European Journal of Social Sciences – Volume 21, Number 3 (2011) Dysfunctional Audit Behaviour: The Effect of Budget Emphasis Leadership Behaviour, and Effectiveness of Audit Review Halil Paino, Azlan Thani Universiti Teknologi MARA Pahang, Malaysia E-mail: [email protected] uitm. edu. my Tel: +60193992676 Fax: +6094602245 Syed Iskandar Zulkarnain Universiti Teknologi MARA Pahang, Malaysia Abstract Dysfunctional audit behaviour is an accepted problem, associated with decreased audit quality. This study develops and tests a theoretical model that identifies factors contributing to dysfunctional audit behaviour.
Budget emphasis, leadership behaviour structure and consideration, and effectiveness of audit review were examined as antecedents of attitudes toward dysfunctional audit behaviour. A path analysis from a partial least square (PLS) approach was employed based on survey results from 225 Audit Managers in Malaysia. The findings produced consistent evidence in support of the theoretical model. It is not an emphasis on meeting budgets that leads to undesirable behaviour, but contextual variables, such as leadership behaviour structure and effectiveness of audit review.
The results of the study should impact auditing procedures, hiring, training and promotion decisions, and help to minimize the occurrence and acceptance of dysfunctional audit behaviour. Keywords: Dysfunctional Audit Behaviour, Budget Emphasis, Leadership Behaviour Structure and Consideration, Audit Review 1. Introduction The panel on Audit Effectiveness, established by AICPA’s Public Oversight Board to examine the issue of audit quality, gathered information from peer reviews and a survey of financial executives, internal auditors, and external auditing professionals.
Their findings indicate that Dysfunctional Audit Behaviour (DAB) is a continuing concern for the auditing profession. This conclusion is significant because dysfunctional audit behaviour can adversely affect the ability of public accounting firms to generate revenue, complete professional quality work on a timely basis and accurately evaluate employee performance. The present study advances Malaysian auditing research by examining the specific factors that contribute to this behaviour including budget emphasis, leadership behaviours and effectiveness of audit review.
In early 2007, the MIA Practice Review Committee issued the first ever Practice Review (PR) report undertaken by MIA, for the period 2003 to 2006, and highlighted some audit quality problems when judged against International Standards on Auditing, Malaysian Standards on Auditing and Companies Act 1965 requirements. These concerns have not been ignored in the academic literature. The underlying premise of much academic research has been that DAB is a dysfunctional reaction to the environment (i. e. , the control system).
This behaviour can, in turn, have both direct and indirect impact on audit quality. Behaviour that directly affects audit quality include premature signing-off of audit steps without completion of the procedure (Otley and Pierce, 1995; Rhode, 1978; Alderman and Deitrick, 1978), 436 European Journal of Social Sciences – Volume 21, Number 3 (2011) gathering of insufficient evidential materials (Alderman and Deitrick, 1982), processing inaccuracy (McDaniel, 1990), and the omission of audit steps (Margheim and Pany, 1986).
Underreporting of audit time has also been shown to have an indirect impact on audit quality (Smith, 1995; Kelly and Margheim, 1990; Lightner, Adam and Lightner, 1982). Underreporting time leads to poor personnel decisions, obscures the need for budget revisions, and results in unrecognised time pressures on future audits (Donnelly, Quirin and Bryan, 2003). Several academic studies have also examined the impact that time pressure has on dysfunctional behaviour (Alderman and Deitrick, 1978, 1982; Margheim and Pany, 1986; Rhode, 1978).
Kelley and Margheim (1990) examined the moderating effects of the interaction between supervisor leadership style and auditor personality. Otley and Pierce (1995) extended this work by examining the moderating effects of audit managers’ leadership style on the behaviour of audit seniors. These studies suggest that an optimal supervisor-subordinate fit can help reduce dysfunctional behaviour and reactions to control systems. Prior literature has identified environmental factors (e. g. time pressure, supervisory style, etc. that contribute to DAB. However, existing literature has not found that individual differences among auditors significantly affect DAB. This study is both timely and significant because of the increased awareness of both the Malaysian Institute of Accountants (MIA), as policy makers, and the Practice Review Committee (PRC) concerning the impairment of audit quality and dysfunctional audit behaviour, following the first PRC Report (publicly available at the MIA website as http://www. mia. org. my/dept/prw/circulars. htm).
The results of this study will help audit firms to understand the harmful impact of these behaviours better and to identify improved means of managing the dysfunctional audit behaviour issue. This study makes a contribution to both the auditing literature and the behavioural literature with respect to the impact of organisational aspects on behaviour. As noted, in relation to the existing studies of audit quality reduction behaviour, it extends earlier studies by simultaneously examining the three different factors that contribute to dysfunctional audit behaviour.
On a practical level, an understanding of the different factors associated with this behaviour should help audit firms in their efforts to dissuade auditors from engaging in dysfunctional audit behaviour. The results of this study could impact auditing procedures, hiring, training and promotion decisions, and help minimize the occurrence and acceptance of dysfunctional audit behaviour. Nevertheless, it should be noted that stronger regulation may not be the only solution to promoting better audit practice and addressing dysfunctional audit behaviour in Malaysia.
Thus, identifying the factors that contribute to auditors’ attitudinal DAB is regarded as an important first step in ascertaining the when, why and how of the incidence of DAB. A theoretical model was developed that relates budget emphasis, leadership behaviour and effectiveness of audit review to auditors’ attitudes toward DABs, namely superficial review of documents, accepted weak client explanations, reduced work below what was considered reasonable, failure to research an accounting principle, and premature signoff of required audit steps. Audit managers were asked to report their attitudes toward DAB.
Using a partial least square analysis, the responses of 225 Audit Managers support the theoretical model and several of the hypothesised relationships. Results of the study suggest that leadership behaviour structure and effectiveness of audit review may have a significant impact on auditor behaviour. These results contribute to our understanding of the correlates of DAB. In addition to the environmental and control variables examined in the previous literature, these findings suggest that individual auditors may differ in their predispositions toward these behaviours.
Further research in this area may assist the profession in identifying personality and individual attributes among auditors that will reduce the incidence of DAB. In addition, improved understanding of DAB can allow firms to target their risk management of DAB to those auditors most likely to have more favourable attitudes toward such behaviours. The remainder of this paper is organised into four sections. The first section presents the theoretical development, while the second section discusses the research method including data 437
European Journal of Social Sciences – Volume 21, Number 3 (2011) collection and measurement. In the third section, empirical results are presented. The final section concludes with a discussion of the findings, together with recognition of limitations, and opportunities for further research. 2. Theoretical Development The theoretical model is presented in Figure 1 and illustrates the structural model of hypotheses for each of the factors studied. Each link in the model is labelled with its respective hypothesis (together with the proposed direction of relatonship) and is discussed subsequently.
The model establishes the proposed links between evaluation of performance (BE), effectiveness of audit review (Arv), consideration in leadership style (LBC) and structure in leadership style (LBS) with dysfunctional audit behaviour (DAB). 2. 1. Direct Association to Dysfunctional Audit Behaviour (DAB) Generally Accepted Auditing Standards (GAAS) and International Standards on Auditing (ISA) both require a firm to maintain a quality control system and require auditors to adequately plan their audits. Several studies have found that a strong audit methodology was associated with higher audit quality.
Malone and Roberts (1996) found that auditors with stronger quality control systems were less likely to participate in dysfunctional audit behaviours. Thus, the independent variables used for this study associated with control systems are time budget emphasis, leadership structure, leadership consideration and the effectiveness of audit review in detecting one type of dysfunctional behaviour, namely premature sign-off (PMSO). Figure 1: Theoretical Model BE H1a + H1 + H1b – LBC H2 – DAB H4a + LBS H4 b H4 – H3 + ARv
Budget related time pressure can occur when the budgeted amount of time is less than total available time and the auditor can respond to the pressure by completing the work in their own personal time and underreporting the amount of time spent on the audit task. Prior research examining auditor time budget pressure has generally been performed using surveys or case studies (e. g. , Rhode, 1978; Kelley and Seiler, 1982; Lightner, Adams and Lightner, 1982; Alderman and Deitrick, 1982; Cook and Kelley, 1991; Raghunathan, 1991). In general, these findings indicate that audit time budget pressure has been increasing over time.
In addition, these researches suggest that audit effectiveness may be negatively affected as a result of auditors engaging in dysfunctional behaviour (such as premature sign off and underreporting of chargeable time) in order to meet their time budgets. Prior research examining auditor time deadline pressure has generally consisted of experiments where auditors were given a task to perform and the amount of available time to complete the task was 438 European Journal of Social Sciences – Volume 21, Number 3 (2011) varied across subjects.
For example, McDaniel (1990) found that as the time deadline pressure utilised in her study increased, auditors’ job effectiveness decreased and their efficiency increased. In contrast, research by Choo (1995) suggests that as time deadline pressure increases from low to moderate levels, auditor judgment performance improves due to the reduction in the usage of non-diagnostic (i. e. irrelevant) cues. However, Choo’s findings also indicate that as deadline pressure increases to higher levels, performance declines because relevant cues are also ignored.
In general, audit time budget pressure occurs when an audit firm allocates (i. e. budgets) a scarce number of audit hours to be used by auditors to complete specified audit procedures. The emphasis which the audit firm places on meeting time budgets can influence individual auditor behaviour (Alderman & Deitrick, 1982; Otley & Pierce, 1996). Time budget achievement is seen by auditors as being critical for performance evaluation (Kelley & Seiler, 1982; Otley & Pierce, 1996), and pressure to meet time budgets is one of the main causes of staff turnover.
Accordingly, the following hypothesis is tested: H1: High perceived emphasis on meeting time budgets in the evaluation of performance (BE) will be associated with an increase in dysfunctional behaviour. Large audit firms operate using a clear hierarchical structure where the audit senior reports directly to the audit manager and the audit manager reports directly to a partner. The formal evaluation of the senior’s and manager’s performance is carried out by the partner. In this situation, the behaviour of the partner, who is cast in a leadership role, is expected to influence the senior’s and manager’s behaviour.
Studies in this area (e. g. , Kelley & Margheim, 1990) have measured leadership style by using two dimensions: consideration and structure. Stogdill (1963) defined ‘consideration’ as “the extent to which an individual is likely to have job relationships characterised by specific action such as mutual trust, respect for subordinate’s ideas, and consideration of their feelings”, whereas ‘structure’ reflects “the extent to which an individual is likely to define his own role and those of his subordinates towards goal attainment”.
Control theory would suggest that, in an auditing environment, rigid application of controls is likely to lead to defensive attitudes and dysfunctional behaviours. Therefore, the following hypotheses were tested: H2: High levels of consideration in the leadership style of partners (LBC) will be associated with low levels of dysfunctional behaviour. H3: High levels of structure in the leadership style of partners (LBS) will be associated with high levels of dysfunctional behaviour. Otley and Pierce (1996) reported that the firm’s review procedures should be adequate to detect Premature Sign-Off (PMSO).
There is a suggestion in previous studies that a manager’s perception of the risk of being caught may be relevant in the decision to sign off prematurely or engage in other forms of dysfunctional behaviour. The Commission on Auditors’ Responsibilities (CAR) (1978) noted that the greatest concern of auditors who engage in PMSO is the risk of discovery by their superior. The results of previous studies support a negative association between the perceived effectiveness of the audit review in detecting PMSO and dysfunctional behaviour.
Accordingly, the following hypothesis was tested: H4: High perceived effectiveness of the audit review in detecting PMSO (ARv) will be associated with low levels of dysfunctional behaviour. 2. 2. Indirect Association to Dysfunctional Audit Behaviour (DAB) The interrelationships between budget emphasis (BE), effectiveness of audit review in detecting PMSO (ARv), leadership structure (LBS) and leadership consideration (LBC) can provide a further understanding of the complex causes of audit firm factors to dysfunctional audit behaviour.
As noted by Fleet and Griffin (2006), leaders most likely play a major role in eliciting dysfunctional behaviour by its members and subordinates. In this way, the leader’s values (consideration or structure) are “taught” to others and shape their behaviour in the organisation. Therefore, the following hypotheses were examined: 439
European Journal of Social Sciences – Volume 21, Number 3 (2011) H1a: There is a positive association between Budget Emphasis (BE) and Leadership behaviour Consideration (LBC) H1b: There is a negative association between Budget Emphasis (BE) and Leadership Behaviour Structure (LBS) H4a: There is a positive association between effectiveness of Audit Review (Arv) and Leadership Behaviour Consideration (LBC) H4b: There is a negative association between effectiveness of Audit Review (Arv) and Leadership Behaviour Structure (LBS) In order to incorporate the indirect effect of the leadership behaviour consideration (LBC) and leadership behaviour structure (LBS) on the relationship between budget emphasis (BE), effectiveness of audit review (ARv) and dysfunctional audit behaviour (DAB), the following hypotheses were tested: H1c: Budget Emphasis (BE) has an indirect effect on Dysfunctional Audit Behaviour (DAB) through the Leadership behaviour consideration (LBC) and Leadership behaviour structure (LBS).
H4c: Effectiveness of audit review (ARv) has an indirect effect on Dysfunctional Audit Behaviour (DAB) through the Leadership behaviour consideration (LBC) and Leadership behaviour structure (LBS). 3. Research Method 3. 1. Data Collection Data was collected using a survey questionnaire sent to all Audit Managers registered with Malaysian Institute of Accountants (MIA), a total of 621 auditors. Questionnaires were sent out to firms of varying size including Big 4, small firms and medium firms. Of the 621 surveys distributed, respondents returned a total of 225 usable surveys for an effective response rate of 36 percent. The average respondent was in the 35-39 year age group and had between 10 and 14 years of audit experience. Female respondents represented approximately 72 percent of the returned instruments. 3. 2.
Measures The variables measured in the questionnaire include budget emphasis, leadership behaviour structure, leadership behaviour consideration, effectiveness of audit review in detecting premature sign-off and attitude toward dysfunctional audit behaviour. Budget emphasis was measured by a direct question about the perceived and desired importance of budget achievement in the overall evaluation of performance. Leadership behaviour structure and leadership behaviour consideration were measured using the instruments adapted for an audit setting by Otley and Pierce (1996) and Pratt and Jiambalvo (1981), and based on Stogdill’s (1963) Leader Behaviour Descriptive Questionnaire (LBDQ). This LBDQ has been used extensively to measure leadership behaviour structure and consideration.
The instrument’s reliability and validity have been deemed acceptable in prior research (Otley and Pierce, 1996). Effectiveness of audit review refers to its effectiveness in detecting premature sign-off. Subjects were asked a direct question on the effectiveness of audit review in detecting PMSO. Dysfunctional audit behaviour (DAB) was measured using five specific behaviours similar to those identified by Otley and Pierce (1996) and Kelley and Margheim (1987). Subjects were asked to indicate the frequency with which these behaviours were encountered in the previous years of audit work. 3. 3. Partial Least Square (PLS) analysis Path analysis using PLS was used to evaluate the proposed hypotheses.
Path analysis, rather than moderated regression analysis (MRA) or ANOVA, was used because the theoretical model presented in this study is viewed as an antecedent framework for DAB. PLS is a constrained form of component modelling, whereas conventional structural equation modelling (SEM) analysis, such as with LISREL, 440 European Journal of Social Sciences – Volume 21, Number 3 (2011) can be seen as modelling with common factors. Conducting PLS analysis involves a two-step procedure: the first step is to evaluate a measurement model for each latent construct, which in practice assesses the validity and reliability of the measures; the second step is to conduct a path analysis.
In PLS analysis, Chin, Barbara and Peter (2003) advise that the adequacy of the measures is assessed by evaluating three components; (1) the reliability of the individual items, (2) the internal consistency of the items measuring the same latent construct, and (3) the discriminant validity of the construct. The reliability of the individual items was assessed by examining the loading of the items on their corresponding construct, where items’ loadings greater than 0. 4 can be considered acceptable. Cronbach Alpha is the most common method used to assess measurement reliability. The measure for internal consistency is assessed by Composite reliability (CR) with a desired value greater than 0. 7. The last indicator is the average variance extracted (AVE), which refers to how much the items explain the variance of the construct.
The desired value for AVE is greater than 0. 5. The Cronbach Alpha for DAB is 0. 72, and the AVE is 0. 64 with individual item loadings ranging from 0. 40 to 0. 85. The Cronbach Alpha for leadership behaviour consideration is 0. 91, and the AVE is 0. 79. The individual item loadings ranged from 0. 60 to 0. 85. The Cronbach Alpha for leadership behaviour structure is 0. 91, and the AVE is 0. 84 with the individual item loadings ranging from 0. 42 to 0. 91. As effectiveness of audit review and budget emphasis each comprised a single item, their loadings were 1. 00. PLS path analysis uses similar indicators to regression analysis to interpret itsresults.
R-square (R? ) can have values 0-1. Higher values mean that the model explains more variance. The size of path coefficients, beta coefficients, refers to the strength of the relationship between independent and dependent variable. The significance of the path, t-values, indicates if a particular path is statistically significant. 4. Empirical Results Table 1 presents the results of the path analysis and lists each hypothesis and its corresponding path coefficient. Figurative representations of the results are also displayed in Figure 2. R? reported for the dependent variable DAB is 0. 56, 0. 31 for LBC and 0. 34 for LBS. Figure 2: Path Analytic Model * p