The Maltese statutory auditing framework encompasses all Maltese registered companies because Article 19(4)(a) of the Income Tax Management Act (2014, Chap.372) continues to impose the obligation of conducting a full scope statutory audit on all companies registered in Malta, irrespective of their size or ownership structure. An external audit of financial statements provides reasonable assurance as to whether the audited financial statements as a whole are prepared, in all material respects, in accordance with an identifiable financial reporting framework. Thus, the auditor is only responsible to detect misstatements that are material to the financial statements as a whole (ISA 200, IAASB, 2009a).

Materiality is defined indirectly in ISA 320 ‘Materiality in Planning and Performing an Audit’ (IAASB, 2009c) as follows:

“Misstatements, including omissions, are considered to be material if they, individually or in the aggregate, could reasonably be expected to influence the economic decisions of users taken on the basis of the financial statements.” – Para. 2

Although at first glance this definition appears to be a relatively simple one, applying it in practice is often fraught with subjectivity and complexity. The auditor has to distinguish between omissions and misstatements that would affect the users of financial statements and those that would not affect such users (Vorhies, 2005). In addition, the fact that general purpose financial statements are prepared for a range of primary users comprising “existing and potential investors, lenders and other creditors” (IFRS Foundation, 2010) adds to the complexity of such assessment since materiality is likely to be unique to each user (Doxey, 2013). However, ISA 320 (IAASB, 2009c, para.10) simplifies the situation by stipulating that when determining materiality, the auditor considers the “common financial information needs of users as a group” and thus ignores the needs of “specific individual users”.

Materiality in the context of an audit takes the form of a numeric threshold which the auditor initially determines by applying a certain percentage to a chosen benchmark (ISA 320, IAASB, 2009c, para.A3; Porter et al., 2003). Materiality is inherently subjective and susceptible to misinterpretation because its seemingly clear-cut numerical form must be interpreted in light of surrounding qualitative circumstances by applying professional judgement (Azzopardi and Baldacchino, 2009; ISA 320, IAASB, 2009c, para.A3). In addition, materiality affects the nature, timing and extent of audit procedures and it is a significant determinant of audit effort and cost (Joldoş et al., 2010). Thus, materiality has to be considered throughout the entire audit process (ISA 320, IAASB, 2009c).

The question as to whether materiality is relevant for users of financial statements was raised as early as 1977 (Fisher, 1990; Leslie, 1977 cited in Davis, 2005); however, there is still no requirement in ISAs to disclose materiality information in the auditor’s report (IAASB, 2014), although such information may be communicated to those charged with governance (ISA 260, IAASB, 2009b, para.A13).

Materiality disclosures are presently not mandatory in Maltese statutory auditing. Moreover, according to a study by Azzopardi and Baldacchino (2009), Maltese audit practitioners were in 2009 largely against the idea of disclosing materiality information in the audit report due to the perceived risks of doing so.

Several foreign studies have shown that materiality disclosures in the audit report could have beneficial effects, while other studies have raised concerns of potential drawbacks. Research from a users’ perspective seems to conclude that materiality should be disclosed, whilst research from the auditors’ perspective is still in its primitive stages, although it seems that auditors are rather apprehensive about disclosing materiality. In fact, Doxey (2013) and the Chartered Financial Analyst (CFA) Institute (2013) observed that users of the audit report seem to be in favour of materiality disclosures, whilst the auditors seem to be against such disclosures. This lack of consensus with regards to materiality disclosures is part of a much larger audit reporting debate which has been going on for many decades.

Key Definitions

The following terms used in this study shall be defined as follows:

Materiality disclosures are the act of presenting materiality thresholds and accompanying qualitative information.

Non-audit accountants are accountants registered with the Malta Accountancy Board (MAB) who had never worked as external auditors at the time they responded to the questionnaire used in this study.

Past auditors are members registered with the MAB who had stopped practicing as auditors prior to the time they responded to the questionnaire.

Present auditors are auditors registered with the MAB who were practicing as auditors at the time they responded to the questionnaire.

Respondents refer collectively to present and past auditors who responded to the questionnaire.

The Regulation refers to Article 11(2)(h) of EU Regulation No. 537/2014, unless otherwise indicated.

Users refers to the readers of audited financial statements and especially to the users of general purpose financial reporting, being “existing and potential investors, lenders and other creditors” as defined by the International Financial Reporting Standards (IFRS) Foundation’s Conceptual Framework (2010, p.9).

Materiality is one of the fundamental auditing concepts throughout the entire audit process, from the initial planning stage, the collection of audit evidence and eventually, the issuance of the audit opinion. At the planning stage, the auditor is required to set a numeric materiality threshold for the financial statements as a whole (ISA 320, IAASB, 2009c), commonly referred to as “overall or planning materiality” (Eilifsen and Messier Jr, 2014, p.5). In addition, when conducting audit testing, the auditor is also required to set a “performance materiality” threshold below overall materiality to reduce to an appropriately low level the probability that the aggregate of undetected and uncorrected errors exceeds overall materiality (Millichamp and Taylor, 2012, pp.134-135). The latter is also known as “tolerable error” (Porter et al., 2003, p.338) or “tolerable misstatement” (Eilifsen and Messier Jr, 2014, p.5) and is the magnitude of detected errors that the auditor will tolerate without requiring management to make corrections, or otherwise consider the implications on the audit opinion according to ISA 705 ‘Modification to the Opinion in the Independent Auditor’s Report’ (IAASB, 2009e). Moreover, the auditor may set as many materiality levels as deemed necessary, including materiality for different classes of transactions, account balances and disclosures. Such thresholds may be set at any level below overall materiality (ISA 320, IAASB, 2009c, para.10; Porter et al., 2003).

Furthermore, ISA 450 ‘Evaluation of Misstatements Identified during the Audit’ (IAASB, 2009d, para.15) requires the auditor to determine and document a Clearly Trivial Threshold (CTT). Therefore, in summary, as a minimum, the auditor establishes the following quantitative materiality thresholds: overall materiality, performance materiality and CTT. In addition to these three thresholds, the auditor determines more thresholds when conducting a group audit and when reporting to an Audit Committee (AC) of the client entity (FRC, 2015b).

There are no prescribed methods or mathematical formulae to calculate such thresholds, although guidelines exist. As a starting point for determining the overall materiality threshold, auditors typically multiply a selected benchmark by a percentage. Auditors may also use blended methods by multiplying several benchmarks with corresponding percentages and then taking their average (Wiley, 2015). Furthermore, sliding scale methods may also be used. These methods consist of percentages which vary in steps depending on the size of the selected benchmark or the entity (Wiley, 2015). The performance materiality threshold and the CTT are then calculated as a percentage of overall materiality (Azzopardi and Baldacchino, 2009; Eilifsen and Messier Jr, 2014; Porter et al., 2003). Although materiality thresholds may be revised in the course of an audit, this paper focuses on the disclosure of the final amounts for such thresholds.

Items that exceed materiality thresholds are material by virtue of their size. However, items falling below such thresholds and purely qualitative disclosures in the financial statements can be material by virtue of their nature and/or impact on the users (Millichamp and Taylor, 2012; Wiley, 2015). For example, items that change a loss into a profit, conceal a management fraud or violate laws and regulations governing the auditee are material regardless of their size (Millichamp and Taylor, 2012; Turner, 2003).

A detailed review of existing literature identified the following benefits and drawbacks of disclosing materiality in the audit report:

Benefits of Materiality Disclosures

(i) Improved market efficiency: Fisher (1990) found that trades in a simplified experimental market setting are more correctly priced when audit materiality levels are disclosed because users can better assess the relative quality of the financial statements. This holds for both public and private disclosure of materiality, with the former appearing to be more effective than the latter.

(ii) More accurate perceptions of the audit report by investors: Non-disclosure of materiality often leads to the misconception that a clean audit report guarantees that the financial statements are correct (Houghton, 2010). In this context, Davis (2005) found that materiality disclosures help investors perceive the audit report more accurately by being less overconfident.

(iii) Alignment of auditors’ materiality judgements with those of users: Since materiality is defined from the users’ perspective, auditors’ materiality judgements should, at least in theory, be identical to (or aligned with) those of the users (Højskov, 1998). However some discrepancy exists and it is undesirable since it allows for the existence of errors deemed immaterial by auditors that can affect users’ decisions on the basis of the financial statements (Doxey, 2013). Positively, Turner (2003), Højskov (1998) and Leslie (1985 cited in Davis, 2005) assert that alignment may be possible via materiality disclosures.

(iv) A reduction of the Audit Expectations Gap (AEG): One of the persistent issues in audit reporting has always been the AEG to which the lack of understanding of the concept of materiality and its non-disclosure is a significant contributing factor (De Martinis and Burrowes, 1996; Houghton, 2010). Shaikh and Talha (2003), Højskov (1998) and De Martinis and Burrowes (1996) sustain that this AEG may be reduced via disclosure of materiality information, by helping readers of the audit report understand what the auditor actually does. In addition, the weak understanding of materiality is a major culprit of the AEG (Houghton, 2010). Disclosing materiality could thus narrow the AEG by improving users’ understanding of materiality and could possibly reduce auditor litigation costs (Doxey, 2013).

(v) Enhancement of the communicative value of the auditor’s report: According to Doxey (2013), disclosing more information in the auditor’s report, such as materiality levels, adds to the report’s value and transparency. Disclosure of quantitative and qualitative aspects of materiality would make the audit report more understandable, improve auditor-user communication, and enable users to assess the precision and level of assurance of the audit (Church et al., 2008; De Martinis and Burrowes, 1996; Doxey, 2013; Manson and Zaman, 2001). Such advantages would lead to the users making better decisions, whilst remaining vigilant of the possibility that audited financial statements still contain undetected errors (De Martinis and Burrowes, 1996).

(vi) The provision of more relevant information to credit institutions: A German study by Ruhnke et al. (2014) documented statistically significant evidence that as overall materiality thresholds are increased, the probability of granting credit decreases and the credit risk premium and credit collateral demanded by the lenders increase. Thus, materiality information is also relevant for credit institutions. This agrees with Litjens et al. (2015) who state that all information in the audit report is considered important by bankers.

Drawbacks of Materiality Disclosures

(i) Revealing the extent of ‘safe’ financial statement manipulation: Houghton et al. (2011) argue that materiality disclosures provide management with useful information as to the extent of financial statement manipulation they can engage in such that if detected, the auditor would still consider the misstatement as immaterial.

(ii) Risk that users misunderstand materiality disclosures: The average investor will find it difficult to comprehend materiality disclosures, especially if multiple thresholds are disclosed (Vanstraelen et al., 2012). The need for more education on materiality identified by Houghton et al. (2011) aggravates the risk that if materiality disclosures are not preceded by proper education, they would be misunderstood. This could create confusion (Litjens et al., 2015) and/or widen the AEG (Houghton, 2010) along with other potentially negative consequences. For instance, the ensuing controversy on materiality reduces the credibility of the audit (Azzopardi and Baldacchino, 2009). This is a real possibility when considering that users comprise a heterogeneous group, for which it is highly unlikely that every user would be satisfied with the auditor’s materiality judgements (CFA Institute, 2013). Positively, exposing such disagreement in the public domain may be an effective way of resolving it (Shaikh and Talha, 2003).

(iii) Higher audit costs, litigation costs and time-pressure on the auditor: Any expansion of the audit report such as the addition of materiality disclosures increases audit costs and puts more pressure on the auditor (Vanstraelen et al., 2012). Adding materiality disclosures to the audit report would inevitably lengthen it. This is a cause for concern, especially since the growth of disclosure may create an information overload which is undesirable to the users (Vanstraelen et al., 2012). Any disclosures added to the audit report could also become boilerplate language (Church et al., 2008) which regulators and investors seek to avoid (PCAOB, 2011). In addition, rather than limiting the auditors’ liability, litigation against auditors may be increased if errors deemed immaterial by the auditor are subsequently proven to be material (Højskov, 1998; Jennings et al., 1996; Vanstraelen et al., 2012). For instance, misstatements below materiality levels may cause material distortions in certain financial ratios (Turner, 1997) which may affect users’ decisions. Moreover, the auditor can easily be held liable for misstatements above materiality thresholds disclosed (PCAOB, 2011). Another risk is that the lack of prescriptive rules or standards on audit materiality and its associated subjectivity exposes the auditor to litigation for applying the concept of materiality inappropriately (Litjens et al., 2015).

(iv) Anchoring: Doxey (2013) claims that explicit disclosure of auditor materiality thresholds pulls users’ materiality judgements towards those of the auditor (anchoring). This implies that if users’ materiality levels are below those used by the auditor, anchoring would cause such levels to be pulled upwards. Therefore, users would end up taking more risk than they otherwise would in the absence of materiality disclosures.

(v) Impairment of professional judgement: The risk is that management would put forth certain arguments in an attempt to influence the auditor’s materiality judgements, and thus impinge on his independence (Houghton et al., 2011). For instance, management could attempt to convince the auditor that certain misstatements or aggressive accounting practices are immaterial. Such disagreement between management and the auditor poses a client retention risk for the auditor (Litjens et al., 2015) which can impair auditor independence.

(vi) Detrimental to firms with low profitability seeking credit: Materiality, along with other disclosures in the audit report may contribute towards a “self-fulfilling prophecy” (Vanstraelen et al., 2012, p.203) whereby revealing the risks that the entity is exposed to hinders its ability to generate external funds. Ruhnke et al. (2014) confirm that disclosure of liberal materiality thresholds makes it more difficult for firms with low profitability to obtain credit from banks. This situation is both a drawback from the auditee’s viewpoint, and an advantage for bankers since it enables them to avoid risky clients.

Other Issues

Existing literature identifies several reasons why auditors may resist materiality disclosures, other than the aforementioned drawbacks. Firstly, auditors sometimes waive misstatements above materiality thresholds. Such practice would no longer be possible if the thresholds are disclosed. Thus, for this reason, auditors may resist materiality disclosures (Wright and Wright, 1997 cited in Doxey, 2013). Secondly, auditors may consider materiality information as sensitive or proprietary (Manson and Zaman, 2001; Waters and Tiller, 1997 cited in Davis, 2005). This barrier has to be overcome if materiality is to be disclosed. Thirdly, if the larger audit firms use higher materiality thresholds than their smaller counterparts (Messier et al., 2005), they might be reluctant to disclose such thresholds in order to avoid giving the impression of a weaker audit and possibly damage their reputation. Fourthly, auditors may fear that disclosing materiality thresholds gives the impression that materiality is not a matter of professional judgement and could mislead users (Manson and Zaman, 2001).

Furthermore, the fact that auditors find it difficult to anticipate users’ reactions to materiality disclosures (Fisher, 1990) could be a reason for their hesitation to disclose materiality. To this effect, Fisher (1990) maintains that auditors may prefer disclosing materiality thresholds to management and others having good knowledge of the entity rather than publicly.

Moreover, the auditee may expect the auditor to use lower materiality levels if they are to be disclosed. This causes management to oppose materiality disclosures in fear of higher audit fees (Litjens et al., 2015). The auditee may also resist materiality disclosures because such information is useful for those seeking to commit fraud against the auditee. Thus the auditor might be reluctant to disclose materiality because the client does not want him to do so.

Materiality Disclosures in Practice

There are currently relatively few practical examples of materiality disclosures. These are discussed below.

PIEs in the UK and Ireland: As a result of ISA 700 ‘The independent auditor’s report on financial statements (UK and Ireland)’ (FRC, 2014) auditors of PIEs in the UK and Ireland which follow the UK Corporate Governance Code have started including materiality disclosures in their audit reports. Such disclosures also utilised diagrams and graphs to improve presentation and readability (FRC, 2015b). Despite encouraging feedback, the UK Financial Reporting Council (FRC) notes that improvements could be made by explaining materiality in greater detail and including which benchmarks and percentages were used (FRC, 2015a). Interestingly, awards were given to the most insightful and innovative audit reports by the UK Investment Management Association (IMA) in 2014 and 2015 (IMA, 2014; FRC, 2016). Moreover, auditors of PIEs in the Netherlands have to state the materiality threshold applied, explain its determination and that both quantitative and qualitative aspects were considered (NBA, 2014a; NBA, 2014b). In the US, the Public Company Accounting Oversight Board (PCAOB) has not yet mandated any materiality disclosures, but is closely monitoring their effects in the UK and Ireland (Bagshaw, 2015).

EU Regulation No. 537/2014: This EU Regulation (The European Parliament and the Council of the European Union, 2014) mandates that as from 17th June 2016, auditors are to submit an additional report to the AC of PIEs which apart from other information, has to:

“disclose the quantitative level of materiality applied to perform the statutory audit for the financial statements as a whole and where applicable the materiality level or levels for particular classes of transactions, account balances or disclosures, and disclose the qualitative factors which were considered when setting the level of materiality;” – Article 11(2)(h)

Methods of Disclosing Materiality

Literature on how materiality should be disclosed is quite limited. According to Fisher (1990) both public and private disclosures improve market efficiency, although public disclosures appeared more effective. This agrees with the Financial Accounting Standards Board’s (FASB) recommendation “that auditor materiality thresholds be publicly reported” (Davis, 2005, p.4). However, these indications are by no means conclusive.

Another issue that is not studied exhaustively in the literature pertains to the extent of rules needed to regulate materiality disclosures. Limited evidence from the UK demonstrates that materiality disclosures mandated by purposefully vague, principle- based rules were successful, as auditors sought to innovate and improve the materiality disclosures in their report (FRC, 2015b). Additionally, mandatory materiality disclosures may deter auditors from falsifying their materiality disclosures (Davis, 2005).

With respect to the presentation of materiality disclosures, several researchers have suggested supplementing paragraphs with web links, tables, diagrams, charts and footnotes (Church et al., 2008; FRC, 2015b; Houghton, 2010; Turner et al., 2010; Vanstraelen et al., 2012).

The Maltese Scenario

The only Maltese study available on materiality disclosures claims that “the proposal of disclosing materiality thresholds to reduce the omnipresent expectations gap was strongly rejected” amongst Maltese auditors (Azzopardi and Baldacchino, 2009, p.13). Several reasons were found justifying Maltese auditors’ apprehension with disclosing materiality. In particular, auditors believed that such disclosures would not be comprehended and might be perceived as giving absolute assurance up to the materiality level stated. Nevertheless, the study by Azzopardi and Baldacchino (2009) also found that materiality disclosures may be beneficial if adequately regulated and properly understood by users.

The purpose of this study is to achieve the following objectives:

Objective 1: to determine the benefits and drawbacks of disclosing the quantitative and qualitative dimensions of audit materiality in Maltese statutory audit reports, from the perspective of Maltese auditors.

Objective 2: to measure and explain the resistance towards, and frequency of materiality disclosures in Maltese statutory auditing.

Objective 3: to identify the preferences of Maltese auditors, on the location, method, content and regulation of materiality disclosures in the case of listed entities and non-listed entities.

Objective 4: to determine whether opinions on materiality disclosures differ significantly between Maltese present auditors and past auditors; and between auditors and non-audit accountants.

A mixed-methods research design consisting of a quantitative phase followed by a qualitative phase was adopted. This pragmatic methodology enabled the study to benefit from the strengths of both approaches.

Population and Sampling

The population consists of all 2,311 registered auditors and accountants listed on the Registered Accountants Register on the website of the Malta Accountancy Board (MAB) (2015a) as at 20^th October 2015. This population was divided into 1,225 registered auditors as per the Registered Auditors Register on the website of the MAB (2015b) as at 20^th October 2015, and the remainder of 1,086 non-audit accountants. This study focuses on the main population of 1,225 registered auditors. The whole population was included in the sample.

The population was targeted directly via email, indirectly through audit firms in Malta and by distributing printed copies of the questionnaire at an MIA Continuing Professional Education (CPE) event (MIA, 2015a).

Research Instruments and Participants

The study employed a self-administered questionnaire for the collection of quantitative data from Maltese auditors and non-audit accountants (vide Appendix 1). Quantitative data was collected between 20^th October and 19^th November 2015 and was subsequently analysed in order to develop an interview schedule for obtaining qualitative data from Maltese audit partners.

The questionnaire utilised a five-point Likert rating scale together with a few category questions and open questions. Section A of the questionnaire measured the auditors’ opinion on the benefits and drawbacks of disclosing materiality in the audit report using a list of statements derived from the literature review, followed by Section B which measured the auditors’ resistance towards materiality disclosures. Section C evaluated the auditors’ opinion on disclosing materiality in several reports (location), using different methods (method), disclosing different aspects of materiality (content) and whether materiality should be mandatory or voluntary (regulation). Finally, Section D elicited some other information which the researcher deemed relevant on the basis of existing literature. Section E concluded the questionnaire by collecting demographic data for statistical analysis. A pilot study of the questionnaire was also conducted and suggestions for improvement were noted.

A total of 241 valid responses were collected, of which 191 were complete responses. The majority of respondents were males (62.1%) and were fairly divided between big-four firms and other firms, and between all age groups. Approximately half of the respondents (50.8%) occupied the position of senior, manager or assistant manager. Furthermore, the majority of respondents (67.6%) had less than ten years of audit experience and 34.3% had three years of audit experience or less.

The interview schedule was developed from an analysis of the 241 valid responses that were collected till 19^th November 2015. The interview questions were carefully worded to explore further the key findings that emerged from the analysis of quantitative data.

The interview schedule consisted of four Sections with headers identical to those of the questionnaire. The most significant finding from each question in the questionnaire was stated, followed by one or two questions on the matter. Five partners from big-four audit firms, five partners from other audit firms and one technical manager at the MIA were selected, for a total of eleven semi-structured interviews.

Data Analysis Procedures

The quantitative data obtained from the questionnaires was analysed using IBM SPSS Statistics Version 23. Initially, frequencies and other descriptive statistics were generated as a first step in understanding the data. Secondly, the Friedman test was used to compare mean rating scores provided for a number of related statements and identify the statistically significant outcomes to be followed up via interviews.

Subsequently, the internal consistency, normality and linearity of the data were tested. The one sample t-test and binomial test were also conducted to determine whether the mean rating scores differed significantly from neutral. The research hypotheses were then tested using the post hoc pairwise comparison test and the Pearson correlation test. Appendix 2 includes a definition of the variables used in the study.

Responses from auditors were segregated into present auditors and past auditors; and responses from non-audit accountants were compared with auditor responses using appropriate statistical tests, in order to make full use of the data collected. A parsimonious univariate general linear regression model was then generated in order to explain quantitatively the resistance of Maltese auditors towards materiality disclosures.

The interview transcriptions were grouped by each question in the interview schedule. This enabled a key theme to be identified for every question. Themes relating to the topic as a whole were also identified.

This Section presents the quantitative and qualitative findings of the study. The findings are grouped according to the four objectives of the study that have been stated earlier on.

Objective 1: Benefits and Drawbacks of Materiality Disclosures in Audit Reports

The benefit that materiality disclosures align the auditor’s judgement of what is material with what really affects users of financial statements obtained the highest mean rating score (x̄=3.73) of all eight benefits that were presented in the questionnaire. In contrast, interviewees generally disagreed (8/11) with this alignment benefit. Such disagreement stemmed from the drawbacks of disclosing materiality, which according to interviewees greatly outweigh its potential benefits.

With respect to the drawbacks, respondents assigned the highest mean rating score (x̄=3.59) to the statement that users will interpret materiality thresholds as absolute assurance up to the level stated. Such drawback was the most prominent of all sixteen drawbacks that were presented to respondents. In addition, interviewees emphasised that the drawbacks of materiality disclosures greatly outweigh its benefits, as shown in Figure 1 (below).

Objective 2: Auditor Materiality Disclosure Resistance and Frequency

When presented with a list of seven statements intended to gauge their resistance to disclosing materiality, auditors assigned the highest score to the statement that it is difficult to anticipate users’ reactions to materiality disclosures (x̄=3.52). It was also found that auditors generally do not disclose materiality to their audit clients (x̄=2.36).

All interviewees (11/11) firmly agreed with this finding and expressed strong resistance towards materiality disclosures due to the drawbacks outlined above. Similarly, five audit firms had a policy of never disclosing materiality whilst in another five firms it was normal practice to avoid such disclosures and to only consider cautious disclosure if the client specifically requests such information. Two of the big-four firms never disclosed materiality thresholds, whilst the other two firms did in rare and exceptional circumstances.

Objective 3: Location, Method, Content and Regulation of Materiality Disclosures

Respondents assigned the highest mean rating score to disclosing materiality in an additional report to the AC at the end of the audit for listed entities (x̄=3.33). Correspondingly, for non-listed entities they assigned the highest score to disclosing materiality in the audit report (x̄=3.08). Such a score is very close to neutrality (x̄=3) and in fact interviewees believed there was no scope for materiality disclosures for non-listed entities.

With respect to the method for disclosing materiality, respondents agreed mostly with paragraphs (Listed x̄=3.20; Nonlisted x̄=3.09) and disagreed mostly with a link to a website (Listed x̄=1.88; Nonlisted x̄=1.84). Interviewees considered the use of paragraphs (10/11) as the most appropriate

and flexible method to disclose materiality, whilst Materiality Per Share (MPS) proposed by Turner (2003) and a link to a website were strongly rejected (MPS: 11/11; Web: 10/11).

Respondents were also presented with ten statements on what materiality disclosures could contain, for listed and non-listed entities respectively. For both types of entities, respondents gave the highest score to the materiality threshold for the financial statements as a whole (Listed x̄=3.61; Nonlisted x̄=3.46). Interviewees (11/11) suggested that in addition to overall materiality, the following should also be disclosed: materiality levels for classes of transactions, account balances or disclosures (1/11), performance materiality (1/11), significant qualitative considerations (2/11), the benchmark used together with the reasons for selection (4/11) and the percentage applied to the selected benchmark (1/11).

Moreover, respondents and interviewees (9/11) preferred making materiality disclosures mandatory for listed entities (x̄=3.19) rather than for non-listed entities (x̄=2.48). Respondents and interviewees also agreed that users should be well educated before introducing materiality disclosures (x̄=3.89, 10/11), and that materiality disclosures must be well regulated (x̄=3.84, 9/11). Interviewees’ opinions were fairly equally distributed between principle-based (3/11), rules-based (4/11) or a mixed style (3/11) of rules.

Furthermore, respondents and interviewees strongly disagreed with giving awards or formal acknowledgement to Maltese audit firms with the best materiality disclosures (x̄=2.19, 11/11). Most of the audit partners in big-four firms knew that materiality disclosures in the audit report are mandatory for PIEs in the UK and Ireland and were also aware of Article 11(2)(h) of EU Regulation No. 537/2014.

Objective 4: Present Auditors, Past Auditors and Non-Audit Accountants

The 241 responses from auditors comprise 133 present and 108 past auditors. Past auditors rated AMD (“Audit Materiality Disclosures”) Benefits, AMD Location Listed and Nonlisted, AMD Method Listed and Nonlisted, AMD Content Nonlisted and AMD Mandatory significantly higher than present auditors. Conversely, past auditors scored AMD Drawbacks and AMD Resistance lower than present auditors.

Non-audit accountants scored AMD Benefits, AMD Mandatory and AMD Awards significantly higher than auditors, whilst they scored AMD Drawbacks lower than auditors. Moreover, non-audit accountants were more in favour of mandatory materiality disclosures for listed entities than auditors.

Hypotheses Testing

The results of hypotheses testing are presented in Table 1.

Linear Regression Model

A parsimonious univariate General Linear Model (GLM) was generated in order to explain quantitatively the resistance of Maltese auditors towards materiality disclosures, thereby achieving Objective 2 more comprehensively. The dependent variable is AMD Resistance. All generated scores range from 1 to 5, where 1 corresponds to ‘no extent’ and 5 corresponds to ‘high extent’. The parsimonious model was identified by using a backward procedure. Equation 1 (Final GLM) defines the model mathematically. This four-predictor model explains 44.2% of the total variance of AMD Resistance.

The discussion presented in this Section is structured according to the four objectives of the study.

Objective 1: Benefits and Drawbacks of Materiality Disclosures in Audit Reports

The drawbacks of disclosing materiality in Maltese statutory audit reports greatly outweighed any potential benefits that might be reaped. Drawbacks revolved around confusion, misinterpretation, the illusion of absolute assurance up to the thresholds disclosed, criticism against the auditors, impairment of professional judgement, manipulation below materiality thresholds, and more audit work. Benefits included alignment of the auditor’s judgement of what is material with what really affects users of financial statements, a possible reduction of the AEG and facilitation of discussions of audit adjustments with clients. However, users would only reap such benefits if they properly understood the materiality disclosures. Otherwise, the drawbacks mentioned above would most likely occur and possibly widen the AEG further.

Furthermore, users of financial statements might use the auditor’s materiality thresholds as a guide for deciding whether an amount is relevant for taking a decision. Thus, the anchoring drawback stipulated by Doxey (2013) was also supported by Maltese auditors. Such behaviour would be problematic since materiality is primarily an audit tool which does not cater for specific requirements of different users. Moreover, in a dynamic environment, thresholds could become obsolete by the time users consider them for decision making and the users comprise a heterogeneous group for which it is highly unlikely that every user would be satisfied with the auditor’s materiality judgements. This amplifies the risk of reduced auditor credibility and might also cause users to pose conflicting demands on the auditors.

Objective 2: Auditor Materiality Disclosure Resistance and Frequency

The second objective of the study was to measure and explain the resistance of Maltese auditors towards materiality disclosures; and their frequency of disclosing it to their clients. AMD Resistance was measured using a five-point Likert scale and explained by the GLM. As expected, a strong link between the aforementioned drawbacks and AMD Resistance emerged.

In addition, auditors preferred to restrict materiality disclosures to the more knowledgeable persons making up the ACs of listed entities. This was probably due to their fear that materiality disclosures would be misunderstood causing unanticipated user reactions. This argument supports similar assertions by Fisher (1990).

Furthermore, auditors may have opposed such disclosures because materiality information is considered sensitive or proprietary and to avoid possibly giving an impression that materiality is merely a mechanical calculation rather than a matter of professional judgement (Manson and Zaman, 2001; Waters and Tiller, 1997 cited in Davis, 2005). Moreover, it seems that some audit clients did not want the auditor to disclose materiality levels in the audit report, whilst other clients had no such reservation. Thus, auditors expressed mixed views of auditee resistance to materiality disclosures.

Interestingly, auditors may also resist disclosing materiality due to the risk of tipping off a potential fraudster when explaining the application of materiality in an audit and because such disclosures would inevitably be historic, thus magnifying the risk of user misinterpretation. These reasons were not previously contemplated in the literature and could lie within the 55.8% of the variability that was not explained by the GLM.

With respect to the frequency of disclosure, auditors never disclosed materiality in the audit report and they rarely disclosed it to their audit clients. This confirms that earlier findings by Azzopardi and Baldacchino (2009) still apply in the local scenario. The non- disclosure of materiality is reinforced at the audit firm level through formal policy or informal common practice. In addition, the frequency of disclosing materiality did not depend on whether the entity was listed or non-listed (H4 was rejected; vide Table 1) or on the job position of the auditor within the firm. Furthermore, auditors’ resistance towards materiality disclosures did not affect how often they disclosed it to their clients (H3 was rejected; vide Table 1), implying that some auditors did so reluctantly.

Objective 3: Location, Method, Content and Regulation of Materiality Disclosures

Location refers to the type of report where materiality could be disclosed. Due to the resistance towards materiality disclosures, auditors were more inclined to agree with a report that was restricted to knowledgeable individuals who were least likely to use materiality for their own advantage. In fact, if materiality is to be disclosed, auditors recommended that for listed entities, it should be disclosed to the AC and for non-listed entities, it should be disclosed to their board of directors, if and only if such disclosure cannot be avoided. In addition, both quantitative (H5 was accepted: vide Table 1) and qualitative findings confirmed that auditors saw no scope for materiality disclosures for non-listed entities.

By opting for the more restrictive locations, auditors disregarded Fisher’s (1990) finding that public materiality disclosures were more effective at improving market efficiency and dismissed the FASB’s recommendation “that auditor materiality thresholds be publicly reported” (Davis, 2005, p.4). It seems that widening the audience of materiality disclosures in Malta would cause its drawbacks to increase more rapidly than its benefits and overwhelm them completely.

Method refers to the way in which materiality could be disclosed in the audit report. Interestingly, a significant positive relationship (+0.131) between AMD Resistance and AMD Method Listed was discovered from the GLM (Vide Equation 1). A plausible explanation of this relationship would be that as auditors’ resistance towards disclosing materiality increased, their fear of its drawbacks also increased and in order to mitigate such drawbacks, they opted for more elaborate methods of disclosing it. This would have been reflected in a higher AMD Method Listed score.

If materiality is to be disclosed in the audit report, according to Maltese auditors, this should be done via paragraphs possibly aided by tables, with no distinction between listed and non-listed entities (H6 was rejected; vide Table 1). Paragraphs have the advantage of being the most flexible method, whilst tables may complement them by adding a structure which could facilitate comparability. Nevertheless, auditors tended to disagree with all methods of disclosing materiality due to their resistance towards these disclosures.

Contrary to Turner et al. (2010) and in disagreement with the FRC’s (2015b) praise of elaborate materiality diagrams in UK audit reports, footnotes, diagrams, MPS and links to websites should not be included. Moreover, the use of a link to a website as suggested by Church et al. (2008), FRC (2015b), Houghton (2010) and Vanstraelen et al. (2012) as well as the MPS proposed by Turner (2003) were strongly rejected. The opposition towards web links stemmed from the concern that they would further complicate the financial statements which are already difficult to read.

Content refers to the information that materiality disclosures would actually contain. The overall threshold for the financial statements as a whole should definitely be included for both types of entities, yet it is essential that materiality is disclosed in a complete and comprehensive manner, which certainly would not be achieved by simply disclosing a number. Thus based on the findings of the study, the information outlined in Table 2 (below) could be disclosed in order to communicate materiality information more effectively. These suggestions are relevant for materiality disclosures in the audit report and to the AC.

Regulation: In agreement with the finding that auditors resist disclosing materiality, they also opposed making such disclosures mandatory. However, if materiality disclosures were to be made mandatory, then auditors believed they should be well regulated and as far as possible they should only be made mandatory for listed entities (H8 was accepted: vide Table 1).

In comparison with limited evidence from the UK and Ireland of the success of principle- based materiality disclosure rules (FRC, 2015b), it seems that if materiality disclosures are introduced in Malta, they would have to be regulated using a mixed style that has a certain element of rules but still allows judgement to be exercised.

Objective 4: Present Auditors, Past Auditors and Non-Audit Accountants

Finally, the fourth objective revealed that the opinions of auditors who were practicing the profession at the time the data was collected (present auditors) were significantly more against disclosing materiality compared with auditors who were no longer active in the profession (past auditors). It was also found that accountants without audit experience (non-audit accountants) were more in favour of materiality disclosures than auditors. Therefore it seems that the farther away from the actual audit work, the higher the support for materiality disclosures. In fact, most benefits of materiality disclosures in the literature emerged from studies whose samples did not include auditors.

Overall, Maltese auditors resist materiality disclosures because their highly probable drawbacks greatly outweigh their potential benefits. Maltese auditors saw no scope for materiality disclosures for non-listed entities. In the case of listed entities, they restricted such disclosures to the AC. Thus the requirement of Article 11(2)(h) of EU Regulation No. 537/2014 to disclose materiality to the AC of PIEs should not be a cause for concern.

Materiality disclosures in the audit report should be avoided because the local market is not mature and knowledgeable enough to appreciate such disclosures as is currently being done in the UK, Ireland and the Netherlands. Moreover, if materiality is to be disclosed, only paragraphs, possibly aided by tables, should be considered. Materiality disclosures should contain the threshold for the financial statements as a whole, along with explanatory information (see Table 2). Regulation of such disclosures should be restricted to listed entities only and should be of a mixed-style which has a certain element of rules but still allows judgement to be exercised.

In conclusion, materiality disclosures in Maltese statutory auditing would have far more drawbacks than benefits and should therefore be avoided. Positively, restricting such disclosures to the AC of PIEs limits their drawbacks and as a big-four audit partner remarked, may “lead to intelligent disclosures and intelligent discussions” between auditors and the AC.

Our study focused on the views of Maltese auditors only. We recommend further research on this subject from the points of view of other users of audited financial statements such as shareholders, banks, tax authorities and creditors. Our study also focused exclusively on materiality disclosures and therefore it did not analyse other ways to enhance the audit reporting model, such as the inclusion of other disclosures in the context of an audit. Finally, external auditing in the public sector was also excluded from the scope of this study, and we believe that this is an interesting avenue for further research.