Underreporting time is a common ethical dilemma in auditing. We examine the effects of professional commitment and experience on auditor acceptance of underreporting time with a sample of 110 practicing auditors at two large national accounting firms. Using a moderated mediation research model, we find that auditor experience moderates the negative relationship between professional commitment and acceptance of underreporting time, such that professional commitment is associated with underreporting acceptance only among less experienced auditors. We discuss the contributions, limitations, and practical implications of our findings.
Data Availability: Please contact the first author.
Underreporting time is a common ethical dilemma among auditors and has several detrimental consequences for public accounting firms (Pickerd, Summers, and Wood 2015). Underreporting time can impact audit fee negotiations with clients, distort time budgets for subsequent years' audits, and cloud assessments of audit effectiveness (Akers, Horngren, and Eaton 1998). Although audit firm policies prohibit auditors from misreporting hours (Sweeney and Pierce 2006), auditors often have incentives to underreport time including the potential for more favorable performance evaluations (Anderson-Gough, Grey, and Robson 2001). Indeed, underreporting time is ubiquitous in public accounting firms (Church 2014; Pickerd et al. 2015). Given the negative effects of underreporting on audit firms and the profession, it is important to consider factors that may influence auditors' acceptance of underreporting time.
Prior research reports mixed results on the seemingly logical effects of auditors' professional commitment and experience on ethical evaluations and decision making, including underreporting time. For example, although Otley and Pierce (1996) find no significant relationship between auditors' professional commitment and underreporting time, Elias (2006) finds that auditing students with higher professional commitment are more likely to perceive underreporting time as unethical.
Our study aims to specify when and how these variables may impact auditor acceptance of underreporting time. We propose and test a moderated mediation model whereby auditor experience predicts professional commitment (mediator) and acceptance of underreporting time (dependent variable), and also moderates the negative relationship between professional commitment and underreporting acceptance. Using a sample of 110 auditors at two large national firms, we find evidence that supports our research model. Specifically, we find that auditor experience moderates the negative relationship between professional commitment and acceptance of underreporting time, such that professional commitment is negatively associated with underreporting acceptance—but only among less experienced auditors. This study contributes to the literature on professional commitment and dysfunctional audit behavior by being the first to document this interaction. This is also the first study to find a significant relationship between auditors' professional commitment and acceptance of underreporting time.
The results of our research suggest that audit firms and business colleges should consider ways in which professional commitment can be enhanced among entry-level auditors. Although accounting firms are responsible for maintaining an organizational culture that promotes professional commitment among their auditors, the socialization process necessary to engender high levels of professional commitment among entry-level auditors must begin years before the auditor joins the firm. College accounting programs and instructors play an instrumental role in students' socialization into the profession (Elias 2006, 2008). Instructors should emphasize to students that accounting is a system designed to meet the needs of financial statement users and encourage them to think of ways in which our system could be improved. Strategies such as this likely promote a sense of professionalism among students. Students should also be encouraged or required to complete internships and attend events sponsored by student (and other) organizations where practicing professionals discuss their organizations. Networking with practicing professionals and instructors can facilitate students' socialization into the auditing profession leading to higher levels of professional commitment upon entry into the profession.
The remainder of this paper is structured as follows. The next section develops our hypothesis by discussing relevant research on underreporting time, professional commitment, and auditor experience. The design of the survey that we conducted is presented in the third section, followed by a results discussion. Finally, we conclude with a review of the findings and discussion of the study's contributions, limitations, and practical implications.
BACKGROUND AND HYPOTHESIS DEVELOPMENT
Auditors' truthfully reporting the actual hours they work on an engagement is critical to public accounting firms (Rhode 1978; Lightner, Leisenring, and Winters 1983; Akers et al. 1998; Pickerd et al. 2015), as underreporting time can threaten audit quality (Donnelly, Quirin, and O' Bryan 2003; Stefaniak and Robertson 2010). Akers et al. (1998) contend that underreporting time can be harmful to firms and auditors for the following reasons: (1) firms use reported time in preparing the time budget for next year's audit thereby forcing future auditors to have their performance measured against an unrealistic budget or to underreport themselves, (2) firms use reported time to negotiate audit fees with clients, (3) firms use reported time to assess the effectiveness of their audit approach on current engagements, (4) firms use reported hours to make resource allocation decisions, and (5) firms use reported hours to bill clients extra in certain situations. These factors have the potential to damage not only the audit firm, but the auditing profession as a whole. For these reasons, audit firm policies specifically prohibit auditors from misreporting hours (Buchheit, Pasewark, and Strawser 2003; Sweeney and Pierce 2006; Smith and Hutton 2011).1
Nevertheless, auditors often have incentives to underreport time including the possibility of more favorable performance reviews (Rhode 1978; Lightner et al. 1983; Anderson-Gough et al. 2001). Indeed, underreporting time is prevalent in public accounting (Otley and Pierce 1996; Shapeero and Killough 1999; Akers and Eaton 2003; Shapeero, Koh, and Killough 2003). While audit firms may have reduced emphasis on formal, explicit incentives to underreport time, recent empirical evidence suggests that auditors still feel at least implicit pressure to underreport (Agoglia, Hatfield, and Lambert 2015). In a recent interview study with auditors at Big 4 and regional firms, Church (2014, A31) documents that auditors feel pressure to work without charging the hours to the budget or “eat time.” Consequently, it is worthwhile to consider factors that may impact auditors' attitudes toward the underreporting of time. The next discussion follows Figure 1.
A factor that may influence auditors' acceptance of underreporting time is the level of their commitment to the auditing profession. Professional commitment has been considered in terms of dedication to a professional career, identification with the profession, and acceptance of the profession's goals and ethics (J. Sorensen and T. Sorensen 1974). In the accounting literature, it has been described with reference to a belief in and acceptance of the goals and values of the profession, a willingness to exert considerable effort on behalf of the profession, and a desire to maintain membership in the profession (Aranya, Pollock, and Amernic 1981; Aranya and Ferris 1984).
Professional commitment is presumably developed during the socialization process that accompanies entry into the profession (Aranya, Lachman, and Amernic 1982). Prior research finds auditor experience level to be positively associated with professional commitment (Aranya and Ferris 1984; Jeffrey and Weatherholt 1996; Smith and Hall 2008; Suddaby, Gendron, and Lam 2009). The increased socialization experiences that go along with advancing through career stages in public accounting may contribute to higher levels of professional commitment (Hall, Smith, and Langfield-Smith 2005). Accordingly, we expect auditor experience to be positively associated with professional commitment.
Aranya et al. (1981) suggest that higher professional commitment should be reflected in greater sensitivity to issues concerning professional ethics. Similarly, Lord and DeZoort (2001, 220) reason that “high professional commitment should orient auditors toward behavior that is in the public interest and away from behavior that has the potential to damage the profession. Alternatively, auditors with lower professional commitment should be more inclined to behave dysfunctionally.” However, empirical studies on the relationship between auditors' professional commitment and ethical evaluations and decisions have yielded mixed results (Elias 2006). To illustrate, Shaub, Finn, and Munter (1993) find no relationship between professional commitment and auditors' ability to identify ethical issues. However, Jeffrey and Weatherholt (1996) and Jeffrey, Weatherholt, and Lo (1996) find a positive relationship between professional commitment and auditors' rule observance attitudes. Taylor and Curtis (2010) find a positive relationship between professional commitment and auditors' intention to report a questionable act, but no relationship between professional commitment and the perseverance of reporting intention.
We are aware of one prior study that has empirically investigated the link between auditors' professional commitment and underreporting time. Contrary to their prediction, Otley and Pierce (1996) did not find a significant relationship between professional commitment and underreporting of time. Their sample consisted of audit seniors in Ireland. However, there is some limited empirical support for the notion that auditors' professional commitment will be negatively related to underreporting time. Using a sample of students from undergraduate auditing classes, Elias (2006) finds that students with higher professional commitment are more likely to perceive underreporting of time as an unethical behavior. Since professional commitment should result in greater sensitivity to issues concerning professional ethics (Aranya et al. 1981; Lord and DeZoort 2001), auditors with high professional commitment may be less accepting of underreporting time. Consequently, we expect professional commitment to be negatively associated with acceptance of underreporting time.
Mixed findings exist on the relationship between auditor experience and ethical decision making (Sweeney, Arnold, and Pierce 2010, 536). Some empirical research on underreporting of time uses pre-manager level auditors, as auditors at the staff and senior levels have been found to engage in higher levels of underreporting behavior (Otley and Pierce 1996; Pierce and Sweeney 2003; Sweeney and Pierce 2006; Sweeney et al. 2010). Nevertheless, other studies on underreporting time use samples consisting of auditors at various hierarchical ranks, from the staff through partner levels (e.g., S. Lightner, Adams, and K. Lightner 1982; Margheim and Pany 1986; Donnelly et al. 2003; Donnelly, Quirin, and O'Bryan 2011; Paino, Smith, and Ismail 2012; Agoglia et al. 2015; Barrainkua and Espinosa-Pike 2015). Lightner et al. (1982) surveyed auditors at the staff through partner levels and did not find a significant relationship between auditor level and underreporting time. However, Donnelly et al. (2011) find experience level to be negatively associated with dysfunctional audit behavior (including underreporting time) and Sweeney et al. (2010) find that auditor experience is negatively related to underreporting time. Accordingly, we expect experience to be negatively associated with underreporting acceptance.
Prior research suggests that as auditors gain experience over the years they are more likely to recognize the adverse consequences of dysfunctional audit behavior (Alderman and Deitrick 1982; Kelley and Seiler 1982; Cook and Kelley 1988; Raghunathan 1991; Shapeero et al. 2003; Donnelly et al. 2011). Lower-level auditors are usually exposed to a limited number of audit areas such as cash and accounts payable. They are likely concerned more with completing their assigned audit sections and getting them through the review process within the allotted time budget and concerned less with the overall audit process. In contrast, partners and managers must adopt a more holistic view of the audit process. They are typically involved in all phases of the audit from planning to reporting, and are ultimately responsible for budgeting, resource allocation, audit execution, billing, and client retention. Accordingly, more experienced auditors better appreciate the harmful effects of underreporting time relative to less experienced auditors, regardless of their level of commitment to the profession. Therefore, auditor experience likely buffers the impact of professional commitment on acceptance of underreporting time. In other words, we expect to find a significant relationship between professional commitment and underreporting acceptance—but only among less experienced auditors. This leads to our hypothesis:
The interaction effect between auditor experience and professional commitment will predict acceptance of underreporting time, such that increased experience will weaken the negative association between professional commitment and acceptance of underreporting time.
Auditors in the American Midwest practice offices of two large national public accounting firms participated in this research.2 Recruitment emails that included a link to an Internet-based survey site hosting the survey were sent by audit partners at our request in the summer of 2014.3 The anonymity of the respondents was protected with the intent of making them less likely to respond in a socially desirable manner (P. Podsakoff, MacKenzie, Lee, and N. Podsakoff 2003).4 Two hundred sixty-three auditors received the email and 110 auditors completed the survey, resulting in a response rate of 41.8 percent.5 The sample is comprised of 40 staff auditors, 27 seniors, 19 managers, 10 senior managers, and 14 partners.
Scales used for the variables are presented in Appendix A. We use the four-item commitment scale developed by Klein, Cooper, Molloy, and Swanson (2014) to measure professional commitment.6 An example item is, “How dedicated are you to the profession?” (1 = not at all, 7 = completely). We use four items taken from Donnelly et al.'s (2003) dysfunctional audit behavior scale to measure acceptance of underreporting time. An example item is, “I am more accepting of auditors underreporting their time if it improves their performance evaluation” (1 = strongly disagree, 7 = strongly agree). Consistent with Donnelly et al. (2003, 2011) and Paino et al. (2012), respondents were asked to indicate their acceptance of, rather than actual engagement in, underreporting time. This approach elicits greater honesty in participants' responses and reduces concerns of social desirability (Randall and Fernandes 1992; Jones, Massey, and Thorne 2003). Auditor experience is measured in years as an auditor.
Table 1 reports descriptive statistics, reliabilities, and correlations for the independent and dependent variables. The reported scale reliabilities (Cronbach's alpha) indicate high internal consistency (Gliner and Morgan 2000). As expected, professional commitment is negatively correlated with acceptance of underreporting time. Also in line with expectations, auditor experience is positively correlated with professional commitment and negatively correlated with acceptance of underreporting time.7
For additional descriptive information on auditor experience level and acceptance of underreporting time, we separated staff and seniors (Staff Group) from managers, senior managers, and partners (Management Group). Table 2 reports descriptive statistics for acceptance of underreporting of time by auditor group. Means for both groups are relatively low, suggesting that participating auditors are generally unaccepting of underreporting time. However, a t-test reveals that the Staff Group mean is significantly higher than the Management Group mean (p < 0.001), indicating that less experienced auditors are more accepting of underreporting time than more experienced auditors.
Panel A of Table 3 reports the regression analysis for tests of the main effects of professional commitment and auditor experience on underreporting acceptance. The coefficient for professional commitment (PC) (−0.132; p = 0.06) provides some evidence that professional commitment is negatively related to acceptance of underreporting time, and the coefficient for auditor experience (EXP) (−0.041; p < 0.001) provides evidence that experience is negatively associated with underreporting acceptance.
For hypothesis testing, we use the PROCESS macro for SPSS (Hayes 2013). Conditional direct effects are evaluated using Model 1 of the PROCESS macro (Hayes 2013) with 10,000 bootstrap samples. H1 posits that the interaction effect between auditor experience and professional commitment will predict acceptance of underreporting time, such that increased experience will weaken the negative association between professional commitment and acceptance of underreporting time. The coefficient for the interaction term in Panel B of Table 3 (0.023; p < 0.01) supports H1. The main effect of professional commitment in Panel B (−0.010; p = 0.09) has weakened, indicating that the interaction term has moderated the relationship between professional commitment and acceptance of underreporting time.
Figure 2 provides a visualization of the conditional direct effects.8 As Figure 2 depicts, high-experience auditors have similar levels of acceptance of underreporting time regardless of their professional commitment levels. This is also the case for average experience auditors. However, low-experience auditors with high professional commitment are significantly less accepting of underreporting time than low-experience auditors with low professional commitment.
Our research model (Figure 1) represents a moderated mediation model (cf. Hayes 2015, 5). We use Model 74 of the PROCESS macro (Hayes 2013) with 10,000 bootstrap samples to test for a mediated relationship along with moderation of the second mediated pathway by the predictor (i.e., to test for conditional indirect effects). Auditor experience is the predictor variable, professional commitment is the mediator, and underreporting acceptance is the dependent variable. Experience is also a moderator between professional commitment and underreporting acceptance. The analysis provides 95 percent confidence interval (CI) values, which indicate that an effect is significant at the 0.05 level if its CI values do not include zero. The results of the analysis reveal that the conditional indirect effect of experience on underreporting acceptance through professional commitment is significant among low-experience auditors (B = −0.014; CI [−0.026, −0.003]), but not significant for average-experience auditors (B = −0.005; CI [−0.014, 0.002]) or high-experience auditors (B = 0.005; CI [−0.004, 0.014]). This analysis supports our research model.
Prior research reports mixed results on the seemingly logical effects of auditors' professional commitment and experience level on ethical evaluations and decision making. The tenuous relationship between professional commitment and ethical decision making reported in prior research could cause some to question the significance of commitment to one's profession in influencing important attitudes and behavior. In an attempt to better specify when and how professional commitment may impact ethical evaluations, we propose and test a moderated mediation model whereby experience predicts professional commitment and acceptance of underreporting time, and also moderates the negative relationship between professional commitment and underreporting acceptance.
Using a sample of 110 practicing auditors at two large national firms, we find evidence that supports our research model. Specifically, we find that auditor experience moderates the negative relationship between professional commitment and acceptance of underreporting time, such that professional commitment is associated with underreporting acceptance—but only among less experienced auditors. This study adds to the literature on professional commitment and dysfunctional audit behavior by being the first to document this interaction. It is also the first study to find a significant relationship between auditors' professional commitment and acceptance of underreporting time. Our study clarifies that professional commitment is indeed an important construct in influencing attitudes among less experienced auditors.
Limitations and Future Research
Our study has limitations associated with survey-based research including the potential for nonresponse and social desirability biases. It is based on the premise that auditors' acceptance of underreporting time serves as an indicator of their actual engagement in underreporting, which may not necessarily be the case. However, this approach is designed to mitigate social desirability bias and is consistent with the literature (Donnelly et al. 2003). Moreover, the theory of planned behavior (Ajzen 1988, 1991) argues that one's attitude toward a behavior contributes to the formation of a behavioral intention, which in turn leads to the actual behavior. Nevertheless, this remains a limitation of our study. The use of a cross-sectional design inhibits us from drawing strong causal inferences.
Future research on underreporting could advance the literature by overcoming some of these limitations. An experimental approach where professional commitment is manipulated and underreporting is measured by asking participants to report the time they spend on a task (with an incentive to underreport) could help eliminate or reduce concerns of social desirability and nonresponse, assess actual behavior rather than attitude toward the behavior, and improve the ability to draw causal inferences.
Our study has practical implications for audit firms and business colleges. Audit firms should effectually train staff to understand the adverse consequences of underreporting time. Such training is consistent with the quality control standards of leadership establishing the appropriate tone at the top. In light of our finding that professional commitment among relatively inexperienced auditors is associated with less acceptance of underreporting time, we now consider ways in which professional commitment can be enhanced among entry-level auditors. Our focus on entry-level auditors is fitting because we find that the impact of professional commitment on underreporting acceptance is only significant among this group of less experienced auditors.
Continuing education, professional accounting qualifications, organizational culture, and professional membership requirements and services are likely to affect the development of professional commitment among auditors (Smith and Hall 2008). Accounting firms are responsible for maintaining a culture within their firm that promotes professional commitment among auditors. Professional commitment is presumably developed during the socialization process that accompanies entry into the profession (Aranya et al. 1982). Firms can help foster professional commitment among entry-level auditors during their internships by communicating the importance of reporting time correctly. Firms should also encourage entry-level auditors to take notice of the positive aspects of their job and consider how they fit into their long-term plans, as this can enhance their level of job commitment (Church 2014). More experienced auditors should take the time to get to know new hires and fully explain their role in the audit as this can facilitate their assimilation and socialization into the profession (Church 2014). Audit firms should increase awareness among more experienced auditors on the importance of their roles in entry-level auditors' successful socialization into the profession, emphasizing the significance of personal interactions (Church 2014). These interactions may be both informal and part of the audit firm's formal mentoring programs.
However, successfully developing high levels of professional commitment among entry-level auditors prior to their serving on audit engagements (often only two weeks after joining the firm) may be an unrealistically tall order for audit firms. The indoctrination of professional commitment must be a collaborative effort between practice and academia. Business colleges should engage with the practice community in an effort to heighten professional commitment among accounting students. We now consider how college accounting programs can assist in the development of professional commitment among entry-level auditors while they are still students.
College accounting programs and instructors play an influential role in students' socialization into the profession (Elias 2006, 2008). Many accounting programs encourage or require students to complete an internship as part of their education. Internships with audit firms immerse students in the auditing profession prior to their graduation, stimulating commitment to the profession while they are still in school. Instructors should encourage or require students to attend events sponsored by accounting student clubs and other organizations where professionals from both public accounting and industry discuss their firms. Student participation in professional development activities both within and outside the university, such as audit competitions, conferences, and state CPA society events, are also worthwhile. Networking with practicing professionals and instructors will likely facilitate accounting students' socialization into the auditing profession and lead to higher levels of professional commitment.
College accounting programs may consider integrating professionalism throughout the accounting curriculum, especially emphasizing the expectations of the profession in a capstone or ethics course for upper-level students. Instructors should encourage students to begin to think like accounting professionals instead of bookkeepers early in the program, such as in intermediate accounting courses. For example, it is important to emphasize to students why we account for events and transactions the way we do, instead of simply how to account for things. Tactics such as this can prompt students to view accounting as a system designed to meet the needs of financial statement users and encourage them to think of ways in which our system could be improved. This is thinking like a professional.
We requested official firm policy on underreporting time from both firms participating in this study. We obtained and reviewed official firm documentation that specifically prohibits underreporting chargeable time.
The participating firms are among the ten largest in the U.S. (SourceMedia 2014) but are not Big 4 firms.
The recruitment email read as follows: “We are assisting with academic research being conducted independently by [author], an accounting professor at [university]. [Author] is a former audit manager and is studying the relationships auditors have with various work-related entities. To assist [him/her] with this research, we are asking that you please complete a short Internet-based survey. Participation is voluntary and completely anonymous. You may quit the survey at any time and your individual survey responses will not be shared with the firm as they are for academic research purposes only.”
The predictor and criterion variables were temporally and proximally separated in our instrument to reduce the perceived relevance of the previously recalled information in short-term memory, thereby mitigating biases by making prior responses less salient, available, or relevant, and diminishing the participant's ability and motivation to use prior responses to answer subsequent questions (Podsakoff et al. 2003, 888). The predictor variables were assessed early in our instrument. After answering demographic questions and questions not relevant to the purpose of the present research, the criterion variable was measured. This temporal and proximal separation reduces priming effect concerns that earlier responses to professional commitment questions influenced participants' later responses to questions on underreporting time.
Second-request emails were sent by the firms after approximately one week. We assessed whether nonrespondents could have produced any significant biases using t-tests comparing early with late respondents on the variables. The reasoning behind this test is that late respondents may share characteristics with nonrespondents. These comparisons revealed no significant differences, mitigating concerns associated with nonresponse bias (Armstrong and Overton 1977). The response rate is similar to those reported in prior accounting survey studies (e.g., Raghunathan 1991; Herrbach 2001; Herda and Lavelle 2012, 2013).
Klein, Molloy, and Brinsfield (2012) reconceptualized the construct of commitment and outlined several advantages of their reconceptualization such as improved construct differentiation and greater applicability across the full array of workplace targets. Consistent with Klein et al.'s (2012) conceptualization, Klein et al. (2014) developed a commitment measure that demonstrates less overlap with other measures, such as job satisfaction and identification, than prior scales. They empirically validated this alternative commitment measure using five samples yielding 2,487 participants. Their analyses included comparisons of the new measure to prior traditional commitment measures. Their measure was most notably correlated with the affective commitment scale from the three-component model (r = 0.69; p < 0.01).
Variance inflation factors (VIFs) were calculated to assess multicollinearity of the variables. Results indicated no concerns about multicollinearity as VIFs were all below 2.
Model 1 of the PROCESS macro provides the information necessary to visualize the moderator effects. When depicting the interaction, the low values of the variables are calculated as one SD below the mean, and the high values are calculated as one SD above the mean. The auditor experience and professional commitment variables were mean centered prior to the analysis.
Please answer the following questions as they relate to the auditing profession (seven-point scale [“not at all” to “completely”]).
How committed are you to the auditing profession?
To what extent do you care about the profession?
How dedicated are you to the profession?
To what extent have you chosen to be committed to the profession?
Acceptance of Underreporting Time
Please indicate the extent to which you agree or disagree with the following statements. I am more accepting of auditors underreporting their time if (seven-point scale [“strongly disagree” to “strongly agree”]):
It improves their chances for promotion and advancement.
It improves their performance evaluations.
It is suggested by their supervisor.
Others underreport their time and it is necessary to compete with them.
We thank J. Gregory Jenkins (editor) and two anonymous reviewers for their helpful comments and suggestions. We also thank the two public accounting firms that participated in this research.
Editor's note: Accepted by J. Gregory Jenkins.