The Life of Audit Sampling Techniques to Test Inventory
Windsor, Sean, Journal of Accountancy
THE USE OF AUDIT SAMPLING TECHNIQUES TO TEST INVENTORY
Although audit sampling is a useful technique, practitioners often have problems using Statement on Auditing Standards no. 39, Audit Sampling, to test inventory substantively. This article identifies some of the problems and possible solutions.
Consider the case of Company X, a manufacturer that is being audited. Company X's fiscal yearend is December 31. However, it will not take a physical inventory until January 31. Therefore, the auditor will have to test the company's rollback to December 31.
It is important for the auditor to observe the physical inventory of Company X. To determine if the client is correctly counting the inventory, the auditor select 30 inventory tags from the 250 used and recounts the related inventory. The auditor finds 3 tags with errors. An investigation of the errors determines that all three of the errors were made by the same person.
Should the auditor
* Assume this is a sampling application and project the error results to the entire population?
* Assume Company X's inventory counting procedures are being tested and (1) have the client recount all the areas counted by the person who made the error or (2) have the client recount the entire inventory?
The answer is not as obvious as it might seem. The auditor could project the inventory count errors to the entire population if he or she believed the errors noted were representative of that population and were not systematic.
However, if the errors involve 3 missing bolts in a box of 50, 2 missing tape recorders in a box of 20 and 1 missing personal computer on a pallet of 30, will the projected error be meaningful? And how does the auditor adjust the inventory tags to reflect the projected error?
The objective of the auditor's test is to determine whether the client has counted the inventory accurately, not to determine the correct amount of inventory items. Since all three errors were made by the same person, it appears more appropriate to have the client recount all the areas originally counted by that individual, instead of recounting the entire inventory. However, if the auditor determines that several counters made the errors, he might consider a recount of the entire inventory, depending on materiality and other factors.
Company X had raw material receipts totaling $5 million between January 1 and January 31. Five of the receipts are each over $500,000 and total $4.5 million. Forty other receipts, all less than $20,000, make up the remaining $500,000. The materiality level is $400,000.
If the auditor vouches all five receipts over $500,000 and selects the next seven largest receipts, which total $125,000, has he
* Performed audit sampling according to SAS no. 39?
* Examined certain items 100%, which is not a sampling application?
Clearly, the auditor has examined less than 100% of the raw material receipts. However, SAS no. 39 says, "Any items that the auditor has decided to examine 100 percent are not part of the items subject to sampling. …