Executives ’ narcissism and decision making: reviewing 20 years of accounting literature

Purpose – Thispaperaimstoprovideasystematicreviewofliteraturepertainingtohowexecutivebehavioral characteristics relate to financial reporting decisions. Design/methodology/approach – The authors review 44 papers published between 2001 and 2021 in top journals that are nested in leading business, economic and accounting journals. Findings – Through the systematic review, the authors provide a framework for the emergence of narcissism and how it relates to decisionmaking and hence, firm performance. Additionally, this paperidentifies different measures of measuring narcissism with their pros and cons and suggest that different measures lead to different outcomes in prior literature. Originality/value – Thestudycontributestoagrowingstreamofresearchonexecutives ’ attributesinfluence on decision making. The authors recommend that future research may focus more on the chief financial officer (CFO) role as the majority of literature in CEO based. Additionally, the authors suggest that different settings may moderate the outcomes, and the authors propose that future research may be conducted to show how the regulatory environment affects or moderates narcissism effect.


Introduction
This paper aims to provide a systematic review of executive narcissism in accounting literature through examining a number of articles published in leading journals in the period 2000-2021, it seeks to provide a better understanding of the importance of behavioral traits of top executives and how it interferes with their decision making. To this aim, first, we discuss the method used and how the systematic review is conducted. Second, the paper provides a framework for valuable theories used in literature to explain decision making and focus on upper echelon theory that is the base for behavioral research in accounting and finance literature. A thorough review of prior literature that encompasses how the influence of executives' personal traits manifested in upper echelon theory in general, and the psychological traits in particular, is made to define narcissism concept. In this context, we acknowledge discrepancies in prior results and that they might be caused by using different measures for narcissism. Hence, to provide a collective reference for future research, we examine measures for narcissism, which have been used in literature and shed light on the pros and cons of each measure. Then, we try to theoretically analyze and clarify the widespread misuse of the concepts of narcissism and overconfidence, as they are occasionally used interchangeably in academic literature. Third, later in the paper, we focus on the effect of executives' narcissism on financial reporting, examine prior literature that links managerial discretion, firm performance and misreporting, and what other factors that could interfere to moderate the association. Finally, we conclude by providing some insight about directions for future research and bring to researchers' attention new angles to tackle narcissism that might be of interest.

Research method and strategy
To postulate an overview about behavioral decision making in general, and narcissism in particular, this study uses a systematic review to provide a conceptual framework about different yet related topics nested in behavioral accounting research.
Initial research encompasses papers published in Elsevier database and Google Scholar; both are major sources for academic literature. The initial search aims at finding articles that deal with how behavioral/personal characteristics at top managerial levels affect their decision making, hence the scope of our interest goes beyond accounting and management literature to included articles from psychology journals that explain specific concepts such as narcissism, overconfidence, hubris. We searched in the database within the title, abstract and provided key words for articles containing the terms "overconfidence", "narcissism" "narcissistic", "executives' characteristics" and "decision making", both as standalone and/or combined. This leads to a total of 78 papers, 8 of which are found in journal of psychology which were the reference for the basic concepts and definitions in our theoretical framework. We use data from the collected papers to provide understanding of major concepts but then exclude books, seminar series, working papers and thesis to focus on articles published in journals. This yields 60 papers as shown in table. These papers are within the stream of accounting, finance and management. We then restricted our selection criteria to include papers that are published under Business, Management and Accounting, Accounting and Finance, and Accounting and Economics section to ensure that articles are related to these fields. Moreover, we narrowed the search to include papers that affect financial reporting. This leads to 44 papers published between 2001 and 2021 in top journals.
3. Background and theoretical framework 3.1 Financial reporting decisions Financial reporting decisions are vital performance outcomes that help in the assessment of the organization by stakeholders and the capital market. Financial reports are the key tool used by many to determine the quality of the information provided about any organization and hence, assess their performance. Therefore, International Accounting Standard Board (IASB) provides a set of standards International Financial Reporting Standard(s) (IFRS) [1] with the main objective of providing a global framework for how public companies prepare and disclose their financial statements.
Financial reporting standards allow for managerial discretion in some cases when it is needed to better reflect the underlying performance and economic position. This allowed discretion may lead to what is called "opportunistic behavior" by managers that in return leads to earnings management (Barth et al., 2008). Earnings management is when managers use their judgment in structuring transactions to alter the numbers on financial reports for the purpose of misleading users about the underlying economic status of the organization. Moreover, managers may use judgment to manipulate numbers to influence some contractual outcomes. Hence, the lower the opportunistic behavior exists, the higher the quality of earnings and financial reporting.

Executives' narcissism and decision making
The main aim of business entity is to maximize profits earned. To achieve this aim, many businesses manipulate accounting numbers to achieve their targets. Hence, earnings management appears to be more of a business culture practiced by many around the world. This culture seems to prevail not only in countries where there is no or weak organized, regulated system, but also in developed countries with strong organized businesses such as the United States. For example, Enron, Wreck and WorldCom are some of the top scandals that are widely known.
Management intention to use accounting estimate and numbers to report in a way that best serves their interest is explained by some theories. Many studies have been conducted to explore what factors influence the quality of earnings. Among those factors that are found to be of influence; profitability, size, debt covenants, financial leverage and ownership structure (Mokhtar, 2017). These studies employed two of the most widely known theories in accounting literature that explain managerial behavior, namely, agency theory and positive accounting theory.
3.1.1 Positive accounting theory. Watts and Zimmerman (1979) were the first to explore factors that are likely to affect a firm's cash flow and are influenced by managerial employment of accounting standards. Factors such as regulation, taxes, managerial compensation, political costs and bookkeeping costs are combined into a model. This theory became known as positive accounting theory in 1986 (Nasution et al., 2020).
Positive accounting theory (PAT) highlights that a company's characteristics direct how the company is organized. Different characteristics among companies lead to different management styles. Hence, the best way to manage a company would depend on its institutional environment, the technology adopted and by the competition level in the market. These different conditions generate a set of different investment opportunities available.
Due to the flexibility that enables mangers to choose from the available accounting sets, a possibility of opportunistic behavior is raised. This possibility generates the important assumption of PAT, which states that managers are believed to be rational persons, and they will, if they have the opportunity, choose the accounting policies that best serve their goals. PAT assumes that managers will act only on their personal interest to maximize profits (Wiratama and Asri, 2020).
3.1.2 Agency theory. According to Meyer and Rowan (1977), agency relationship is characterized by the conflict of interest, known as the "agency conflict". The conflict happens between the interests of investors which are the owners/principal and the mangers which in this case are the agent. Owners of capital mainly aim to efficiently use the funds with the lowest risk possible, while managers (agents) aim to maximize profits regardless the risk. Different interests lead to the conflict, managers as agents should act for the owners' interest, but because the risk level that managers are willing to accept to achieve goals differs from the one tolerated by owners, conflict appears and hence comes the agency problem.
In financial management, the agency theory discusses the relationship between the agent and the principal (Jensen and Meckling, 1976), regarding the separation between ownership and management. The owner of capital (principal) should be able to change the agent that does not fulfill his purpose and increase the owner's prosperity. Since the agent in reality has more experience and better understanding of the company's operation, that creates a risk that the agent will use this advantage to act in their own interest through financial reporting manipulation.
Though researchers went back and forth on reasons behind this opportunistic behavior in accounting choices using different theories that are relevant to elements of the organization or the surrounding environment, a more recent panel of efforts tries to reach out to other disciplines that can affect financial reporting. In today's business world, where opportunistic behavior is apparent still, researchers attempt to broaden the theoretical perspective to the empirical testing of this opportunistic behavior in accounting choices to add more explanatory power to the existing knowledge. This line of research is based on the pioneer studies of Hambrick and Mason (1984) and Hambrick (2007) who suggest that individuals in any organization have an important influence and that different individuals will react differently because of individual idiosyncrasies. This is the essence of "Upper Echelons Theory".
3.1.3 Upper echelon theory. Prior research in accounting and finance postulates that the rational behavior of management is predominant as portrayed in neoclassical and agency perspectives (Pl€ ockinger et al., 2016). These perspectives give no or little room for personal idiosyncrasies, discretion, human error and decision misconduct. Nevertheless, many studies in psychology and socioeconomy that focus on judgment and decision making have showed evidence that different individual characteristics do affect decision making outcomes (Stumpf and Dunbar, 1991;Foster et al., 2009). This latter view was adopted by Hambrick and Mason (1984) to postulate that the different characteristics of individuals play a role in decision making, in other words, top management individual characteristics affect their strategic choices and accordingly, firm performance. This view is what is known as "upper echelon theory".
The roots of upper echelon theory lie within the behavioral theory of the firm which says that choices at top management are not always a result of rational motives, but they are influenced also by the limitations of mangers human nature. Hence, behavioral factors such as conflicting goals, aspiration levels, bounded rationality, etc. are considered in strategic decisions made by top executives that influence firm performance. Hambrick and Mason (1984) is the first study that combines the roots of organizational outcomes with theories on the effect of cognitive bases in an organization. Upper echelon theory is based on the notion of bounded rationality, that is, in situations where individuals should make a strategic choice, such as financial reporting decisions, they are confronted with complex tasks that they are unable to comprehend and thoroughly process (Pl€ ockinger et al., 2016). Hence individuals in that position tend to curb the details for the purpose of simplification. This simplification reflects upon one's personal characteristics, in other words, this can be thought of as a lens through which an individual alters the real world's situation into his own perspective. As this process is structured by an individual's values, cognitive base and other tangible personal characteristics, upper echelon theory suggests that the resulting decisions and outcomes are a reflection of one's personality, experience and other idiosyncrasies (Hambrick and Mason, 1984;Hambrick, 2007;Finkelstein et al., 2009).
Since psychological factors are usually hard to measure, it is recommended to measure personality dimensions with demographic proxies in order to achieve additional reliability and validity (Hambrick and Mason, 1984). One example of these demographic proxies is executive age, in their study, Hambrick and Mason argue that age reflects risk taking appetite and suggest that younger executives are more likely to pursue risky strategies such as financial leverage. Hence, upper echelon theory suggests that individual characteristics may be used to predict firm performance. Other external and internal influences affect strategic choices of upper echelon theory, Hambrick and Mason (1984) suggest that managerial characteristics directly affect organizational performance, or indirectly when mediated by other organizational outcomes as shown in Figure 1.
Ever since the original theory was first introduced in 1984, an extensive number of empirical research analyzed the relationship between executives' characteristics and strategic choices. The basic proposition of the theory is supported by empirical findings (Dejong and Ling, 2013;Pl€ ockinger et al., 2016;Utama et al., 2020). In financial reporting, empirical findings support upper echelon theory as they provide evidence that financial reporting information reflects upon top executives' characteristics (Pl€ ockinger et al., 2016).
Top executive's idiosyncrasies mirrored in their financial reporting decisions differs than other decisions making issues in literature due to the existence of many regulations, auditing and other constraints that curb top executives' discretion (Ge et al., 2011). Nevertheless, upper Executives' narcissism and decision making echelon theory faced some challenges stemming from opposing perspectives and theories which lead to introducing refinements to the original theory as explained next. Hambrick (2007) introduced two important moderators, namely, managerial discretion and executive job demands that are believed to empower the theory's relationship with strategic choices. Managerial discretion was first introduced by Hambrick and Finkelstein (1987) as an attempt to reconcile the opposing views relevant to top executives' effect on organizational outcome. One view is that executives do influence organizational outcomes, while the other view states that executives have little if no influence as they are outweighed by the external forces and norms. The introduced moderator reconciles these views as both are valid but depend on how much discretion exists. Discretion can be defined as the extent of possible latitude and the absence of constraints; it exists also when there are multiple alternatives that lead to ambiguity. The inference of managerial discretion moderator is that upper echelon theory provides valid interpretation of organizational outcomes relevant to the degree to which discretion exists. Several studies show that managerial discretion is a crucial moderator of upper echelon theory (Chatterjee and Hambrick, 2007;Ham et al., 2017;Janahi et al., 2021;Donatella and Tagesson, 2021).
The second moderator to the upper echelon theory that was introduced by Hambrick (2007) is executive job demands. Job demands refers to the level of difficulty, challenges and needs that face executives. These challenges may be performance related (e.g. meeting shareholder's needs, achieving targets), task related (e.g. surrounding conditions) and/or executives' personal desires (e.g. promotion, compensation and outperforming peers). The idea is similar to managerial discretion; job demands play a moderator role on the extent to which individual characteristics affect decision outcomes. Executives who are under pressure to achieve specific needs, targets or who would face hard consequences for their decisions are more likely to depend on their cognitive thinking, previous experience to make decisions than to do a thorough analysis of available alternatives to reach rational, objective decision outcomes. Hence, as with managerial discretion, executive job demands when introduced as a moderator to upper echelon theory, gives better prediction to the reflection of individual cholesterics on decision outcomes.
Empirical research on upper echelon theory began shortly after its introduction 1984. The theory drew much attention in the fields of business and economics. However, most of the work inclines towards studying the relationship between corporate strategic decisions and  (Nielsen, 2010). Upper echelon theory predictions at the first glance seem to be more linked to strategic decisions in settings where no or minimal regulations exist than in the more regulated setting of financial reporting. Hence, it was not until recently that we find studies on upper echelon theory applications in the field of finance and accounting (Nielsen, 2010). In other words, managerial style and other idiosyncrasies are believed to be constrained by accounting standards. However, literature shows that even with the presence of regulation, managerial influence in accounting choices exists. This influence may appear in financial reporting either by engaging in aggressive accounting policies [2], or through an opportunistic behavior to move earning upwards or downwards depending on which policy better serve management's interest [3]. Accounting choices are a part of the strategic decisions made by executives; hence they are the medium through which the management interacts with the outer market. Literature has shown that management individual characteristics are reflected in reporting outcomes as will be discussed in the next section.
Researchers (e.g. Fields et al., 2001) suggest that upper echelon theory contributes to a comprehensive theory of accounting choices; moreover they regret the lack of attention to individual management characteristics in the 1990s and argue that this leads to a slow progress of empirical research on the determinants and outcomes of accounting choices. By focusing on individuals' personality and attributes of those who are involved in decision making, upper echelon theory adds a new perspective by paying more attention to individuals' idiosyncrasies as an important input to cognitive processes.

Influence of executives' personal characteristics
The common belief in accounting literature before the 90s assumed that managers are perfect substitutes, which is based on the neoclassical view of the firm, in other words, managers' decisions are a result of economic incentives. Latter literature recognizes that individual attributes of top management/executives have an influence on the decisions outcome. This assumption is supported by upper echelon theory proposition of how individual's personal characteristics affect their judgment and decision-making process (Hambrick and Mason, 1984;Hambrick, 2007). Pl€ ockinger et al. (2016) comprehensive review on literature of individual characteristics influence on financial reporting provides that influence could be of many forms; a general influence of top management without getting into the details of the individual personal characteristics, an influence of the demographic characteristics on reporting decisions, and psychological or behavioral attributes relation with decision outcomes. Table 1 is adopted from that review paper and is amended to include more recent studies that examine similar relations.
The majority of research evidence supports the postulate that upper echelon theory characteristics influence financial reporting. Individual characteristics influence different aspect of financial reporting, which will be discussed more thoroughly in the next section. For instance, the majority of papers show these characteristics affect earnings management and earnings quality . The next in order is how individual characteristics lead to financial irregularities like misstatements, fraud and restatements (Feng et al., 2011;Olsen et al., 2014). A similar line of research deals with their influence on the extent and the frequency of disclosure (Hribar and Yang, 2016). Also, some papers focus on accounting conservatism and how individual style can affect it (Ham et al., 2014). Few papers examine individual characteristics influence on other accounting related choices such as, asset impairments, tax and timeliness of audit reports (Dyreng et al., 2010).
Top executive characteristics in general, without focusing on a specific characteristic, are found to be of influence on financial accounting choices. Ge et al. (2011) find that chief financial officer (CFO) fixed effects have a significant effect on accounting decisions which is reflected more strongly in settings where higher managerial discretion and/or higher job demands     exists. Executives manage earnings using different methods to manipulate earnings whether through real earnings management activities or using accrual earnings (Roychowdhury, 2006;Dejong and Ling, 2013). This is more adherent when decision makers face difficult performance targets, or when they are under pressure to meet analysts' forecasts. Other research studies executives' demographic characteristics such as, age, gender, tenure, education and experience. Evidence in prior literature suggests that in terms of gender, female executives in general are found to be more conservative when it comes to adopting accounting policies. Moreover, female executives are found to engage less in earnings management, and are less risk tolerant especially in high litigation settings (Liu et al., 2016;Janahi et al., 2021). Nevertheless, there is no common agreement on such effect as some other studies did not find such relationships Davis et al., 2015). Another attribute is age, where younger executives are found to disclose more than elder peers, while older executives engage less in fraudulent accounting. Executives who have higher tenure are less involved in restatements, less fraud and provide more timely reports (Schrand and Zechman, 2012;Donatella and Tagesson, 2021). Studies that examine other demographic characteristics, such as education and prior experience, suggest that executives that received more sophisticated education (e.g. MBA holders) are associated with conservatism and provide better quality disclosure. Evidence also suggests that chief executive officers (CEOs) with accounting and finance experience provide more timely reports and engage less in earnings management (Donatella and Tagesson, 2021).
Earlier research on financial accounting choices overlooks the psychological and the behavioral characteristics of top management as a factor that influence their decision outcomes. One reason could be that it is not easy to reliability measure psychological attributes and managerial behavior (Hambrick, 2007). Thus, later studies that explore such influences abstain from directly measuring executives' values and psyches. They usually seek one of two ways to assess executive's psychological attributes, either define a scoring model from available data on executives, or they extract from observable characteristics, and assume specific psychological characteristics. Behavioral aspects of top executives that have been examined in accounting and finance literature are overconfidence, executives' ethical behavior and narcissism. Though many might confuse each for the other, each aspect has a different meaning in psychological literature as will be discussed in the sections that follow. The concept of narcissism will be discussed more thoroughly as it is the focus of this paper.
Confidence refers to one's excess certainty about his own knowledge, ability and judgment that is not justified (Merriam-Webster definition [4], as cited in Brunzel, 2020). Another definition of overconfidence by Chen (2010) is that it inflates the subjective probability that a specific outcome will occur. Moore and Healy (2008) postulate three-fold traits that construct overconfidence; first, an illusion of control by overestimating individual's actual performance; second, better-than average illusion by over placing individual's performance relative to others; third, illusion of excessive precision in individual's belief. Literature posits that overconfidence can have an unfavorable implication on organizational performance. Scholars provide evidence that overconfidence may explain why new ventures fail, may lead to incomplete and ineffective information, lead to a greater likelihood of committing fraud and restatements, and that overconfident executives may pursue higher-risk strategies as they are more risk insensitive compared to their peers (Hayward et al., 2006;Schrand and Zechman, 2012). Moreover, Hribar and Yang (2016) examine overconfidence influence on issuing forecasts and how accurate they turn out to be. They find that overconfident CEOs are issue more earnings forecasts and subsequently they tend to miss their own forecasts. In general, research shows that overconfidence effect manifests in risky and high discretion contexts (Brunzel, 2020). This is consistent with Hambrick (2007) that suggest better upper echelon theory predictions in settings of high discretion and/or high job demands.

Executives' narcissism and decision making
Few studies examine the ethical behavior of executives and its effect on financial reporting decisions. For example, Beaudoin et al. (2015) study the ethical scores of CFOs and find that in the presence of conflicting corporate financial and personal financial incentives, CFOs tend to resist self-motivated earnings management if they are with higher personal ethics. Other studies adopted the methodology of executives' language in conferences or shareholder meetings to predict their tendency to financial misreporting. Literature shows that executives with low rate of honesty during earnings conference calls and/or shareholder letters are associated with higher probability of latter misreporting and lower earnings quality (Larcker and Zakolyukina, 2012;Patelli and Pedrini, 2015).
Another characteristic examined in literature is narcissism, which will be discussed in detail in the next section. Narcissism trait in top executives, and its influence on reporting decisions, has captured an increasing attention in the 2000s. Many researchers examined narcissism in different ways, Olsen et al. (2014) for example use a as an unobtrusive measure in prior literature to capture narcissistic tendencies which is the CEO photograph in annual reports. They find that narcissistic CEOs are more inclined to manage earnings via real activities. Ham et al. (2017) use CFO signature as a proxy for narcissism and find that narcissistic CFOs are more aggressive in managing earnings, report at a lower conservative manner, and tend to have more restatements. Using an experimental approach, Murphy (2012) studies more than 200 participant using questionnaires that assess personal inclinations and finds that those who show higher attitude towards misreporting, are more likely to misreport and they tend to feel less guilty about it. A review on accounting executives narcissism provides that highly narcissistic executives think of themselves as leaders, using their discretion and knowledge of accounting procedures to protect themselves and the company (Utama et al., 2020). The concept of narcissism is easily confused with overconfidence; hence, some argue that it does not necessarily have a negative influence on performance outcomes. In the section below, the concept of narcissism, how it differs from overconfidence and the argument about its positive and negative influence will be discussed.

The concept of narcissism
The American Psychiatric Association (APA) provides guidelines that define narcissism as a multilayered personality attribute that combines grandiosity, an unrealistic overstated selfview, attention seeking, a need for that self-view to be continuously reinforced through selfregulation and a lack of empathy (APA, 2013). APA considers narcissism a part of diagnostic and statistical manual for mental disorders (DSM) (look at Table 2). Grandiosity implies entitlement, a self-centered image and a conviction that a person is better than the rest.

Criterion
(1) A grandiose sense of self-importance, firmly holding to the belief that one is better than others, condescending toward others (2) Preoccupation with fantasies of unlimited success, power and brilliance (3) Belief in "special" or unique status (including fixating on associating with high-status people or institutions) (4) Requirement of excessive admiration (5) Unreasonable sense of entitlement or expectations (6) Interpersonal exploitative, i.e. takes advantage of others to achieve his or her own ends (7) Lack of empathy, unwilling to recognize the feelings and needs of others (8) Envious of others or believes that others are envious of him or her (9) Arrogant behaviors or attitudes Source(s): APA (2013) Attention seeking suggests that a person desperately strives to be the center of attention. An unrealistic self-view implies that an individual's identity is exaggerated in an inaccurate way. Self-regulation refers to employing all what is in hand from mechanisms, tactics, to strategies to control how others see them. Finally, a lack of empathy refers to an inclination to exploit situations and people for a personal gain (Cragun et al., 2020). Other studies in literature viewed narcissism as consistent but multidimensional personality trait that can be defined as the extent to which an individual has an exaggerated sense of oneself and is obsessed with reinforcing this sense continuously.
The main symptoms of narcissism comprise feelings of entitlement, a constant need for attention and admiration and feelings of superiority. Narcissist develops a biased selfperception through a false upward evaluation of his abilities and performance, an evaluation that lacks objective evidence and feedback (Ham et al., 2014;Lin et al., 2019). In the realm of top executives, the previously mentioned attributes provoke leaders to appear as charismatic, thereby help leaders to maintain power over others. Evidence in literature provides that narcissistic leader causes suboptimal outcomes by taking over the decision process while disregarding feedback and/or ideas from other members involved in the same process (Resick et al., 2009;Nevicka et al., 2013). Likewise, Goncalo et al. (2010) show that narcissists not only believe that they are more creative but can convince others of their superior creativity, although their work is of equal creativity in reality.
3.3.1 Narcissism measures in literature. Researchers and psychologists in prior literature have tried to conceptualize narcissism to be more than just a syndrome. Cragun et al. (2020), in their meta-analysis of CEO narcissism literature, have identified five narcissism measurements, namely: narcissism index, psychometric third-party, psychometric selfreport, use of pronouns in conferences and annual reports, and signature size. These different measures come to reduce the need to complete psychological scales. Perhaps the first, and most widely recognized, and frequently used is Narcissism Index Hambrick (2007, 2011) are the pioneers in the field of management and accounting to use this index, then followed by other researchers (Liu, 2009;Schrand and Zechman, 2012;Olsen et al., 2014;Olsen and Stekelberg, 2016;Buyl et al., 2019). The original index included five elements: (1) The size of the top executive's picture in the annual report, (2) The number of top executive's mentions in company press releases, (3) The number of first-person singular pronouns used by the top executive during interviews, The relative cash pay of the top executive to the next-highest paid executive, and (5) The relative noncash pay of the top executive to the next-highest paid executive.
Nevertheless, the index has some limitations. First, it has been argued that this narcissism index does not directly link to all subdimensions of narcissism (e.g. grandiosity). Second, two of the components in the index could be affected by factors not within the executive's control as they are based compensation and firm size (Cragun et al., 2020). Another measure used in literature is based on a psychometric self-report, which is called Narcissistic Personality Inventory (NPI) developed by Raskin and Hall (1979) and later modified by Raskin and Terry (1988). NPI is widely used as an assessment of narcissism and is considered to be valid and accurate. The original version consists of 40 questions that capture narcissism, as a conceptual validation of narcissism index. Though this measure would seem to be researchers' first choice, it is rarely used due to difficulty in obtaining the data, as it is directed to high-level executives who would most likely overlook any questionnaire sent towards them. A third measure sometimes used as complementary to NPI, is a psychometric third party where a Executives' narcissism and decision making qualified observers are hired to directly observe the executive's behavior and rate their narcissism based on their biographies. Nevertheless, this is hardly used due to difficulty of accessing such observers. Pronoun usage has been employed as a measure of narcissism in some studies (Capalbo et al., 2018). It draws on narcissists' style of speech and calculated by comparing the use of singular pronouns versus plural pronouns in annual reports, shareholders letters and press conferences (e.g. I, me, vs. we, us). Pronoun usage on its own is rarely used as a measure as it requires further validation, usually is combined with other components of Narcissism Index (Chatterjee and Hambrick, 2011).
A more recent measure for narcissism in literature is signature size, which is used as an alternative unobtrusive measure. It was employed by Ham et al. (2017) and Ham et al. (2018) who suggest that a larger signature reflects the grandiosity of a narcissistic executive. They validated this measure by constructing an experiment on a number of students and matched their signature with the score they obtain to a set of questions that captures narcissism nature. This measure has an advantage that it is directly under the control of an individual; however it may not fully reflect the multilayered nature of narcissism.
Researchers have used and developed multiple ways to measure narcissism, however, the difficulty of obtaining some measures lead them to employ others. Some measures provide more validity than others but are difficult to obtain their data, while other measures are easier and more useful, but they do not provide rich evaluation as scaled measures. Thus, researchers choose the measure based on their sample while balancing between availability of data, advantages and disadvantage of alternative measures of choice.
3.3.2 Narcissism versus overconfidence. The concept of narcissism and overconfidence are sometimes falsely used reciprocally. Though the two concepts are related, they have distinctive meanings. Hence, it is important to distinguish between the two terms specially that their use is gaining popularity in accounting and finance research.
Overconfidence in psychology research refers to the likelihood that one makes more optimistic predictions than relying on actual/objective prediction probabilities, which in return suggests higher levels of risk-taking (Olsen et al., 2014). This optimism nature is why in finance literature, the concept of overconfidence is sometimes substituted by the term wishful thinking, or unrealistic optimism, as overconfident individuals tend to make upward-biased estimations of future outcomes. Executive overconfidence is a rather growing stream in literature. Many studies have examined the relationship between CEO overconfidence and corporate strategic decisions (Hribar and Yang, 2016;Brunzel, 2020;Chen et al., 2021). One particular study that is closely related to accounting decisions is a study by Schrand and Zechman (2012). They examine whether specific personality tendencies could lead to a greater likelihood of earnings management. They measure the upward bias/optimism manifested by overconfident executives to overestimate earnings. They show that overconfident executives unintentionally misreport because of their optimistic decisions.
A distinguishing difference between overconfidence and narcissism is that narcissists exhibit a constant need for admiration, recognition and attention from others. Narcissists have an inflated belief in their abilities, and hence are likely to have magnified overconfidence, but they pursue personal rewards at the expense of others, which could aggravate poor decisions-making compared to overconfident mangers (Olsen et al., 2014;Ham et al., 2017). Buyl et al. (2019) argue that narcissism mirrors overconfidence but associated with a rather strong focus on personal incentives. This conclusion comes in line with Chatterjee and Hambrick (2007) argument that the two concepts are closely related but narcissism posit the need for constant self-assurance from outside parties to confirm superiority. It is vital to know the difference between these two concepts when conducting research to be able to choose the suitable measure for the construct of interest.

Narcissism: good or bad.
Narcissism is usually interpreted in the negative sense whether in psychology literature, accounting literature or even in the press and social media platforms. Many academics find a negative effect when narcissism is a personality trait in top executives. Hornett and Fredricks (2005) for example construct a case study that analyzes a number of executives who have been accused of ethical misbehavior. They find that many of these executives do have a narcissistic attribute that drove them to pursue individual rather than corporate goals. Other researchers show that narcissism is negatively affecting the firm, whether through high earnings management or unrealistic inflated firm performance (Frino et al., 2015;Capalbo et al., 2018;Ham et al., 2017). Ham et al. (2017) also provide that narcissistic executives are associated with less timely loss recognition as measured by their responsiveness to good versus bad news have weaker internal control systems and are more likely to experience restatements. O'Reilly et al. (2014) find that narcissistic CEOs exploit their influence and how others confide them to extract higher compensations compared to other executives. In risk related research, Foster et al. (2009) posit that narcissists show motivation to achieve desirable goals but tend to be negligent to avoid negative outcomes.
In contrast, a stream of research argue that narcissism sometimes has a positive effect in the notion that many of the traits embedded in narcissists overlap with ones that are in good leaders. High self-confidence, charisma and a vision for future are all traits of a strong leader, hence having a narcissist at top-level management may turn positive outcomes. For example, Chatterjee and Hambrick (2007) show that narcissism exhibited by CEOs is positively associated with the number and the size of acquisitions, as well as firm performance as measured by shareholder return and retun on assets (ROA). In consistent with the idea, Olsen et al. (2014) provide that firms have higher earnings per share (EPS) and a higher share price when there is a narcissistic CEO on board. Chatterjee and Hambrick (2011) added to their argument about this positive effect and suggest that CEOs may be seen as charming, competent and charismatic in the short term; however, over time, they demonstrate entitlement, hubris and overstated overconfidence. In their review of narcissism literature, Campbell et al. (2011) raised the question of when narcissism should be considered as good or bad and suggest that researchers should consider the conditions under which narcissists thrive. To answer this, Perez (2017) explored some of the conditions and suggests that narcissists thrive in times of financial distress as it induces risk-taking, provides an opportunity to move quickly without oversight and encourages proving one's ability.
Narcissism, per se, is neither good nor bad (Cragun et al., 2020); however, narcissism can cause harm if it is left unchecked. Therefore, scholars should try to seek a better understanding of the concept and suggest proper mechanisms necessary to mitigate or avoid the potential downsides of narcissism.
3.3.4 CEO and CFO narcissism. Executives on the top are in charge of organizational strategies which means their decisions are directly related to performance. The influence of personal characteristics of top executives on the organizational performance is rooted in the upper echelon theory of Hambrick and Mason (1984). The upper echelon theory posits that the cognitions, emotions and values of executives, manifest in their decision-making process. Psychology literature provides evidence on the relation between narcissism and unethical behavior, arguing that narcissistic managers are characterized as over-confident, self-benefit seekers and are expected to make decisions regardless the consequences (Resick et al., 2009;Goncalo et al., 2010;Tamborski et al., 2012).
Literature on executives' personal characteristics influence on decision making and performance have mostly focused on CEOs as they are responsible for majority of firm's activities. Also, CEOs data is relatively much easier to obtain, hence it is easier to construct proxies based on observable elements. For example, CEO's picture in the annual reports, the use of personal pronouns in shareholders letters and press releases are easily accessible and more likely to lie within the control of the CEOs. Research has provided some evidence on Executives' narcissism and decision making CEO narcissism link to performance, aggressive merger and acquisition strategies, tax avoidance, return volatility and bad investments (Chatterjee and Hambrick, 2007;Aktas et al., 2016;Olsen and Stekelberg, 2016;Ham et al., 2014Ham et al., , 2018. Existing literature do provide evidence that personality traits could influence executives' decisions, however CEOs personality traits are more likely related to strategic domain for making operating not financial decisions (Malmendier et al., 2022). Hence, while CEO traits are of importance to many operating decisions, it is more likely that CFO personal traits are the focus when examining financial reporting decisions and outcomes (Ham et al., 2017). CFO personality attributes research is limited because of the less accessible data and difficulty in constructing valid proxies. A pioneer study in CFO traits is by Ge et al. (2011) who provides that CFO fixed effects are associated with a number of accounting practices. Ham et al. (2017) complement these findings by adding that narcissism is an important factor to consider when examines financial reporting decisions because CFOs bear the responsibility for reporting accurate and timely financial disclosures. Moreover, they oversee internal control systems and financial reporting decisions.
Prior literature shows that narcissistic CFOs are likely to falsify financial reporting for several reasons. First, narcissistic people tend to have a high level of self-entitlement and selffocus; they seek appreciation and recognition and hence, are likely to misreport reporting to appear in a better image. Moreover, they tend to take decisions that reflect their self-interest behavior regardless of the others (investors and other stakeholders). Second, CFOs who possess narcissistic traits tend to dominate decision processes and unwilling to accept advice from others (Chatterjee and Hambrick, 2011). Their overconfidence prevents them from reacting to bad news, which in some case results in fallacious portrayal of the entity's financial status. Finally, narcissistic CFOs exert time and effort obsessing over their profile instead of directing time towards ensuring that the financial reports are free of misstatements. These shortcomings could create a variety of negative outcomes, such as financial restatements, internal control weaknesses as well as lawsuits. Yet only few papers tackled how CFO personality traits can affect financial reporting outcomes (Ham et al., 2014).

Literature review 4.1 Executive characteristics and firm performance
A growing literature in accounting and finance examines how managers make financial reporting decisions . Earlier studies adapt the neoclassical view that assumes managers are perfect substitutes; this means that when they are presented with the similar economic circumstances, they would make the same accounting choices. Hence, under this view, managers' decisions are not influenced by their individual traits. However, recent studies shift to the upper echelons theory that takes into consideration specificindividual attributes shaped by their personalities, experience and values when analyzing managerial strategies and how these attributes may affect their decision making (Hambrick, 2007;Hambrick and Mason, 1984). The psychology literature suggests that factors such as confidence and risk attitudes affect the decision-making process (Bonner, 2008). In the recent years, researchers have tried to pinpoint the observable characteristics of managers that may affect their accounting choices. Nevertheless, the results are somewhat contradictory and take limited space in the literature on accounting choice (Dauth et al., 2017).
In support of upper echelons theory, literature has provided empirical evidence that the manager fixed effects have influenced firm policies and outcomes (Nielsen, 2010;Hiebl, 2014;Yamak et al., 2014). Bertrand and Schoar (2003) investigated different firm's policies and find that manager's individual characteristics affect the firm's operational strategies (research and development (R&D), ROA and advertising), financing policies (leverage, dividends to earnings, interest coverage and leverage) and investing policies (capital expenditures, Tobin Q and cash flow sensitivity). Financial reporting literature also provide some empirical findings showing that executives' characteristics are reflected in the financial reporting information that is given by listed companies (Pl€ ockinger et al., 2016) and public sector organizations (Anessi-Pessina and Sicilia, 2020) to external stakeholders.
In accounting literature, executives' characteristics have been found to have an effect on the firm's financial performance, earnings quality, tax avoidance, risk taking and disclosure. Several studies have documented executive-specific fixed effects, or styles, in the context of conference call tone (Davis et al., 2015), voluntary disclosure (Bamber et al., 2010;Yang, 2012), the firm's investment behavior and financing policy (Bertrand and Schoar, 2003), tax avoidance (Dyreng et al., 2010) and accounting practices and earnings management De Almeida and Lemes, 2019). Dejong and Ling (2013) show that firm's accruals are affected by the individual's personal operating style. According to Davis et al. (2015) manager's characteristics (such as their career path and their banking and consulting experience) affect their optimistic tone in conference calls. Within the same line of research, Bamber et al. (2010) find that the disclosure styles of mangers (such as precision, frequency and bias of disclosures) are affected by the demographic characteristics of these managers, namely, age, professional background or education.
Within the same context and to better understand how individual's psychological attributes affect their decisions, Utama et al. (2020) performed a content analysis of 52 articles with narcissism as the main theme as one of different personality traits that affect the decisionmaking process. The purpose was to understand how executives' personalities can influence their decisions. They show that low earnings quality is associated with narcissistic executives' discretionary choices in accruals. They also provide that executives' optimism in disclosure may reflect an obscure picture about firm performance. Other research has identified some manager specific effects on accounting choices. Dyreng et al. (2010) argue that individual executives' characteristics affect their choices for tax avoidance as measured by Cash and GAAP [5] effective tax rates. Their results show that these characteristics have a significant role in adopting tax avoidance behavior, but no evidence to explain the variation in these effects.
A strand of research focuses on the impact of personal attributes on accounting practices and earning management, which uses personal attributes to explain the variation between executives in making decisions and how this is reflected in firm's financial performance. Ge et al. (2011) investigate how executives' style affects their accounting decisions. They focus on three observable characteristics including risk attitudes and confidence, on financial reporting quality. Nevertheless, they find only limited evidence of the impact of these observable CFO characteristics on CFOs' reporting choices. Habib and Hossain (2013) review how CEO/CFO characteristics are associated with different properties of accounting information. They focus on executives; turnover, overconfidence and gender to explore their relation to financial reporting quality and reporting outcomes. Their review did not reach conclusive results with regard to the association between turnover and reporting quality, however they argue that empirical findings in literature show overconfident managers to use more income-increasing accruals choices and issue more optimistic forecasts, a choice that would require them to manage earnings to reach earnings expectation benchmarks. While other researchers may share the same lack of evidence of an association between personal attributes and financial reporting, they provide new insights to literature as the economic consequences of executives' behavioral aspects may go beyond merely affecting financial reporting. More specifically, the behavioral integrity of CEOs has been found to affect the audit fees imposed on an entity (Dikolli et al., 2020). By investigating computational linguistics in CEOs' stakeholders' letters to conceptualize behavioral integrity, they find that audit fees are decreasing as CEO behavioral integrity increase. They rationalize this result as the higher audit fees reflect greater audit effort to mitigate the effects of low behavioral integrity. Moreover, although they find no relation between CEO behavioral integrity and Executives' narcissism and decision making misreporting, results show that low-integrity CEOs impose a higher likelihood of lower future firm performance. While a many research papers focused on CEOs personal characteristics in association with firm performance and managerial choices, only some shifted the focus to CFOs attributes, stemming from the notion that CEOs are more likely to affect strategic decisions while CFOs, due to their job nature and their ability to exercise discretion, are more likely to be influencing financial and reporting decisions Feng et al., 2011;Dejong and Ling, 2013;Ham et al., 2017). For example, Ge et al. (2011) track 359 CFOs across different firms over time and investigate whether individual CFO characteristics, which they referred to as style, impact firm's discretionary accounting choices. They find that CFO style affects accounting choices, and that this effect is more evidenced when CFOs' job discretion and job demands are high, in conformity with the assumptions of upper echelons theory. More recently, De Almeida and Lemes (2019) examine the association between the observable characteristics (gender, age, tenure, educational level, diversity of functional background and level of internationalization) of CFOs and their accounting choices. Their results indicate that CFOs can adopt policies to increase earnings or operating cash flow that fit their personal characteristics because of the flexibility inherited in the accounting choices.
Literature has documented that those behavioral and personality characteristics affect firm outcomes documented in management, psychological, accounting and economic literature (Brunzel, 2020), one stream of literature has been focusing on a specific personality attribute, namely narcissism (Ham et al., 2014(Ham et al., , 2017Lin et al., 2019;Pl€ ockinger et al., 2016;Young et al., 2016). The intent of this direction in literature is to draw the attention towards individual's psychological standpoint in their decision-making process. The framework underlying these studies is referred to as narcissistic accounting. This paper works within the same framework to explore how the narcissistic attribute of key executives affects their accounting decisions, and whether the association could be mitigated or moderated by other factors such as governance mechanism.

Narcissism and firm performance
The underlying philosophy of narcissistic accounting is that highly narcissistic persons occupying executive positions deem themselves as leaders who can use financial and accounting information, procedures and/or loopholes to protect themselves and the company (Pl€ ockinger et al., 2016;Lin et al., 2019).
The mention of narcissism is usually interpreted in a negative sense in public media and daily discourse. In the narcissistic accounting literature, there have been some opposing views on the value of having a narcissist as a top executive. On the one hand, a line of research finds that narcissism as a trait in leaders is associated with negative effects. On many occasions, literature portrays narcissism as an unethical behavior because of the pertaining nature of excessive self-entitlement that under some circumstances stirs up actions that benefit narcissists themselves at the expense of others. They tend to be deceitful and presume that rules are not abiding to them; hence they are more likely to violate norms to benefit themselves (Ham et al., 2017). Moreover, narcissists are less willing to accept advice from others as they believe that they are always right and hence, dominate the decision-making process. Prior literature also suggests that narcissists tend to be less tolerant to monitoring and some even may prefer to hire personal who are not effective monitors (Young et al., 2016;Chatterjee and Pollock, 2017). Narcissists therefore are likely to manipulate the system or else, design a system that will render them flexibility.
Several academics study how narcissism is associated with earnings management, measured by abnormal discretionary accruals, discretionary expenses, abnormal production costs and abnormal operational cash flows (Frino et al., 2015;Ham et al., 2017), less timely loss recognition (as in the timeliness of good versus bad news); high probability of restatements and ineffective or weak internal control (Ham et al., 2017). Other researchers on the same note studied narcissism effect on a firm and found that it could be via; high earning management, higher probability of misstatement, weaker internal controls, resistance to board strategies and sometimes higher compensation schemes (Hales et al., 2012;Zhu and Chen, 2015;O'Reilly et al., 2014). Capalbo et al. (2018) find that narcissistic CEOs overidentify themselves and are more likely to exert efforts to reach their goal even if it means to adopt unethical policies. Hornett and Fredricks' (2005) conducted a case study analyzing how individuals perceive corporate executives who were featured at that time in the press for their criminal acts and ethical violations. They document that these accused leaders have a narcissistic trait that led them to work for their own benefit rather than working to achieve corporate goals. Strategic decisions have been found to be affected by narcissistic CEOs, for example Aktas et al. (2016) find that on mergers and acquisitions (M&A) firms, the acquired shareholders react in a less favorable way takeover news if the CEO of the target firm is more narcissistic.
Another research stream shows that some traits found in a narcissistic individual overlap with good traits of strong leaders, including future strategic vision, high confidence and charisma. Moreover, some research shows that narcissistic traits might actually have a positive effect on the firm via higher EPS and share price (Olsen et al., 2014). Hence, it is not widely agreed upon that hiring a narcissistic at the top is always a bad choice.
Psychology literature supports this later view and suggests that narcissism could be a desirable leadership trait under some circumstances. Maccoby (2004) for example provides that narcissists may bring benefit in chaotic times but might be of a disadvantage in other tranquil times. Within the same line of literature, a study by Campbell et al. (2011) suggests that researchers should explore the conditions under which narcissists flourish or become short. Nevicka et al. (2013) show narcissistic leaders are chosen in the face of high uncertainty about the business environment, in spite of their negative traits. In other words, in times when things could not get any worse, narcissists are to be chosen despite their negative traits, that is, if the company has lost market share, the company's status is one in which the company is in difficulty, the share price has dropped rapidly, the work environment is unpredictable and/ or where a sense of stress is spreading through the company.
Phillips (2019) examines how CEOs specific characteristic, which is narcissism, affect their financial reporting behavior under different market conditions. She finds that although narcissistic CEOs tend to have an aggressive behavior in financial reporting and earnings management, this behavior is toned down by the market conditions. The results are not in line with relevant research by Foster et al. (2009), who argue that in periods of market euphoria, people are more risk tolerant and narcissistic managers are inherently more likely to take risks because of their heightened perceptions of benefits. While in periods of market crashes, the general belief is that the public would be more risk averse, however narcissistic CEOs will keep on their financial reporting aggressiveness. They interpret this result as when narcissists feel that they are questioned, doubted or wronged in some way, they want to prove the public that their decisions were not at fault, so they keep on the same reporting pattern instead of trying to mitigate the firm's risk. These opposing views altogether suggest that narcissism in itself maybe desirable in some conditions of high distress when there is a need for speedy strategies.

Implications
Narcissistic accounting is a fundamental concept that could be adopted in different types of research. The analysis of different articles in the past few years from highly reputable journals provide some insight that can be useful for future research. First, different results in literature have been attributed to the different measures used for narcissism; literature provides no consensus on which measure to use. While interviews and personal data extraction provides reliable data, the Executives' narcissism and decision making difficulty in obtaining such data prevents many scholars from depending on this type of measurement. Moreover, a time series analysis might be impossible, because it will be very difficult to hunt down executives that have been in one company and conduct surveys or interviews with them. Unobtrusive measures are commonly used in narcissism research, but yet each has its advantages and disadvantages; hence, future scholars may combine measures to increase validity or even innovate new ones. Second, most of the research studies on narcissism focus on CEOs as the decision maker, very few examined CFOs influence. Future researchers should focus on CFOs when they examine financial performance as the responsibility for financial reporting preparation and supervision lies within their hands, also, they possess the influence over discretionary accounting choices. Third, prior research does not provide much insight on how to mitigate or avoid the harmful outcomes of narcissistic behavior; hence future researchers could try to investigate moderating factors such as control systems, board attributes and corporate governance mechanisms. Fourth, some researchers study how narcissism influence differs under different settings Liu et al., 2016;Janahi et al., 2021). Future research can be conducted to show how the regulatory environment affect or moderate narcissism effect, moreover, future research can be conducted to examine how the capital market (e.g. analysts' pressure) affect this relation. Furthermore, researchers may examine whether executives' narcissistic nature is suppressed under greater regulatory constraints or greater capital market pressure. Finally, this paper sheds the light on the importance of including behavioral assessment in selecting individuals in top positions. Since behavioral aspects of individuals are more believed to be a major factor in nowadays environment, and since the data is very difficult to find, practitioners and regulatory bodies shall take into consideration the importance of providing an accessible database to enable further in-depth analysis. Future researchers, with data accessibility, may benefit from the use of robotics and data analytics indices/platforms.

Conclusion
This review aims to provide a systematic review on how narcissism affects financial reporting decisions in accounting literature. It does not claim to resolve the debate about whether narcissism affect decision making in top executives' level, but it does provide insights for future research. Narcissism continues to attract the attention of many academic scholars as well as practitioners; hence there is still ample room for narcissism research. We begin by providing a framework for valuable theories used in literature to explain decision making and focus on upper echelon theory that is the base for behavioral research in accounting and finance literature. Thorough the review of prior literature, we find that although many papers find that narcissism encourages a person in top-level management to make decisions that have a negative effect on the financial performance, some papers provide evidence that narcissism serves the company in other settings. We acknowledge these discrepancies and suggest that using different measures for narcissism may be one of the reasons that lead to these different outcomes. We then examine these measures and shed light on the pros and cons of each measure. Nevertheless, future research could try to explore other measures and control mechanisms that can mitigate or avoid the downsides of executives' narcissism. This paper postulates that it is crucial to take narcissism into consideration when investigating performance and encourages future research to explore and innovate new narcissism measures, new theoretical approaches other than known theories and investigate performance issues that lie beyond CEO's narcissism direct/immediate effects.

2.
A review on the literature on accounting conservatism can be found in Ruch and Taylor (2015).