Preventing Fraud: Users and Preparers Speak
A nationwide survey of the four groups most affected by new auditing standards to detect and prevent fraud shows a negative consensus on many of the crucial assumptions behind the standards and a pervasive skepticism toward both the means and ends of Congressional and professional reform.
To discover how financial statement users and independent auditors view the CPAs' responsibilities for detecting fraud, in early 1987 we randomly surveyed 1600 professionals nationwide: 400 chartered financial analysts chosen from the nationwide list of Chartered Financial Analysts (CFA); 400 bank credit evaluators from the nationwide list of Robert Morris Associates (RMA); 400 financial executives from the nationwide list of the Financial Executives Institute (FEI); and 400 CPAs from the nationwide list of members of the American Institute of Certified Public Accountants (AICPA). The equal sample size was large enough to allow significant conclusions, achieving equal levels of precision from each of the four groups. Finite population correction factors (FPCF) were computed to ascertain if using identical samples on varying populations was affected because we sampled without replacement. The four FCPFs varied from .981 to .993, which exceeded Bailey's level of .95, the point at which we can ignore the potential problem [1, p. 32]. Our purpose was to solicit opinions on how the financial reporting process should be changed in order to prevent, detect, and report fraud.
The questionnaire was pre-tested with ten members from each of the four groups. We sent out two waves of questionnaires and received 453 valid responses (106 CFA responses, 121 RMA responses, 93 FEI responses, and 133 AICPA responses), a 28.3% response rate. Tests were made for non-response bias using Oppenheim's early-late hypothesis. The results indicate no significant ([is less than] .05) differences between early and late respondents. Tests for non-response bias and the high response rate strongly suggest that there is no significant non-response bias associated with the study.
The impetus for our survey was, first, that in 1982 fraud amounted to $55 billion and has been rising since [3, p. 2]; and, second, recent Congressional hearings concerned with the bankruptcy of several companies (e.g., ESM, Drysdale Securities, and Penn Central) after they received unqualified audit opinions. Hearings by the House Energy and Commerce Subcommittee on Oversight and Investigations, beginning in June 1986, and subsequent legislation recently introduced by Congressman Wyden (D-Oregon) have led the AICPA to propose nine new standards as a potential substitute for Congressional or Securities and Exchange Commission (SEC) action. Congressman Dingell (D-Michigan), who heads the oversight committee, was quoted in September by The New York Times as saying that the accounting profession has one year to improve its performance before the government steps in.
It is not a new pattern. Over the past ten years, the AICPA has issued, under Congressional pressure, five standards to combat fraud:
1. Statement on Auditing Standard (SAS) No. 16, The Independent Auditor's
Responsibilities for the Detection of Errors or Irregularities,
2. SAS No. 17, Illegal Acts by Clients,
3. SAS No. 19, Client Representation,
4. SAS No. 20, Required Communication of Material Weaknesses in Internal
Accounting Control, and
5. SAS No. 45, Related Parties.
However, by 1986 numerous shortcomings in these standards and their application had been noted. Users were unclear about the general nature of the auditor's responsibilities as well as those regarding error and fraud. "Subject to" opinions confused users. Auditors lacked formal steps by which they could address "going concern" questions. In addition, they were required to search only for "material" errors and irregularities--without strong guidance. …