Discussant's response to "Analytical procedure results as substantive evidence";

              Discussant's Response to
"Analytical Procedure Results as Substantive
Evidence"
Abraham D. Akresh
Independent Consultant, former Partner, Laventhol & Horwath
I generally agree with Bill Kinney and Christine Haynes' analysis of the usefulness of analytical procedure results as substantive evidence. As I told the conference organizers when I took this assignment, I won't challenge the mathematical aspects of the paper. Instead, I would like to put the paper into a real world context and discuss how we consider their suggestions and how we use analytical procedures.
Kinney and Haynes point out the dangers of using analytical procedure results as substantive evidence. One could read the paper to mean that they believe analytical procedures should never be used as substantive evidence. I am glad that this is not their view. I agree that there are dangers in analyt-ical procedures and the auditor needs to recognize these dangers and use analytical procedures in the right time, in the right place, in the right way. I will try to explain how and when analytical procedures can be used in light of Kinney and Haynes' paper.
I have no problem with the discussion of the history of analytical proce-dures. It's nice, but it doesn't add much. Auditors have always performed an-alytical procedures. That is, they have looked at the forest and said "does it make sense?" Tests of details have gotten them into the trees; the good au-ditor needs to see both the forest and the trees. Analytical procedures are important, especially when testing the completeness and valuation asser-tions.
There are different kinds of analytical procedures and they serve differ-ent purposes. SAS 56 requires analytical procedures in planning the en-gagement and in wrapping up the engagement In these, analytical procedures serve a useful purpose as attention-directing techniques. SAS 56 does not re-quire analytical procedures to be used as substantive tests. The wording of paragraph four makes it clear - "In some cases, they can be more effective or efficient than tests of details." Obviously in some cases they are less ef-fective.
In substantive testing, there are two different ways of using analytical pro-cedures. The first is as the primary substantive test. Here, analytical proce-dures are the most important form of audit evidence. In these situations, the auditor does little or no tests of details. The second use is as a corroborative test. Here the auditor performs analytical procedures, but he or she also per-forms detailed substantive tests. In determining sample size, the auditor
104