Local governmental entities in the United States are usually required to prepare comprehensive annual financial reports (CAFR) which include the general purpose financial statements (GPFS) and combining and individual fund type and account group financial statements. (See Appendix A for explanations of these terms and a brief discussion of the local government audit environment in the United States.) These reports are generally subject to external audit; the reporting entity frequently has a choice of two levels of external audit scope. At a minimum, the auditor’s report must express an opinion on the GPFS and include an opinion as to whether the combining and individual fund and account group financial statements are stated fairly in all material respects in relation to the GPFS taken as a whole (limited scope engagements). Alternatively, the audit firm can expand its scope by expressing an opinion on the GPFS and on each of the combining and individual fund and account group financial statements (expanded scope engagements).

Given consistent audit quality and efficient production of the audit, audits of comprehensive annual financial reports must use more resources than audits of the general purpose financial statements only. Generally accepted auditing standards (GAAS) suggest those additional resources should be allocated in a particular manner. No previous study has examined local government audit production costs to determine whether auditors’ adjust their audit effort as expected when audit scope differs. Anecdotally, audit practitioners have expressed concern that the audit staff may over-audit limited scope engagements. That is, they may conduct an examination sufficient to express an opinion on the entire CAFR, regardless of the scope indicated in the engagement letter. (Prior to 1981, local governmental audits encompassed the entire CAFR. Limited scope engagements were not an option. Audit staff may not have modified the audit program sufficiently to accommodate the change in scope.) If audit resources are not reduced when the scope of the engagement narrows, practitioners may not be conducting audits of the GPFS in an efficient manner. Alternatively, the quality of audits of the entire CAFR may be lower than the quality of audits with a more limited scope.

Generally accepted auditing standards (GAAS) also suggest that initial audits of a specific client will require more resources than subsequent engagements. GAAS requires that certain additional steps be performed during an initial audit. Among these are an investigation of the beginning balances in the balance sheet accounts, communications with the previous auditor, review of the prior year working papers (if any), and review and documentation of the client’s accounting system and internal controls. In addition to these initial audit start-up costs, a learning effect may result in auditor efficiency increasing with auditor tenure. As members of the audit staff become increasingly familiar with a client, they may perform their tasks more efficiently, reducing the overall audit costs. Start-up costs should be concentrated in the initial year of a continuing engagement; the learning effect may result in costs decreasing over a longer period of time. GAAS suggests that the additional resources expended on initial engagements be allocated in a particular manner.

While the presence of initial audit start-up costs may seem intuitively obvious, consistent empirical evidence has not been found by previous researchers using cross-sectional study designs. O’Keefe et al., 1994 T.B. O’Keefe, D.A. Simunic and M.T. Stein, The production of audit services: Evidence from a major public accounting firm, Journal of Accounting Research (1994) (Autumn), pp. 241–261.O’Keefe, Simunic, and Stein (1994) find no systematic pattern of labor hours across auditor tenure in a study of the hours employed in the audits conducted by a Big Six accounting firm during 1989. Conversely, Deis and Giroux (1996) find audit hours to be significantly higher on initial engagements in a cross-sectional study of 232 Texas school districts. Direct evidence of audit start-up costs has been limited because cost information is considered proprietary and is not generally made public by audit firms. A regional audit firm has made available for this study the actual production costs of their governmental audits over a 6-year period. This panel of data allows a longitudinal examination of audit costs across 49 audit clients.

I examine the relationship between audit scope, auditor tenure, and certain other engagement characteristics and the hours of labor charged to audit activities (e.g. planning, internal control evaluation and testing, etc.) by a regional public accounting firm across a sample of governmental audits. I find that more resources are used on expanded scope and initial engagements and that they are generally allocated in the expected manner across audit activities.

Because production cost information is not publicly available, relatively few studies directly examine the determinants of audit hours. Hackenbrack and Knechel (1997) perform a study which examines the determinants of labor hours charged by a single audit firm to particular audit activities across a number of client industries. My research provides similar insight into the production of governmental audits.

Of course, the interpretation of this study depends on the ability to generalize one regional firm’s resource allocation behavior to the population of municipal audits as a whole. Because audit production costs are not publicly available, studies such as this must expand knowledge by examining one audit firm at a time. This research expands the knowledge obtained by similar studies such as O’Keefe et al. (1994) and Hackenbrack and Knechel (1997).

1. Hypotheses development

1.1. Audit scope

The governmental entities included in this analysis are required to prepare comprehensive annual financial reports (CAFR) which include the general purpose financial statements (GPFS) and combining and individual fund type and account group financial statements. These reports are subject to external audit; the reporting entity has a choice of two levels of external audit scope. At a minimum, the auditor’s report must express an opinion on the GPFS and include an opinion as to whether the combining and individual fund and account group financial statements are stated fairly in all material respects in relation to the GPFS taken as a whole. Alternatively, the audit firm can expand its scope by expressing an opinion on the GPFS and on each of the combining and individual fund and account group financial statements.

If the entity requests an audit opinion on the GPFS taken as a whole, “existing audit practice is that audit scope should be set and materiality evaluations should be applied at the fund type and account group level” (AICPA, 1986, p. 37). “If the auditor is engaged to examine the combining and individual fund and account group financial statements in addition to the GPFS, the auditor’s opinion addresses each presentation as a primary statement. Ordinarily, in such circumstances the auditor will need to expand the auditing procedures applied …,” (AICPA, 1986, p. 150) because materiality will be evaluated separately for each individual fund and account group. An increase in audit scope to encompass the separate combining and individual fund and account group financial statements should consequently require an increase in the total amount of time necessary to perform the audit.

A review of the AICPA Audit and Accounting Guide, “Audits of State and Local Governmental Units,” suggests several functional areas of the audit which should be affected by the audit scope. In planning the audit approach, the auditor should develop preliminary estimates of materiality. When the audit scope encompasses the combining and individual fund financial statements, separate materiality levels must be estimated for each individual fund, rather than for each fund type. Audit procedures must also be planned for each fund, whereas some smaller funds may not be considered material when an opinion is expressed on the GPFS taken as a whole. Consequently, the amount of time expended on audit planning is expected to be greater when the scope includes the combining and individual funds and account groups.