Audit partner narcissism and audit reports readability: evidence from lead and review audit partners Open Access

This study examines the impact of auditor narcissism on the readability of audit reports in Iran, an increasingly pertinent topic as financial reporting standards globalize. Despite the unique regulatory requirements and auditing practices in Iran, the influence of auditor personality traits, particularly narcissism, on audit outcomes has been underexplored. This research seeks to fill this gap by investigating how narcissistic traits among audit partners might affect the clarity and overall quality of audit reports.

Auditing serves as an intermediary function that enhances the credibility of financial information for businesses and investors (Healy and Palepu, 2001). The outcomes of audits are significantly influenced by the personalities and behaviors of the individuals responsible for overseeing and conducting the audit process. Recent studies have highlighted the substantial role of audit partners in shaping these outcomes. For example, research by Chen et al. (2020) and Pham et al. (2023) illustrates how factors such as partners’ workload and their cultural alignment with CFOs can affect audit quality and the likelihood of issuing modified opinions. In addition, Ittonen et al. (2015) and Maghakyan et al. (2024) emphasize the critical importance of specialized knowledge – including digital competencies and experience with public clients – in enhancing audit effectiveness. Strategic decisions by audit firms to assign partners with sector-specific expertise to higher-risk clients have been shown to improve audit quality by mitigating engagement risks (Wu et al., 2023). Moreover, ongoing collaborations between audit committee chairs and lead partners have been proven to enhance financial reporting quality while preserving the independence of audit fees (Bedford et al., 2023).

The characteristics and behaviors of audit partners profoundly affect the integrity and effectiveness of audits (Zerni, 2012). Given their pivotal role, the personality traits of these partners, especially narcissism, are of significant interest to stakeholders who rely on transparent and accurate financial reporting and the extent to which the audit function supports these qualities (Duff, 2009). Narcissism is typified by traits such as grandiosity, entitlement, dominance and a heightened sense of self-importance, which may influence decision-making processes and interpersonal interactions within the audit context (Li et al., 2023). Research indicates that while the overconfidence associated with narcissism may lead to riskier behaviors (Olsen and Stekelberg, 2016; Chatterjee and Hambrick, 2007; Nevicka et al., 2011), it can also enhance auditor independence and assertiveness, potentially improving audit quality (Chou et al., 2021; Church et al., 2020).

Recognizing the critical role of audit partners in determining audit quality, it is essential to consider the readability [¹] of audit reports as a direct indicator of this quality (Rajabalizadeh, 2024b). The clarity and comprehensibility of these reports are crucial for their usability and for enabling stakeholders to understand and assess the credibility and validity of the financial information presented. However, audit reports often exhibit low readability, which complicates comprehension (Boolaky and Quick, 2016; Coram et al., 2011). Enhancing the readability of audit reports is not merely about improving the communicative value of these documents; it is also about strengthening the integrity and reliability perceived by the market (De Franco et al., 2015; Loughran and McDonald, 2014; Seebeck and Kaya, 2022). Presenting complex audit findings in an accessible manner necessitates carefully crafted narratives that tackle the intricacies of detailed accounting and financial issues, as underscored by Xu et al. (2020). Sections such as the audit opinion and the basis for opinion are essential for providing a comprehensive understanding but contribute to the complexity of the reports. Thus, improving readability can have significant economic implications, enhancing the transparency necessary for informed investment decisions and alleviating information asymmetries and incentive problems that impede the efficient allocation of resources in capital markets.

Despite the collectivist culture of Iran, which generally discourages self-promotion, the structure of its auditing practices might inadvertently promote such traits, creating a unique amalgamation of cultural norms and professional behaviors (Rajabalizadeh, 2024a). In Iran, the absence of international audit firms and distinct auditing standards provides a specific context for examining how narcissism could influence the clarity of audit reports. A concise analysis of the differences between Iranian Standards on Auditing and International Standards on Auditing shows that while the international standards typically emphasize uniformity and comparability across borders, Iranian standards appear more aligned with domestic business practices and regulatory requirements. This alignment may lead to variations in audit scope, documentation and reporting responsibilities (Mirshekary and Saudagaran, 2005). The substantial influence individual auditors have over these reports, within Iran’s specific regulatory framework and auditing practices, adds significant complexity to this dynamic. This intricate interplay between is further explored in Section 2, where the impact of narcissism on audit outcomes is detailed, incorporating insights from recent research (Wan Hussin et al., 2018; Nazarian and Atkinson, 2013; Cooper and Pullig, 2013; Kurpis et al., 2008).

This study uses signature size as an indicator of narcissism (Ham et al., 2017, 2018; Chou et al., 2021) and uses the FOG index to assess the readability of audit reports (Li, 2008; Biddle et al., 2009; Lehavy et al., 2011; Hesarzadeh and Rajabalizadeh, 2019; Salehi et al., 2020; Hesarzadeh et al., 2020). Our analysis is informed by a multifaceted theoretical framework: Upper Echelons Theory suggests that the personal traits of audit partners, particularly narcissism, influence audit outcomes, leading to complex and less readable reports (Wang et al., 2016; Joullié et al., 2021). Institutional Theory elucidates how auditors might adjust report readability to conform to ethical norms (Filatotchev and Nakajima, 2014). Agency Theory examines how client power dynamics affect readability adjustments (Adams, 1994). A comprehensive discussion of these theoretical perspectives and their relevance to our study is presented in Section 3.1, titled “Theoretical Underpinning.”

Our analysis substantiates the first hypothesis (H1), positing a significant relationship between the narcissism of audit partners and the readability of audit reports. We found that audit partners with narcissistic tendencies generally produce reports of lower readability, as indicated by higher FOG index scores. This phenomenon is rooted in their approach to managing accounting conflicts and their preference for autonomous decision-making. The heightened scrutiny of narcissistic partners toward potential client deception and errors manifests in reports that are both more comprehensive and complex. These findings align with previous studies (Ma and Jaeger, 2005; Hüffmeier et al., 2014), confirming the impact of auditor personality traits on the attributes of audit reports.

In addition, our research supports the second hypothesis (H2), elucidating the distinct impacts of narcissism among both lead and review audit partners on report readability. Notably, the narcissism of lead partners significantly exacerbates the complexity of audit reports. This finding highlights the divergent roles that lead and review partners play in determining the outcomes of reports. The association between the partners’ narcissism and report complexity is consistent across various readability metrics, including the SMOG (Simple Measure of Gobbledygook) and the ARI (Automated Readability Index). This relationship is demonstrated to be robust through analyses correlating partners’ narcissism rankings and the sizes of their signatures, thus underscoring the pervasive influence of narcissism on the presentation of audit reports.

Further analysis reveals that narcissistic traits contribute to increased complexity in detailed audits, consistently resulting in decreased readability, even in specialized or intricate audit scenarios. The study also examines the effects of client economic significance and partner narcissism on readability. We found that reports concerning larger clients tend to be more readable, suggesting adjustments are made for clients of substantial economic importance. However, our findings show no significant relationship between partner quality and readability, highlighting that narcissism, rather than professional competence, predominantly influences readability. A positive relationship between the rank of an audit firm and partner narcissism indicates that higher-ranking firms tend to use more narcissistic auditors who produce complex reports, thus emphasizing the influence of firm ranking on the presentation of audit reports. In addition, our research suggests that the combined narcissism of auditors and CEOs may exacerbate challenges, resulting in less clear and potentially more biased audit reports, especially in contentious negotiation scenarios. The implementation of International Auditing Standard 701 has not significantly mitigated these effects, indicating that while the standard aims to enhance transparency, entrenched narcissistic traits continue to overshadow these regulatory efforts. This underscores the challenge of mitigating inherent personality influences within audit practices through standard regulatory measures alone.

Our study contributes to the advancement of accounting theory by integrating the psychological trait of narcissism into the analysis of audit outcomes, adopting a behavioral accounting perspective. This approach investigates how individual behaviors and psychological traits, particularly narcissism as identified among Iranian auditors by Banimahd et al. (2013), influence accounting and auditing practices. In addition, we use concepts from Agency Theory and Upper Echelons Theory to assess how narcissistic traits affect the professional judgments of audit partners. Agency Theory is expanded to explore the interaction between auditors’ psychological characteristics and their independence and objectivity, supported by studies from Hambrick (2007) and Liu (2023). Moreover, Upper Echelons Theory, which asserts that organizational outcomes reflect leaders’ characteristics, supports our examination of how leadership narcissism impacts audit practices, further explored by Brunzel (2021) and Zengin-Karaibrahimoglu et al. (2021).

Building upon prior research into auditors’ professional attributes, such as education and experience (Salehi et al., 2022), our study explores personality traits within the framework of Iran’s mandatory disclosure regulations. This context allows us to examine the influence of transparency on report presentation and contributes to global discussions on audit quality and partner identification, as detailed by Fisher et al. (2016). In addition, our research ties into investigations of negative personality traits and behaviors in accounting, including tax evasion (Olsen and Stekelberg, 2016) and dishonesty (Cohen et al., 2017; Campbell et al., 2011). We investigate how narcissism affects auditors’ decision-making and transparency and analyze the correlation between signature size – a direct indicator of personal accountability – and reporting patterns, following DeZoort et al. (2006). This analysis aids in devising strategies that align auditor traits with client needs, thus enhancing our understanding of auditor style and its impact on audit outcomes (Goodwin and Wu, 2016).

Building on this foundation, our study introduces a novel contribution to the literature by examining the specific effects of auditor narcissism on audit report readability within the Iranian audit context, which significantly differs from those previously studied. While existing research such as that by Church et al. (2020) and Johnson et al. (2021) has illuminated the effects of auditor narcissism in China and globally, focusing on various audit outcomes and auditor-client negotiations, our investigation delves into this phenomenon within the distinctive Iranian setting. Here, the manifestations of auditor narcissism are particularly apparent in less readable audit reports. This finding contrasts with prior research and underscores the significant role of regulatory environments and auditing practices in shaping the consequences of auditor narcissism. By expanding the geographical focus of the research and deepening the understanding of how auditor narcissism affects audit processes and outcomes in non-Western contexts, our study provides essential insights. The observed differences highlight the need for a nuanced approach to assessing auditor behavior across diverse regulatory and cultural landscapes, which is crucial for academics and practitioners aiming to improve audit quality and transparency globally.

The subsequent sections of the paper are structured as follows: Section 3 outlines the theoretical relationship between auditors’ narcissism and the readability of audit reports. Section 4 details the data and research design, while Sections 5 and 6 present the empirical findings, additional tests and a sensitivity analysis. Section 7 discusses the implications and concludes the paper.

Iran presents a distinct context for investigating the influence of auditor narcissism on audit report readability. In Iran, social and professional behaviors are deeply influenced by Islamic values (Dastmalchian et al., 2001). Traits associated with narcissism, such as a heightened sense of self-importance, might arise, potentially conflicting with these religious norms (Cooper and Pullig, 2013; Kurpis et al., 2008). Such traits could lead auditors to produce more complex, less comprehensible audit reports as a means to demonstrate their expertise. Conversely, recent research suggests that narcissism might also promote audit quality by enhancing auditors’ independence, indicating a complex role for personal traits in the efficacy of auditing in Iran (Salehi et al., 2022).

The auditing process in Iran, delineated through the roles of lead and review partners, illustrates a confluence of individual motivations and broader socioeconomic factors. Although Iranian audit reports adhere to a traditional format that diverges from Western practices – often emphasizing key audit matters – they still provide critical market insights. The auditing environment is marked by significant government oversight, lax governance and opaque practices, emphasizing the importance of personal relationships and risk tolerance in audit outcomes (Nazarian and Atkinson, 2013). This setting forms a framework in which the effects of auditor narcissism may be either mitigated or exacerbated, especially during periods of increasing alignment with International Standards on Auditing. In this challenging environment, auditors in Iran are under considerable pressure to adhere to both national and international standards, influencing their professional behavior and the clarity of their reports. This scenario facilitates a comprehensive examination of how these pressures, along with the distinctive regulatory framework in Iran, affect auditor behavior, offering insights potentially unobservable in more liberal or less regulated contexts. Recent studies by Safarzadeh and Mohammadian (2024) indicate that narcissism among auditors positively affects professional skepticism. Nonetheless, the research also points to variations in narcissism across different demographics, noting higher levels among younger auditors and those in the private sector (Banimahd et al., 2013). These observations underscore the intricate relationship between auditor traits, market dynamics and reporting practices within Iran’s unique regulatory landscape.

Furthermore, challenges such as international sanctions and political constraints distinctively position Iran’s audit sector. These factors, while restricting the presence of global firms, foster a vibrant and diverse local auditing environment. The absence of large, international firms encourages a broader variety of audit practices and decision-making processes. This situation potentially strengthens local auditors’ skills, reduces audit costs, minimizes conflicts of interest and enhances competition, which could improve transparency. This unique structure may influence how narcissism manifests in audit reports. Local firms, perhaps striving to compete with or differentiate from global standards, may adopt reporting styles that more prominently reflect individual auditors’ traits, making the study of narcissism’s impact on report readability particularly pertinent. Research indicates that narcissism among audit partners, evident from signature size, enhances both perceived and actual audit quality, particularly for larger clients and when auditors have personal relationships with client executives (Chou et al., 2021). Recent findings by Zudana et al. (2023) demonstrate that CEO narcissism improves the readability of tax footnotes, affirming that executive traits significantly impact reporting clarity. These studies underscore the complex relationship between narcissism and various facets of financial reporting and audit quality within distinct regulatory and market contexts like those in Iran.

Finally, mandatory audit procedures and regulatory oversight in Iran play an essential role in upholding high standards of practice within a multifaceted regulatory and economic landscape. Audits require signatures from certified partners and mandate firm rotations every four years, complemented by rigorous annual evaluations conducted by the Iranian Association of Certified Public Accountants (IACPA), which assesses firms based on detailed quality criteria, assigning grades from A to D (MohammadRezaei et al., 2018). Top-ranked firms, accredited by the Security Exchange Organization, are able to command higher fees and provide superior services, capitalizing on their extensive experience and broad service range. Despite these stringent regulations, the actual content of audit reports often varies significantly, reflecting disparities in how individual firms interpret and implement these standards. Such variability can result in marked discrepancies in report readability, underscoring the challenges in ensuring consistent audit report quality across the industry. This inconsistency is further affected by factors such as auditor tenure, fees and specialization, which have been shown to negatively impact report readability. Conversely, intense competition in the audit market, particularly among private firms, tends to improve it (Rajabalizadeh, 2024b). This competitive environment is especially pronounced in contexts where audit teams remain unchanged over long periods and in smaller market scenarios.

These dynamics illustrate the intricate interplay between personal traits, professional norms and market conditions that shape audit practices and report quality in Iran. The next section will provide a theoretical foundation for these observations, aiming to deepen our understanding of how these various elements influence audit quality and financial reporting within the unique Iranian context.

The role of individual auditor traits, particularly narcissism, and their impact on audit outcomes is crucial for understanding auditing practices, especially within the distinct regulatory and cultural landscapes of Iran. This section focuses on the significant influence of narcissistic traits among auditors on the readability of audit reports, integrating insights from upper echelons theory, institutional theory and agency theory. This analysis not only deepens our understanding of the auditor’s role but also sets the stage for subsequent hypothesis development.

Upper Echelons Theory suggests that organizational outcomes, such as strategic decisions and performance levels, are influenced by the background characteristics of top executives, including CEOs and audit partners (Wang et al., 2016). In the Iranian context, where individual traits significantly affect professional behaviors due to cultural and religious factors (Naslmosavi and Jahanzeb, 2017), narcissistic traits in audit partners could lead to more complex audit reports. These leaders might use elaborate language to project superiority or distinctiveness, thereby imprinting their personal traits onto organizational outcomes (Joullié et al., 2021). In this regard, pronounced narcissistic traits could compel audit partners to enhance report readability to demonstrate their competence and earn respect and admiration within the collectivist frameworks of Iranian society. Conversely, they might opt to obscure readability to create an illusion of complexity, believing that it reflects their intellectual prowess in an environment that values modesty.

Transitioning from the discussion on Upper Echelons Theory, we delve into Institutional Theory, which examines how the broader regulatory and cultural environment in Iran could shape auditors’ motivations and behaviors concerning report readability. Institutional Theory posits that organizations adopt practices more for legitimacy than for efficiency, often in response to normative pressures within their environment (Filatotchev and Nakajima, 2014). Given the strong religious and cultural ethos in Iran, which significantly shapes business practices (Zamani-Farahani and Musa, 2012), auditors may be compelled to enhance the readability of their reports to align with the ethical and professional standards expected in Iranian society. This alignment with institutional norms serves to gain legitimacy. However, if the institutional environment is perceived as valuing detailed and intricate reporting as a sign of thoroughness and expertise, auditors might increase the complexity of their reports to meet these expectations.

Finally, we expand our theoretical exploration to include Agency Theory, which primarily investigates the conflicts between principals (clients) and agents (auditors). Agents may act in their own interests, diverging from the interests of their principals (Adams, 1994). In addition, we consider how the bargaining power of clients influences auditors’ behaviors and decision-making processes (Windsor and Ashkanasy, 1995). In the context of Iranian audits, clients may possess substantial bargaining power due to their economic or political ties (MohammadRezaei and Mohd-Saleh, 2017; MohammadRezaei et al., 2016). Such power can impact how auditors respond to client preferences in their audit reports. For example, if powerful clients demand transparency and straightforwardness to meet regulatory requirements or enhance market reputation, they might pressure auditors to produce clearer audit reports. In this scenario, narcissistic auditors might view compliance as an opportunity to curry favor and boost their own prestige (Young et al., 2016). On the other hand, if clients with significant bargaining power prefer less transparency – perhaps to conceal poor performance or reduce tax liabilities – they may push for more opaque reports. Narcissistic auditors, driven by the desire to sustain profitable client relationships or showcase their capability in managing complex situations, may accommodate these demands (Johnson et al., 2013), thus diminishing report readability.

With a robust theoretical framework in place, we now shift our focus to reviewing literature and developing hypotheses. The forthcoming section will integrate empirical studies and theoretical perspectives to deepen our understanding of how auditor characteristics influence the readability of audit reports. Our examination of the existing literature will highlight gaps that our study seeks to address, thus paving the way for the presentation of our hypotheses.

The influence of narcissism on auditing effectiveness is multifaceted. Narcissistic traits, such as risk underestimation and overconfidence, can compromise financial decisions and leadership, thereby undermining the teamwork essential for accurate audits. These traits may lead to the oversight of critical errors, producing shorter, ostensibly clearer audit reports (Olsen and Stekelberg, 2016; Chatterjee and Hambrick, 2007; Nevicka et al., 2011). Conversely, narcissistic resistance to conformity and persistence in views can result in deeper audits, yielding longer, more detailed reports that might impair readability (Maccoby, 2000, 2003; Xiang and Song, 2021). This dual impact of narcissism necessitates a nuanced exploration of its effects on audit report readability.

Previous studies, such as Mock et al. (2013), have underscored the importance of both the informational content of audit reports and users’ reactions to this information. The quality of the narrative in audit reports is pivotal; however, research on factors influencing readability, as suggested by Lo et al. (2017), is limited. Readability is crucial for narrative quality and can be assessed by the presentation of information, which directly affects its perceived clarity to users. Further research by Abdelfattah et al. (2021) reveals that auditor characteristics, such as gender, can influence report readability, with female audit partners often producing less readable reports. Conversely, Smith (2023) notes that the inclusion of key audit matters in UK reports enhances readability, significantly affecting user interpretation and response.

Regarding the balance between standardization and specificity, Fakhfakh (2016) demonstrates that standardized audit reports can be difficult for financial statement users to understand, highlighting a gap between the intent of standardization and its practical effects. Conversely, Dyer et al. (2017) argue that the use of specific and verifiable language, including numerical data and organizational names, enhances readability and the reliability of the information provided. Regulatory bodies have initiated efforts to enhance financial disclosure clarity, with significant contributions from the Securities and Exchange Commission (SEC, 1998) and the International Auditing and Assurance Standards Board (IAASB, 2015). Supporting this, research by Biddle et al. (2009) and Asay et al. (2017) shows that less readable disclosures provoke weaker investor reactions, leading to greater reliance on external information. This highlights the crucial role of audit report readability as a reliable external information source. Zeng et al. (2021) extend this discussion by examining how textual characteristics such as readability, length and specificity correlate with audit quality, suggesting that improving these textual features can significantly enhance audit quality. This comprehensive review underscores the fundamental importance of readability in the effectiveness of financial communication and its regulatory implications, closely aligning with the central themes of audit report quality.

In the context of the literature reviewed, the potential of narcissism to either enhance or impair readability is particularly relevant in situations where lead and review audit partners in Iran must collaborate. High levels of narcissism might lead auditors to generate reports that are either overly complex or strategically straightforward, influenced by their professional motives and team dynamics. The dual-signature requirement aims to improve disclosure quality and reduce agency costs, but it also introduces complexity by illustrating how personal ambitions can either conflict with or complement collective audit objectives (Ittonen and Tronnes, 2015). Considering these dynamics and the nuanced role of narcissism described in our theoretical framework, the study proposes the following nondirectional null hypotheses to capture the intricate interactions affecting the readability of audit reports:

This hypothesis addresses the direct impact of narcissistic traits on the clarity of audit reports, reflecting the potential for narcissistic auditors to create reports that are either excessively complex or overly simplistic, based on their personal tendencies to either display competence or conceal deficiencies:

This hypothesis investigates how the relationship between lead and review audit partners, particularly within the structured framework of Iranian audit practices, may affect how narcissistic traits influence report readability. It explores whether collaborative dynamics might either lessen or intensify the effects of narcissism, depending on how the requirement for dual signatures affects the auditors’ approach to preparing reports.

Recent studies in accounting suggest that the size of a signature can serve as a simple and unobtrusive measure of an individual’s narcissism (Chou et al., 2021; Church et al., 2020; Ham et al., 2017, 2018). These studies indicate that larger signatures are generally associated with higher self-esteem, dominance and self-regard. This correlation supports the use of signature size as a proxy for assessing narcissism.

In our study, focusing on listed Iranian companies, we analyzed the annual reports from 2011 to 2023. Our analysis specifically targeted the mandatory handwritten signatures of two partners in the audit reports section of the financial statements. This method helped ensure the exclusion of automated signatures, which are less reliable for measuring narcissism. Following the methodology of Ham et al. (2017, 2018), we used ImageJ – a custom software program – to enclose each signature within a tight rectangle, ensuring each side of the rectangle touched the most extreme endpoint of the signature. The area of the rectangle (length multiplied by width) was then computed, and to adjust for variations in the length of the auditor’s name, the area was divided by the number of characters in the partner’s name. We calculated the average of these measurements for each partner across all signatures during the sample period to determine the average signature size. Finally, the natural logarithm of this average signature size was used to quantify each partner’s level of narcissism (AUD_NAR).

Readability in this paper is assessed using the FOG index, defined as:

In Iran, the FOG index has been used in various studies to assess the readability of financial statement notes (Hesarzadeh and Rajabalizadeh, 2019, 2020; Salehi et al., 2020) and the Management Discussion and Analysis (MD&A) sections (Hesarzadeh et al., 2020). Our research, however, concentrates on audit reports, as they are a key communication medium between auditors and financial statement users. We specifically compute the FOG index for the audit report opinions, which includes sections on the financial statement opinions, the basis for these opinions, emphasis on specific matters and other explanatory clauses.

The annual reports containing auditor signatures were obtained from CODAL, a comprehensive database that archives all annual reports since a company’s listing on the Tehran Stock Exchange (TSE). This database also provided financial and accounting data, as well as information on signing partner names, audit firms, audit opinions and accounting restatements. Our sample spans from 2011 to 2023, comprising a total of 2,691 firm-year observations. Data related to financial institutions such as banks, insurance companies and investment and holding firms – totaling 960 firm-year observations – were excluded due to their distinctive accounting practices and regulations. In addition, data were removed for firms where information was inaccessible, accounting for 295 firm-year observations (see Table 1).

For the textual analysis of audit reports, we manually downloaded the relevant reports from CODAL for the specified research period and prepared them for analysis. The text was extracted using Python scripts and saved in TXT format, which facilitated the necessary processing for readability assessments. This preparation involved removing special characters from the Persian text while meticulously preserving sentence-ending punctuation to ensure accurate readability calculations. We used a suite of Natural Language Processing tools tailored to the Persian language, such as the Natural Language Toolkit and Hazm, to extract and compute readability scores from these audit reports.

The main analysis is performed using the following regression:

where FOG and AUD_NAR represent the measures of audit report readability and partners’ narcissism, respectively. The selection and definition of the control variables are based on recent research (Hesarzadeh and Rajabalizadeh, 2019; Rajabalizadeh and Oradi, 2022; Rajabalizadeh, 2023; Church et al., 2020; Chou et al., 2021; Mohammadi et al., 2021). These variables include audit partners’ and audit firms’ industry specialization (AUD_SPEC), tenure (AUD_TENURE), audit firms’ mergers (AUD_MERGE), client and audit firms’ locations (DISTANCE), auditor rotation (AUD_CHANGE) and auditor opinions (AUD_OPINIOn). The data set also accounts for the gender of the lead (LEAD_GENDER) and review partners (REVIEW_GENDER), as well as the CPA license tenure for both lead (LEAD_CPA) and review partners (REVIEW_CPA). Additional control variables encompass a broad range of client firm characteristics such as earning quality (AQ), size (SIZE), sales growth (SALESGROWTH), operating cash flows (CFO), loss status (LOSS), financial distress (BANKRUPTCY), the market-to-book ratio (MB), firm lifecycle (AGE), operating cycle (OPERCYCLE), restatement (RESTATE), quick ratio (QUICK) and share concentration (TOPSH). The analysis controls for firm, year and auditor fixed effects (FIRM FE, YEAR FE, AUDITOR FE) [²]. To mitigate the influence of outliers and potential database coding errors, all continuous variables used in the estimations are winsorized at the 1% and 99% percentiles. Standard errors are clustered by firm for all regression test statistics.  Appendix provides the specifications for all regression variables.

Descriptive statistics presented in Table 2 illustrate that the mean readability (FOG) of audit reports is 21.561, which is lower compared to the FOG scores reported by Zeng et al. (2021) and Seebeck and Kaya (2022), at 25.775 and 25.570, respectively. Given that Persian sentences are typically 15% shorter than their English counterparts, these figures are consistent with international standards. Earlier research within Iran, using the FOG index, observed comparable trends across financial statement notes and MD&A sections (Hesarzadeh and Rajabalizadeh, 2019, 2020; Salehi et al., 2020). The average narcissism score of audit partners (AUD_NAR) is 9.659, which aligns with findings on CFO narcissism by Xiang and Song (2021).

The correlation matrix in Table 3 indicates a positive, significant correlation between AUD_NAR and FOG, suggesting that higher levels of narcissism may decrease audit report readability. The low variance inflation factors for independent variables, all below 5, effectively dismiss multicollinearity concerns.

Building on our extensive theoretical discourse, the empirical analysis seeks to definitively test the proposed mechanisms linking auditor narcissism with audit report readability. The findings detailed in Table 4 offer crucial empirical support, accurately reflecting the impacts hypothesized under various theoretical frameworks. The regression analysis reveals a positive correlation between the narcissism of audit partners (AUD_NAR) and the FOG index of audit reports, with a significant coefficient of 0.185 at the 1% level. This relationship suggests that higher narcissism correlates with higher FOG scores, hence lower readability [³].

This observation supports Upper Echelons Theory (Wang et al., 2016), which posits that senior executives’ personal traits, such as narcissism, can shape decision-making to reflect their individual styles – manifesting here as complex and less accessible audit reports. In addition, Institutional Theory (Filatotchev and Nakajima, 2014) suggests that such complexity may stem from institutional pressures within the Iranian audit environment, where auditors may use complexity either to navigate or resist normative expectations. Moreover, the differentiated impact of narcissism between lead and review partners (coefficients of 0.160 and 0.140 at the 1% and 5% levels, respectively) highlights how roles within the audit team influence report outcomes. The focus on different roles within the team ties directly back to the traits described by Upper Echelons Theory and the dynamics explored in Agency Theory. Agency Theory (Adams, 1994) further posits that the readability of audit reports can be influenced by the dynamics between auditors (agents) and their clients (principals). Here, the bargaining power of clients can compel auditors to adjust readability to better serve client interests, potentially varying based on the auditor’s narcissistic traits and their approach to client relationship management.

Using the methodology introduced by Mitton (2024), the economic significance of the relationship between AUD_NAR and FOG was assessed. This result is calculated using the $ESIQR$ = $|b (p75−p25)||Sy|$ where b is the regression coefficient for partners narcissism (AUD_NAR), p₇₅ and p₂₅ are the 75th and 25th percentiles of AUD_NAR and s_y is the standard deviation of FOG [⁴]. The analysis demonstrates that an increment in partner’s narcissism across its interquartile range corresponds to a 0.081 increase in the FOG index when expressed as a percentage of its standard deviation. This metric offers a standardized measure of the effect relative to the natural dispersion of the outcome, highlighting the substantial influence of partner’s narcissism on the clarity and readability of audit reports. The interpretation of this metric underscores that higher levels of partner’s narcissism are associated with less readable audit reports, emphasizing the considerable impact of auditor personality traits on the transparency of financial reporting. This finding underscores the critical role these traits play in the readability and transparency of financial documents, suggesting that increased narcissism may lead to more opaque and complex audit reports [⁵].

In this section, we critically examine how specific contextual variables – auditing complexity, auditor specialization and the type of auditor opinion – affect the interplay between auditor narcissism and the readability of audit reports. These aspects are pivotal as they represent key facets of the auditing environment that can either amplify or mitigate the inherent traits of auditor narcissism, influencing their professional conduct and the clarity of financial reporting (Gaynor et al., 2016; Habib, 2013; El-Masry and Hansen, 2008; Beattie and Fearnley, 1995). Understanding these relationships is crucial for enhancing audit quality and tailoring auditor training and oversight to match the demands of varied audit contexts.

Complex audit engagements, measured by the ratio of inventories and accounts receivable to total assets (INVREC), generally require more detailed explanations and nuanced financial reporting. Our findings, as documented in Table 5, reveal that when narcissistic auditors handle such complex engagements, the readability of audit reports deteriorates further, as indicated by the positive interaction term (INVREC × AUD_NAR). This suggests that the assertiveness and overconfidence typically associated with narcissism may lead to overly complex or less precise explanations in already complex scenarios, thus increasing the FOG index.

For auditor specialization, defined by an indicator variable (AUD_SPEC) that marks auditors with the highest number of clients in a client’s industry for a given fiscal year, we also observe a positive interaction with auditor narcissism (AUD_SPEC × AUD_NAR). This result implies that in specialized firms, where one might expect a deeper industry knowledge to aid clarity, the narcissistic traits instead lead to a decrease in report readability. This could be because narcissistic auditors, despite their industry knowledge, might prefer showcasing their expertise through complex language, adversely impacting the clarity perceived by less specialized readers.

The analysis further extends to the impact of narcissism in settings where auditors issue different types of opinions. Positive coefficients in scenarios involving qualified opinions would suggest that narcissistic auditors produce even less readable reports under conditions where clarity and precision are paramount. This might reflect a misalignment between the auditor’s self-perceived competence and the actual clarity of their reporting, particularly in contentious or scrutinized audit situations.

The observed increase in the FOG index in contexts involving complex audits, specialized auditors and qualified opinions can be theoretically linked to the personality traits associated with narcissism. Narcissistic auditors might be inclined to use more complex language to assert dominance or showcase expertise, particularly in environments that are inherently complex or where they hold a lot of industry-specific knowledge (Kaszak et al., 2024). Moreover, the demanding nature of specialized audits and the critical scrutiny in qualified opinions could exacerbate the auditors’ tendencies toward verbosity or technical jargon, which, while potentially impressing some peers, does not necessarily translate to clear and accessible reporting. Thus, the presence of narcissistic traits might amplify the inherent challenges of these audit contexts, leading to poorer readability as auditors opt to emphasize complexity over clarity.

To enhance the robustness of our initial findings which used the FOG index, we expanded our analysis to include the SMOG and ARI readability indices. These indices are well-recognized for assessing text complexity in accounting (Hesarzadeh et al., 2020). Our results, though untabled, consistently indicate that higher levels of auditor narcissism lead to less readable audit reports. This finding corroborates our initial results, demonstrating a clear negative impact of auditor narcissism on report clarity. In addition, we analyzed auditor narcissism using different metrics, initially focusing on the size of audit partners’ signatures and then categorizing these sizes into quantiles. Higher quantiles suggest greater narcissism. Our findings confirm a significant correlation between higher narcissism levels and increased FOG index scores, indicating less readable reports.

Concerns about endogeneity in auditor-client relationships present substantial challenges, as the pairing of audit partners with clients involves complex factors, including partner skills and client needs (Lennox and Wu, 2018). To navigate these complexities, our study uses firm and audit firm fixed effects in the main regression analysis. However, the relationship between partners’ narcissism and the readability of audit reports may be influenced by unobservable omitted variable bias. To address this issue, we use a quasi-experimental switching regression difference-in-differences (DiD) methodology, as reported in column 1 of Table 6. Following the approaches outlined by Garcia-Blandon et al. (2019), Hardies et al. (2015) and Ittonen et al. (2013), we investigate the changes in audit reports’ readability following the appointment of audit partners with varying degrees of narcissism. Our primary narcissism measure, AUD_NAR, is divided into high-narcissism (coded as 1) and low-narcissism (coded as 0) groups based on the median score.

The DiD analysis contrasts the readability of audit reports before and after the switch from non-narcissistic to narcissistic audit partners, compared with a control group that experienced consistent non-narcissistic switches. We define a dummy variable, NNTN, which is set to 1 for switches to narcissistic partners and 0 for consistent non-narcissistic switches. Our sample now comprises 491 non-narcissistic and 143 narcissistic partner switches, resulting in a total of 634 firm-year observations. The variable NNTN captures the incremental impact of switching from a non-narcissistic to a narcissistic audit partner.

Adhering to the methods of Hardies et al. (2016), we reestimate the regression models presented in Table 4 with NNTN as the primary independent variable. The dependent and control variables measure the changes (Δ) from year t−1 to year t, with each auditee acting as its own control to enhance causal inference by accounting for firm-specific, time-invariant characteristics. The estimated coefficient for the change in narcissism score, Δ AUD_NAR (0.561), is statistically significant at the 5% level in our primary model. This finding suggests that audit reports become significantly less readable when high-narcissistic partners are assigned, compared to their low-narcissistic counterparts [⁶].

We also explored the potential for self-selection bias regarding the roles that audit partners assume within the audit process. Given that leadership roles typically require traits such as assertiveness and risk-taking – characteristics frequently associated with higher levels of narcissism – it raises the question of whether narcissistic tendencies influence an auditor’s preference for leading rather than reviewing. Conversely, review roles, which demand precision and a keen eye for detail, might attract those with lower levels of narcissism. To examine this dynamic, we used logistic regression to assess how levels of narcissism impact the likelihood of auditors assuming lead versus review roles. Our analysis, presented in column 2, reveals a coefficient of 0.238 for narcissism scores, statistically significant at the 5% level. This indicates a positive correlation between higher levels of narcissism and the propensity to assume leadership roles, supporting the hypothesis that auditors with more narcissistic tendencies are more inclined to occupy leadership positions within audit teams. This finding underscores the complex interplay between personality traits and professional roles.

To mitigate the potential selection bias inherent in observational studies, we applied the entropy balancing approach to refine our analysis of the impact of auditor narcissism on audit report complexity. Proposed by Hainmueller (2012), entropy balancing is a reweighting technique that precisely adjusts covariates to achieve a predetermined balance between treatment and control groups. This method is particularly effective in ensuring that the distributions of covariates are comparable across groups, thereby minimizing the influence of confounding variables.

We used the entropy balancing method to adjust for various covariates, including auditor characteristics and firm-specific factors, which could influence the complexity of audit reports as measured by the FOG index. Our goal was to isolate the effect of auditor narcissism from these potential confounders. This approach involved adjusting weights for each observation in the treatment group (high narcissism auditors based on the median) to match specific moment conditions (mean, variance and skewness) in the control group (low narcissism auditors).

Panel A of Table 7 presents the first, second and third moments of key covariates before entropy balancing. It shows significant differences between the treatment and control groups across various metrics, indicating initial imbalance. Following the application of entropy balancing, Panel B confirms that the moment conditions for the covariates are closely aligned between the two groups post-balancing, reflecting successful adjustment. For example, the means and variances for key variables such as auditor tenure, auditor mergers and financial metrics now closely mirror each other between the treatment and control groups, validating the effectiveness of entropy balancing in creating comparable groups for subsequent analysis.

Panel C outlines the regression results post-entropy balancing. The models incorporate the balanced weights and further control for firm, year and auditor fixed effects to ensure robustness. The results show a significant positive association between auditor narcissism and the complexity of audit reports. The coefficient for overall partner narcissism (AUD_NAR) is statistically significant at the 1% level, suggesting that higher levels of narcissism among audit partners are associated with less readable audit reports. The results also differentiate between lead and review partners, indicating that narcissism in lead partners has a more pronounced effect on report complexity. These findings address concerns regarding the robustness of our results and reinforce the validity of our conclusions that audit report complexity can indeed be attributed to the narcissistic traits of auditors, rather than other extraneous factors.

Our research examines how auditor narcissism interacts with the economic significance of clients to influence audit report readability. Addressing the conflicts between auditors’ financial interests, professional independence and the risk of client loss as discussed by Mautz and Sharaf (1961) and DeAngelo (1981), we analyze the impact of auditors’ stakes in client relationships on audit quality. This is particularly relevant in markets like Iran, where protections are less stringent. In our regression analysis, we use a methodology adapted from Chen et al. (2010) to measure client importance and its interaction with auditor narcissism. The results, detailed in Table 8, indicate that while significant clients generally receive clearer reports, the introduction of auditor narcissism leads to increased report complexity. The coefficients for the interaction terms with client importance (CI_A and CI_S) are positive and significant [CI_A: 1.012** (t-statistic = 1.970), CI_S: 0.529* (t-statistic = 1.870)], suggesting that narcissistic auditors might prioritize asserting their expertise and maintaining professional standing over simplifying reports for better client comprehension, aligning with findings by Maccoby (2003), Ma and Jaeger (2005) and Hüffmeier et al. (2014).

In Iran, audit restatements are notably prevalent, affecting 60–70% of audit opinions and 75% of financial reports from TSE-listed firms. This high incidence raises substantial concerns about investor trust and market stability. Our study exploits Iranian disclosure norms that mandate the naming of lead and review partners in audit reports. We investigate audit partner quality (QLead and QReview) through the lens of audit failures, defined as restatements following unqualified opinions (Wang et al., 2015). Contrary to initial expectations, our findings (untabled) reveal no significant correlation between the readability of audit reports and audit partner performance. This suggests that less readable reports, which are typically more detailed and complex, do not necessarily indicate lower audit quality. Instead, they may reflect the auditor’s diligence, potentially associated with narcissistic traits.

The IACPA grades audit firms from A to D based on audit quality criteria, with rankings reflecting various evaluative measures including partner and staff evaluation, firm structure, diversity in service provision and overall audit quality (MohammadRezaei et al., 2018). Our study assessed the correlation between auditor narcissism and employment in higher-ranked audit firms, assigning values of 4 to A-ranked firms down to 1 for D-ranked ones, with the majority of firms in our sample being A-ranked. Our analysis (untabled) indicated a significant positive link between firm ranking and auditor narcissism in relation to audit report readability. This suggests that auditors with traits like assertiveness or self-confidence, which are advantageous in high-pressure, prestigious environments, may be more prevalent in top-tier firms. This finding points to a potential attraction of narcissistic auditors to prestigious firms for recognition, impacting audit report readability.

Recent advancements in behavioral accounting have emphasized the role of individual traits in influencing corporate outcomes (Anessi-Pessina and Sicilia, 2020; Liem and Hien, 2020). The application of Upper Echelons Theory in accounting research illustrates how the personal characteristics of top executives can significantly impact corporate decision-making and financial reporting practices (Hiebl, 2014; Bassyouny et al., 2020). Similarly, by relaxing the assumption of auditor rationality, traits such as narcissism become pivotal in understanding the auditing process and its outcomes (D’Souza et al., 2019). While our research has focused on auditor narcissism, the potential influence of client traits, particularly the narcissism of CEOs [⁷], on audit negotiations and outcomes cannot be ignored. This section introduces a new variable – CEO narcissism, CEO_NAR – calculated based on the size of CEOs’ signatures, a proxy previously validated in the literature as indicative of narcissistic traits (Ham et al., 2017). We explore how the interplay between auditor and CEO narcissism might jointly influence audit outcomes.

To investigate the combined effect of auditor and CEO narcissism on audit outcomes, we introduce an interaction term between auditor narcissism and CEO narcissism (AUD_NAR × CEO_NAR) in our regression models. This approach allows us to examine whether the negotiation dynamics between narcissistic auditors and executives exacerbate or mitigate issues in audit quality and financial reporting clarity. We hypothesize that the interaction between auditor and CEO narcissism may lead to heightened conflict during audit negotiations, particularly in scenarios involving significant audit adjustments (Church et al., 2020; Chou et al., 2021). Our regression analysis reveals a positive and statistically significant coefficient at the 10% level for the interaction term, suggesting reduced clarity in audit reports when both auditor and CEO narcissism levels are high, as further detailed in Table 9. CEO narcissism (CEO_NAR) presents a positively significant coefficient at the 10%, indicating a stronger, significant influence on audit report readability. Unexpectedly, the individual effect of auditor narcissism (AUD_NAR) becomes nonsignificant when controlling for CEO narcissism. This indicates that the predominant influence of CEO narcissism in shaping audit outcomes is a critical aspect that merits further exploration.

This observation compels us to explore further the impact of CEOs’ narcissistic traits on audit processes. In the realm of financial reporting and audit processes, the narcissistic traits of CEOs – characterized by grandiosity, entitlement, dominance and a sense of superiority – are particularly consequential (Amernic and Craig, 2010; Zhu and Chen, 2015). Such traits may prompt CEOs to pursue aggressive corporate strategies that directly influence audit processes and enable them to assert their desired outcomes (Judd et al., 2017). Institutional Theory supports this notion, proposing that while individuals in top executive roles generally conform to the norms of their environments (Liang et al., 2007), they also tailor these environments to suit their personal traits and agendas (Hambrick et al., 1993; Alvarez and Svejenova, 2005). Furthermore, Agency Theory suggests that conflicts of interest between principals (shareholders) and agents (CEOs) intensify when agents operate with significant discretion (Dagnino et al., 2019), a situation that may be exacerbated by their narcissistic traits (Kashmiri et al., 2017). This exacerbation of agency problems can pose substantial governance challenges, particularly affecting financial transparency and audit integrity (Head, 2012; Salehi et al., 2023). Narcissistic CEOs, likely to resist scrutiny and challenge audit adjustments, paired with similarly narcissistic auditors who may assert their judgments more aggressively (Hou et al., 2024), could lead to more contentious and less effective audit processes. This might result in audit reports that are less clear and potentially more biased.

The context of Iran provides a unique setting to examine the influence of CEO traits on audit outcomes. In environments where regulatory frameworks may permit significant executive discretion (Rajabalizadeh, 2023), the personal traits of CEOs, such as narcissism, could have a more pronounced impact on corporate governance practices. Factors such as the bargaining power of narcissistic CEOs, auditor tenure and the prevalence of opinion shopping are crucial in this regard (Tarighi et al., 2022; MohammadRezaei et al., 2016). Narcissistic CEOs may strongly resist audit adjustments (Sedikides and Campbell, 2017) and engage in more aggressive negotiations during audit report preparation, influenced by their bargaining power (Bhaskar et al., 2023). Moreover, prolonged auditor tenures might suggest excessively close relationships between auditors and executive teams, which, when combined with CEO narcissism, subtly yet powerfully influence audit negotiations (Liu et al., 2021; Wang and Tuttle, 2009). The practice of opinion shopping, where executives seek auditors who may align with their views, further illustrates how CEO narcissism might manifest, affecting audit quality and financial reporting clarity (Berglund et al., 2022; Seifzadeh et al., 2021).

In conclusion, our study emphasizes the significant influence of CEO narcissism on audit outcomes, more so than auditor narcissism. This predominance of CEO narcissism, especially in contexts like Iran where executive discretion is extensive, underscores the need for robust governance mechanisms to mitigate the potential adverse effects of executive personality traits on audit quality. Our findings enrich the literature by detailing the complex interplay between executive traits and corporate governance, offering insights that could inform both policy and practice in the realms of financial reporting and audit oversight.

Since December 2016, specific criteria have been established for communicating key audit matters to corporate governance stakeholders (management) and disclosing them in the auditor’s report (Kitiwong and Sarapaivanich, 2020; Nguyen and Kend, 2021; Rautiainen et al., 2021). In Iran, International Auditing Standard 701 was adopted in 2021, with the revisions becoming effective for auditing financial statements for periods ending on or after March 21, 2022 (Ghaemi et al., 2024). This standard marks a significant advancement in enhancing transparency and accountability in audit reporting by requiring auditors to explicitly address those matters that, in their professional judgment, were most significant in the audit of the financial statements for the current period.

To assess the potential moderating effect of this new standard on the relationship between auditor narcissism and the readability of audit reports, we divided our sample into pre- and post-reform periods. The pre-reform period encompasses the years 2011–2021, while the post-reform period includes the years 2022 and 2023. We introduced a dummy variable, POST_REFORM, to denote these periods, with the post-reform period coded as “1” and the pre-reform as “0.” In addition, we generated an interaction term, POST_REFORM × AUD_NAR, to represent the interaction between auditor narcissism and the implementation of the reform. Our untabled regression analysis revealed that the coefficient for auditor narcissism was positive (0.216) and significant (p = 0.004), indicating that higher levels of narcissism are associated with less readable audit reports, as measured by higher FOG index scores. The coefficient for the post-reform period was not significant (p = 0.386), indicating that the introduction of the standard alone did not have a discernible impact on the readability of audit reports. The interaction term was negative (−0.203) but not significant (p = 0.290), suggesting that the reform does not significantly alter the effect of auditor narcissism on report readability.

This analysis indicates that the implementation of International Auditing Standard 701 in Iran, which mandates the disclosure of key audit matters, does not significantly moderate the impact of auditor narcissism on the readability of audit reports. These findings suggest that despite the regulatory intent to enhance transparency and clarity, the personal characteristics of auditors, such as narcissism, continue to play a dominant role in shaping audit report outcomes [⁸]. This may imply that the standard, while theoretically beneficial, has not been effective in practice in mitigating the influence of individual auditor traits on report readability within the Iranian auditing environment.

This study presents compelling evidence that the narcissism of audit partners, particularly within the Iranian context, significantly influences the readability of audit reports listed on the TSE. The use of signature size as a proxy for narcissism has elucidated the impact of auditor personality traits on the clarity of audit reports. Notably, narcissism, especially prominent in lead audit partners, is correlated with lower readability of audit reports, as indicated by higher FOG index scores. This trend highlights the complexity introduced into audit reports by narcissistic auditors, who likely embed the reports with detailed and complex narratives to assert their dominance and expertise.

The implications of this research are profound, extending beyond academic circles into practical applications for audit firms and regulatory bodies. By demonstrating that narcissism adversely affects audit report readability, the study underscores the need for meticulous selection and training of audit partners. Audit firms may need to incorporate personality assessments into their recruitment and team assembly processes to ensure a balance that promotes assertive independence and clear communication in audit reporting. In addition, the study illuminates the broader impact of cultural and regulatory environments on audit practices. Despite the general discouragement of self-promotion within Iran’s collectivist culture, the structure of its audit practices may inadvertently foster narcissistic traits among auditors. This observation suggests that both cultural nuances and professional practices should be considered in the development of standards and training programs for auditors in Iran and potentially in other similar regulatory contexts.

Our study has its limitations, which necessitate a cautious interpretation of the results. We used signature sizes as a proxy for narcissism, a creative yet indirect method that may not fully capture the complexity of this personality trait. Despite efforts to control for variables such as company size and risk, challenges such as data availability and the potential variability of narcissism over time remain. Moreover, focusing on Iranian audit firms may limit the generalizability of our findings due to geographical specificity, though the insights could be relevant to both developed and emerging markets. Future research could empirically evaluate how readability affects stakeholder perceptions, conduct behavioral experiments on readability and trust and explore additional personality traits like conscientiousness. Further investigation into how readability influences different market participants and the regulatory impacts on audit practices could yield deeper insights for both academia and practice.

The authors extend the gratitude to Professor David Hay, Professor Reza Hesarzadeh and Associate Professor Tony Kang for their invaluable feedback and insights, which have greatly enriched this study. Special thanks are also due to CPA Hossein Fateh Goush, whose expertise on Iran's audit market provided essential context to their analysis. Lastly, the authors acknowledge the constructive critiques and suggestions received from participants at the Parallel Session of the 45th Annual Congress of the European Accounting Association. The authors, however, retain responsibility for the final interpretations.

Declarations of interest: None.