigger. The F&G audit planning model is also based on the assumption that transactions for which recording is triggered by external events have different propensities for errors and irregularities than those for which recording is not triggered by external events.
Table 1 summarizes the audit planning implications that fall out of their analysis of propensity for errors and irregularities for each of these classes of transactions.
According to F&G, externally-triggered completed transactions have a low risk of random error or fraud because of the confluence of the control systems of all involved parties over the recording process. F&G suggest that completed transactions for which recording is not triggered by external events should be tested for asset misappropriation and random error because they are not subjected to the same degree of control. However, they suggest the risk of fraudulent account balance manipulations is very low in all completed transactions because they would normally be detected by routine testing of the ending cash balance. Finally, F&G suggest that all incomplete transactions should be tested for random errors and fraud, but not for asset misappropriation with concealment,
F&G define "incomplete" transactions as "those economic exchanges that remain incomplete at the end of the fiscal period." This is my attempt to make the definition more precise.
155