McGladrey & Pullen, LLP
400 Locust Street, Suite 640
Des Moines, Iowa 50309
O 515.284.8660 F 515.284.1545
December 31, 2002
Johnathan G. Katz, Secretary
U.S. Securities and Exchange Commission
450 Fifth Street NW
Washington, DC 20549-0609
RE: File No. S7-46-02
Dear Mr. Katz:
We apologize for being late with our submission, but hope that you will nevertheless consider the following general and specific comments regarding the Commission's Proposed Rule, Retention of Records Relevant to Audits and Reviews:
Record retention rules should enhance audit quality by creating a reasonable standard for retention of relevant records created, sent or received by an accountant in connection with an audit or review.
The Commission has the authority and the obligation, under Section 802 of the Sarbanes-Oxley Act, to promulgate "such rules and regulations as are reasonably necessary relating to the retention of relevant records such as workpapers, documents that form the basis of the audit or review, memoranda, correspondence communications, other documents and records (including electronic records) which are created, sent or received in connection with an audit or review and contain conclusions, opinions, analyses or financial data relating to such an audit or review" [emphasis added].
Record retention rules should facilitate and enhance audit quality. Rules designed to facilitate investigations or litigation against accountants would have the opposite effect. The key to promulgating record retention rules that enhance audit quality lies in the word "relevant". The proposed rule would require an accountant to capture and retain, at significant cost and for no worthwhile purpose, irrelevant, incomplete and immaterial records.
We support the following principles for retention of records relevant to audits or reviews:
- An accountant performing an audit or review should document procedures performed, evidence obtained and conclusions reached as required by applicable professional standards and should retain such documentation for at least the period required by law or regulation.
- An accountant performing an audit or review should not be expected or required to retain drafts or other documentation expressing tentative opinions or conclusions, or incomplete or irrelevant analyses or financial data.
- An accountant performing an audit or review should retain all records created, sent or received in connection with the audit or review if they are material to his or her final opinions, conclusions and completed analyses.
- An accountant performing an audit or review should not be expected or required to retain all evidence obtained or document all procedures performed during the course of the audit or review.
We support a requirement for an accountant performing an audit or review to document procedures performed, evidence obtained and conclusions reached as required by professional standards and retain such document for at least the period required by law or regulation.
Currently, professional standards with respect to audit documentation (also referred to as workpapers or working papers) are contained in Statement on Auditing Standards (SAS) No. 96, Audit Documentation and in various other SASs as indicated in Appendix A to SAS No. 96. Existing professional standards do not contemplate that the accountant would retain all evidence obtained or create a record of all procedures performed during an audit or review. Rather, the accountant creates a record of those procedures performed and the evidence obtained that (a) relate to material matters, and (b) is relevant to the opinions or other conclusions expressed in his or her work product. We believe that where a procedure performed or evidence obtained is deemed by the accountant to either, (a) not relate to material matters, or (b) to be irrelevant or unnecessary, it need not be (and customarily is not) documented or retained.
A requirement to retain all records that cast doubt upon the opinions or other conclusions expressed in the accountant's work product is counterproductive to the goal of enhancing audit quality.
Current professional standards require an accountant performing an audit or review to evaluate all evidence obtained and the results of all procedures performed, regardless of whether they corroborate or cast doubt upon management's assertions. These same standards require an accountant to obtain sufficient relevant evidence in support of the opinions or other conclusions expressed in his or her work product to overcome any evidence that casts doubt upon those opinions or other conclusions.
Under current professional standards, any unresolved difference of opinion among members of the engagement team or between members of the engagement team and a firm consultant, along with the firm's resolution of the matter, must be documented. Initial differences of opinion that are later resolved to the satisfaction of all involved need not be documented; however, the resolution of the matter would ordinarily be documented as a significant audit finding or issue.
A rule that embodies the foregoing principles would enhance audit quality; however, the proposed rule goes well beyond these principles without producing any incremental benefit in the form of improved audit quality.
We support a requirement to retain workpapers and other records created, sent or received in connection with an audit or review provided they relate to material matters and are complete.
The proposed rule would require an accountant to retain workpapers (defined as documentation of auditing or review procedures applied, evidence obtained, and conclusions reached by the accountant in the audit or review engagement as required by standards established or adopted by the Commission or by the Public Company Accounting Oversight Board) and other documents that form the basis of the audit or review. Hereinafter, we will refer to this category of records as "workpapers". We interpret this requirement to permit an accountant to exercise some discretion in determining whether, once created, a "workpaper" must be retained.
The proposed rule would also require an accountant to retain memoranda, correspondence, communications, other documents, and records (including electronic records which, (a) are created, sent or received in connection with the audit or review, and (b) contain conclusions, opinions, analyses, or financial data related to the audit or review. Hereinafter, we will refer to this category of records as "other records". We interpret this requirement to require the retention of all records within this category, regardless of whether they relate to material matters or are complete.
During the course of an audit or review, many records containing conclusions, opinions, analyses, or financial data related to the audit or review are created, received or sent. We believe that those records that contain the accountant's final conclusions and opinions with respect to material matters and all relevant analyses and financial data should, under existing professional standards, be incorporated in the audit documentation (workpapers) retained by the auditor. However, the proposed rule appears to require the retention of all such records. If so, the accountant would be required to retain tentative conclusions, draft opinions, incomplete analyses or research, and irrelevant financial data merely because it was created, received or sent in connection with the audit or review. We believe that retention of these types of records is contrary to the public interest because they serve to confuse, rather than clarify, the accountant's final judgments, conclusions and opinions.
Different standards for retention of "workpapers" and "other records" would lead to unnecessary debate and legal wrangling.
The creation of different standards for the retention of the two categories of records would result in protracted and unnecessary professional debate and legal wrangling over whether a particular record is a "workpaper" or an "other record". The standard for retention of "other records" should be the same as the standard for retention of "workpapers". The appropriate standard for both categories of records should be those that form the basis for the accountant's work product, including records of circumstances, evidence, findings or results related to material matters that cast doubt on the conclusions expressed in the final work product. This standard would allow the accountant to exercise appropriate professional judgment concerning whether the records related to material matters and whether initial doubt had been overcome or tentative conclusions had been revised based upon the accountant having obtained additional evidence or facts pertaining to the matter.
Contrary to the public interest, the proposed rule would inhibit open dialogue between the client and the engagement team, amongst members of the engagement team, and between the engagement team and firm consultants.
Technological advancements, especially the Internet, have created an environment in which information can be shared quickly and easily without the need for direct interactive communication. As a result, clients send data files and other information to their accountants via electronic mail, and accountants use electronic mail for both internal and external communication. These technological advances have greatly contributed to the efficiency of an audit or review and are especially useful in global engagements spanning worldwide time zones. Requiring the retention of virtually all such communications containing tentative conclusions or opinions, and incomplete analyses or research, would inhibit the free flow of information between clients and their accountants, members of the engagement team and the engagement team and firm consultants. Accountants would resort to inherently more unreliable oral communication, the results of which would be less likely to be universally understood or appropriately documented.
The cost of complying with the proposed rule would be significant.
The Commission's conclusion that there would be no significant increase in costs for accounting firms or issuers as a result of the proposed rules because all of the records exist at some point during the course of an audit or review is erroneous. Many records that are required to be retained under the proposed rule are currently retained for only a very brief period of time. In the case of our firm alone, we believe the cost of the systems and procedures necessary to provide reasonable assurance that all "other records" are captured and retained would be substantial.
Our firm currently has no standard firm-wide systems for storing and retrieving electronic records other than those records contained in our "paperless audit" (electronic workpapers) database. The cost of implementing, and monitoring compliance with, systems to capture, store and retrieve "other records" would be significant. For example, assuring that we retained all e-mail communications meeting the criteria for retention as "other records" would require us to implement a new system that would require an identification of the client and the engagement in all communications, as well as the adoption of different systems and procedures for backing up our file servers and retaining back-up tape files.
As a practical matter, most e-mail systems retain messages based on the instructions of either the sender or receiver(s) of those messages. Frequently, those messages are received by multiple parties who individually decide whether to respond to the message and whether to retain or delete the message. In addition, senders can "blind copy" some recipients so that other recipients are not aware of everyone who received the message, and respondents can elect to include the original text (or a portion of the original text) with their reply and to respond to all or only certain of the recipients. Finally, either the sender or any recipients can forward the message to others, and in doing so, can modify any prior text they elect to include. As a result, the integrity and security of these systems are not reliable and it would be nearly impossible to assure that all communications that are created are retained.
Are the "workpapers" and other documents that would be required to be retained under this proposed rule sufficiently described? If not, what changes should be made to provide for greater clarity? Are there alternative definitions that would better implement section 802?
It is doubtful that any record retention rule is capable of providing clear and adequate notice of the difference between "workpapers" and "other records". As a result, it is not practical to establish separate standards for their retention. In addition, for the reasons previously cited, we are opposed to separate retention standards even assuming clearly distinguishable definitions can be developed.
Would auditors have to implement significant changes to their retention policies or internal control processes and procedures, as well as system upgrades, to ensure compliance with the proposed rule? If so, what types of changes most likely would be required? How can we minimize any required changes consistent with section 802?
Significant policy and procedure changes would be required and accountants would incur significant expense if the rule regarding the retention of "other records" is adopted as proposed. Consistent with section 802, the proposed rule should be modified to provide the same standard for retention of "other records" as for the retention of "workpapers".
Would auditors circumvent the proposed record retention requirements by, for example, replacing written communications with oral communications? If so, what additional measures should be taken?
Contrary to the public interest, the proposed rule would cause accountants to significantly limit written communications to ensure that they are captured in accordance with the rule. As a result, the proposed rule would encourage accountants to substitute oral communication for written communication, and that any additional measures taken to the contrary would not be effective.
Section 103 of the Sarbanes-Oxley Act directs the Public Company Accounting Oversight Board to adopt an auditing standard that requires each registered public accounting firm to retain for a period of not less than seven years audit workpapers and other information that support the conclusions in the auditor's report. Should the retention period in the proposed rules be extended to seven years to coincide with the retention period in section 103? Why?
In order to avoid debate and argument over whether a particular record is a "workpaper" or an "other record", the retention period as well as the standard for retention of "workpapers" and "other records" should be the same.
Should the retention period be for some other appropriate period based on consideration of other factors, such as the utility of the records to investors, regulators or litigants, the cost of retaining the records, or the size of the accounting firm?
Consistent with our current internal requirement and other common legal and regulatory requirements, we support a retention period of seven years.
Audits of the financial statements of many investment advisers and broker-dealers would not be subject to the proposed rules because they are not "issuers" of securities. Should the proposals be amended to apply the retention period to audits of the financial statements of these entities? Why?
As "public interest" entities, the requirements should extend to investment advisers and broker-dealers.
The proposed rules would incorporate the definition of "issuer" in new section 10A(f) of the Exchange Act? Should "issuer" be defined more broadly to include any issuer of securities with respect to which a registration statement or report is filed with the Commission? Why?
As a practical matter, we expect that most firms will adopt record retention policies that assure compliance with the "lowest common denominator". As a result, broadening the definition would have little impact on accountants.
Should there be a document retention requirement for issuers as well as auditors? If yes, what would be the scope and nature of that requirement? For example, should issuers be required to retain records that the auditor reviewed but did not include in the audit workpapers? Should issuers be required to keep copies of all correspondence with the auditors and copies of documents provided to the auditors?
Issuers are subject to a variety of legal and regulatory requirements related to the creation and retention of books and records. It is challenge enough for an issuer to assure compliance with those requirements, and it would be literally impossible for an issuer to assure retention of all records reviewed by the auditor. We believe that requiring issuers to retain copies of all correspondence with the auditors would be difficult, if not impossible, to implement in global audit environments and would result in significant additional cost to the issuers.
Does the "cast doubt on the final conclusions reached by the auditor" provision in the proposed rules adequately capture the scope of the retention requirements under the Sarbanes-Oxley Act? Should the scope be narrower or broader? Would a different test be more appropriate, such as "significant differences in professional judgment," or "differences of opinion on issues that are material to the issuer's financial statements or to the auditor's final conclusions regarding any audit or review"?
The proposed scope is too broad and would have unintended consequences. We support a different test along the lines of those suggested.
Should the rules include other examples of materials that "cast doubt" on auditors' conclusions? If so, what examples should be included?
During the course of performing an audit or review, an accountant frequently obtains evidence that casts doubt on management's assertions and the auditors' conclusions. For example, tests of controls may identify deviations from control procedures or substantive tests may identify financial statement misstatements. Singling out "differences of opinion concerning accounting and auditing issues", especially if such differences resulted from an incomplete understanding of all relevant facts and were later resolved, may serve to inappropriately narrow the focus of the requirement, and at the same time focus the documentation on the process by which matters were resolved rather than the resolution of the matter.
Are there any other costs or benefits that we have not identified? For example, would the cost of audits increase? Please describe any such costs and provide relevant data.
The proposed rules would increase costs because they would require accountants to develop systems and procedures for retention of "other records" that don't exist currently and because they would encourage the substitution of less efficient and effective forms of communication. In our firm, we estimate that the costs of developing and maintaining the systems and procedures for the storage and retrieval of "other records" would be substantial.
Are there additional costs related to the proposed rules? If there are, please identify them and provide supporting data.
If the rules are adopted as proposed, we expect to incur significant costs to monitor compliance with the retention requirements for "other records". In addition, we expect there will be increased professional liability insurance costs, as well as increased legal costs associated with proving, on a case-by-case basis, that all required records were retained.
We request comments on the reasonableness of the burden hours, cost estimates, and underlying assumptions related to the proposed disclosures.
As previously stated, the Commission's assumptions regarding the existence of systems for the capture, storage and retrieval of "other records" and the development and monitoring efforts required to comply with the proposed rules are unrealistic, resulting in estimates of burden hours that are understated by significant multiples.
Thank you for considering our comments. Questions regarding these or other matters may be directed to Bill Travis, Managing Partner (952.921.7780) or Bruce Webb, National Director of Auditing (515.281.9240).
McGladrey & Pullen, LLP