High ability managers and corporate ethical decisions: an empirical evidence from China Open Access

The importance of corporate ethical decision-making extends to financial reporting practices, impacting organizational behaviors and approaches. Compliance with regulations is just one aspect; considering the broader impact on stakeholders and society is crucial. Earnings management, a form of managerial discretion in financial reporting, presents a key area for ethical decision-making, emphasizing transparency, accuracy and fairness. Balancing a true portrayal of financial performance with market expectations is essential. Evaluating corporate ethics in earnings management involves assessing contributions to societal well-being and aligning with expectations of corporate responsibility. Demonstrating commitment to ethical practices includes transparency, accountability and adherence to ethical standards. Prioritizing ethics in earnings management enhances trust, mitigates risks and fosters societal welfare, underscoring its importance for long-term sustainability and success.

As a result of big corporate giants’ downfalls in the last few decades, corporate stakeholders have remained concerned about earnings manipulation for several years. There have been extensive debates among researchers concerning the ethical issues involved in earnings management (Nekhili et al., 2021). However, the personal traits due to which the managers are likely to fall into this practice remain inconclusive (Huang and Sun, 2017). An enormous research stream documents the nexus of earnings management (EM hereafter) with corporate governance (CG) mechanisms (Peasnell et al., 2005), firms’ financial aspects (McNichols and Stubben, 2008), audit characteristics (Klein, 2002) and the macroeconomic institutional environment (Enomoto et al., 2018; Khan et al., 2021a). The need has been instigated to ascertain the characteristics of chief executive officers (CEOs) and managers who are (not) involved in earnings maneuvering activities. Considering this, the study uncovers whether management with superior ability is more likely to be (not) interested in earnings manipulations.

Acknowledging the multifaceted ethical dimensions inherent in earnings management practices is crucial when examining the relationship between earnings management and corporate ethics. Earnings management, characterized by the manipulation of financial statements to portray a desired financial performance, constitutes a significant ethical issue within the realm of CG and financial reporting (Khan et al., 2021a). This practice raises concerns related to the misrepresentation of financial performance, violation of stakeholder trust, distortion of capital markets, potential legal and regulatory infringements and adverse long-term consequences. By engaging in earnings management, companies risk compromising the principles of transparency, integrity and fairness that underpin ethical financial reporting practices (Böcking et al., 2015). Moreover, the positive correlation between high-level managers and earnings management, as elucidated in this paper, underscores the critical importance of understanding the ethical implications of managerial decision-making.

According to the postulates of upper echelons theory (UET) (Hambrick, 2007), managerial traits influence organizational outcomes (at least in part). EM encompasses using management discretion to select accounting practices and estimates to manipulate the results of the financial reporting process. Accounting earnings are handled to conceal information and protect personal gains from the control that enables businesses to avoid reporting earnings losses (Leuz et al., 2003). While earnings management itself may not necessarily contradict accounting principles, managers who engage in earnings management are not considered ethical as they negatively impact their external stakeholders (Dechow et al., 1996).

This study produces a compelling insight. To begin, the outcome of this research adds to the body of knowledge on managerial skill and earnings manipulations (Huang and Sun, 2017). We also found that more highly skilled managers are connected to lower accrual quality, and this is in line with earlier research (Francis et al., 2008). Contrary to popular belief, our findings (Huang and Sun, 2017) suggest that companies managed by managers with greater ability have more inclination toward earnings manipulation activities. Although this result would contribute significantly, it does not undermine the importance of hiring managers with higher abilities. Second, contrary to most studies investigating the significant firm-level and institutional attributes that impact the EM, this study directly explores the effect of individual characteristics on EM magnitude. More and more studies examining the motivations, methods and consequences of EM focus on corporate traits (Huang and Sun, 2017; Hambrick, 2007); managers’ personalities remain inconclusive. Extant studies have explored the association of accounting measures (accounting practices, quality of information, conservatism, investments, misrepresentation) with CEO reputation (Demerjian et al., 2012), directors (Ge et al., 2011), CEO overconfidence (Schrand and Zechman, 2012) and CEO optimism (Malmendier and Tate, 2005). However, the links between management skills and EM were not primarily discussed in these studies.

To determine whether or if highly capable managers are more likely to be (or are not) interested in earnings manipulation, panel data regression models are estimated using data from companies listed on Chinese stock exchanges between 2010 and 2019. According to the findings, abler managers are expected to use their abilities and talents to enhance earnings management. The research concludes that stringent monitoring effectively instills ethical behavior in managers more than the manager’s ability. The results recommend using highly skilled management and auditing the businesses using reputable auditing firms.

In addition, the findings are robust to various managerial ability and earning management indicators and free from endogeneity problems. Our findings suggest that while managers’ ability to smooth earnings effectively varies, deliberate cooking of earnings management by highly skilled managers appears to be a cost-saving mechanism for the firm to consistently meet earnings expectations. It helps in avoiding violation of debt covenants but is not helpful to facilitate insider trading or agency issue. This should be instructive for boards to evaluate management’s worth and weigh the merits of purposeful smoothing.

Some econometric issue, including endogeneity and self-selection biasness, may have influenced our findings, which is understandable. To address such anticipated issues, instrumental variable approach has been used in this study. Besides, earnings management is widely recognized for being a multifaceted strategy (Demerjian et al., 2020), relatively irreplaceable with earnings smoothing (Doukas and Zhang, 2020), income smoothing (Baik et al., 2020) and quality of earnings (Demerjian et al., 2013; Francis et al., 2008) and is very difficult to measure (Dechow et al., 2010; Dechow and Dichev, 2002). Therefore, we run several tests using alternative earnings management measures to alleviate measurement error concerns.

The current paper makes the following contributions to the existing literature. First, this study makes pivotal contributions to the understanding of earnings management by bridging theoretical gaps and offering practical insights. Second, our study contributes to the knowledge of what Hambrick (2007) refers to as the “different complexions” that UET may take by analyzing data from China, which has a significantly different socioeconomic environment than the developed countries (Khan et al., 2017; Chi and Gooda, 2024). In line with the predictions of UET, but in a non-US context, we uncover on how managerial traits correlate with earnings manipulation, challenging the notion that higher ability equates to ethical behavior. Third, through robust panel data regression analyses, it enhances methodological rigor within the financial and accounting research domains. Fourth, the findings highlight the necessity of effective CG mechanisms, advocating for the strategic hiring of competent managers along with rigorous monitoring and auditing to mitigate unethical financial maneuvers. Fifth, ultimately, the study provides comprehensive implications for both scholarly research and practical application in improving corporate decision-making and ethical standards in financial reporting. Lastly, we also add to the growing literature that documents how the characteristics of CEOs impact earnings quality at the firms they manage. As examples, earnings quality is impacted by the “talent” or managerial ability of the CEO (Cho and Choi, 2023), the tenure of the CEO (Ali and Zhang, 2015) and CEO’s general ability, etc. Our evidence suggests higher ability managers use their talent and skills to mask the firm performance rather than improving the organic performance by manipulating the earnings. We enrich the literature on economic consequences of managerial ability. Recent empirical studies have shown that managerial ability is positively associated with sustainability performance (Khan et al., 2022), earnings quality (Demerjian et al., 2013), investment efficiency (Habib and Hossain, 2013), stock returns (Chen et al., 2018) and financial constraints (Huang et al., 2021). These studies document the bright side of MA. In contrast, several studies appear to uncover the dark side of MA. For instance, Mishra, 2014 shows that CEOs with higher managerial skills may result in more agency problems. Our research contributes to supporting the bright side of managerial ability and reveals that CEO’s use their capabilities to beautify the firm performance by doing earnings management.

To investigate the effect of managerial ability on earnings management, China offers itself as the perfect case study. China is the largest transitional economy and plays an increasingly critical role in the global economy. However, because of the tilt of the financial system toward strict government control rather than market control (Khan et al., 2017), the incidence of financial and accounting scandals has increased dramatically over the past two decades (Wang et al., 2017) . The resulting decline in investor confidence in the capital market and firm financial reporting has received increasing attention from academics, practitioners and regulators. China launched its “Thousand Talents Plan” at the end of 2008 (Yuan and Wen, 2018 ) to recruit the highly skilled CEOs. The Thousand Talents Program, a flagship initiative aimed at attracting high-level talent to China, has garnered significant attention both domestically and internationally. So it becomes very interesting to investigate whether highly capable managers are successful in catering to the earnings-manipulating activities or not.

For the remainder of the draft, we have included the following sections: the next section introduces the literary context in which we formed our hypothesis. The subsequent parts present the research design (data and variable descriptions, empirical model), discussion of results and conclusion.

Many studies have listed the factors that influence the quality of accounting information and concluded that firm characteristics dominate the corporate decisions (Meek et al., 2007; Kontesa et al., 2020; Khan et al., 2021b). A distinct line of research examines the effect of executives and managers on company choices. Meanwhile, the UET suggests that top management behavior is a determinant of organizational behavior, including earnings management (Hambrick, 2007). The UET, developed by Hambrick and Mason (1984), posits that the characteristics and experiences of top executives significantly influence their strategic decisions and, consequently, organizational outcomes. The theory suggests that managers’ cognitive bases, values and perceptions, shaped by their background, education and career experiences, play a critical role in how they interpret situations and make decisions. These psychological and demographic factors act as filters through which executives view the world, affecting their choices and behaviors. For instance, highly capable managers, often characterized by superior expertise and ethical grounding, are likely to prioritize long-term organizational success over short-term gains, reducing the likelihood of engaging in earnings management. By focusing on the human element in decision-making, the UET provides a framework for understanding how managerial ability influences corporate strategies and outcomes, including financial reporting practices. This theory is particularly relevant to our study as it helps explain why more capable managers may exhibit lower tendencies toward earnings management, aligning with our hypothesis.

Based on this theory, the seminal work of Bertrand and Schoar (2003) laid the groundwork for the literature on “management styles.” They demonstrate that CEOs have distinct management styles that influence a wide variety of organizational decisions. Research on management style continues to investigate the relationship further. Bamber et al. (2010) discover that business leaders exhibit distinct “styles” connected with their inclination to offer guidance and the features of the advice they issue. Similarly, both Ge et al. (2011) and Dejong and Ling (2013) explore managerial specific effects on quality of financial information and demonstrate that manager as a person matters because businesses’ accounting and disclosure policies differ as a result of manager fixed effects. The background of a CEO, for example, has been found to possess a significant impact on financial disclosure (Reeb and Zhao, 2013) and financial fraud (Demerjian et al., 2012), reputational status of CEO and earnings quality (Francis et al., 2008), management style and corporate tax avoidance (Dyreng et al., 2010) and firm voluntary disclosure (Bamber et al., 2010). High-ability managers tend to provide higher quality and more informative non-GAAP (Generally Accepted Accounting Principals) earnings than low-ability managers (Kim, 2022) and also better access to the finances during financial constraints (Zahid et al., 2023). The earnings management is the phenomenon that is affected by range of CEO’s personal characteristics (Putra and Setiawan, 2024). These studies demonstrate the critical role of management traits in business decision-making and performance. However, the impact of manageability on income management behavior has been understudied to date.

Agency theory postulates that the CEO manages the company’s earnings to achieve shareholders’ objectives. CEOs take advantage of loopholes in accrual accounting to manipulate a company’s earnings to meet shareholder expectations and increase their bonus and compensation (Dechow and Skinner, 2000). According to Healy and Wahlen, the highest level of management is responsible for shifting the financial statements to meet or influence the targets depending on the numbers reported (Healy and Wahlen, 1999). Most of the time, companies use earnings management to deceive investors or influence contractual outcomes by manipulating financial reports (Healy and Wahlen, 1999). Companies’ earnings performance is adjusted to meet investors’ expectations. These approaches are designed to camouflage any improvements in the firms’ success and development, thus preventing investors from being alarmed.

Management capacities have recently gained renowned recognition as a significant management feature. Managers’ capability relates to a manager’s knowledge, competence and experience (Kor, 2003). Management capacities derive from domain expertise, including sector, company strategies and technology understanding (Kor, 2003). Higher management capability is requisite for every organization because more able managers have a lower propensity to undermine audit quality (Demerjian et al., 2013). For a specific scale of resources, managers with higher abilities can generate more sales revenue and thus are less pressured by the need to meet earnings targets. CEOs with high general ability may manage earnings more given their higher tolerance for failure in the labor market relative to other CEOs (Zbib et al., 2024).

Moreover, managers with a high ability to understand the long-term adverse effect of EM (Demerjian et al., 2020), and thus, they refrain from doing it. Aier et al. (2005) establish a link between Chief Financial Officer’s competence (number of years spent as CFO, experience at another business, higher qualification and certifications) and misreporting, suggesting that organizations with more experienced CFOs had fewer misstatements. Francis et al. (2008) investigate the relationship between earnings information and CEO reputation, quantified by how many times a CEO’s name appears in the business press. In their research, they find that the quality of a company’s profits is negatively correlated with its CEO’s public image. They find that “boards of directors employ particular managers because of their reputation and ability in managing these enterprises” in more complicated and variable operational settings. Huang found that high-ability management can reduce Real EM (REM)’s negative effect on future firm performance because they choose less value-destroying REM (Huang and Sun, 2017). Since managers have less time and energy, they would instead devote more to daily operations versus EM. Highly able managers can also forecast future earnings and predict ups and downs. Hasan et al. (2022) showed that high-ability managers operating in highly competitive industries are more inclined to engage in accrual-based earnings management (AEM) but are less likely to resort to real earnings management (REM). These findings suggest that such managers strategically weigh the relative costs associated with different forms of earnings management and opt for the method that is comparatively less costly. This trade-off reflects their ability to make informed, cost-effective decisions in competitive environments.

Earnings have long been thought to reveal information about management’s capability (Trueman, 1986). According to Baik et al. (2011) and Baik et al. (2020), greater information flow occurs with higher-performing managers than lower-performing ones. Because a tactical smoothing strategy is intricate, it calls for significant insight. The smoother the firm’s earnings, the more accurately managers must be able to predict future profits and reports, and the greater the reporting must reflect the actual operations of the firm (Beidleman, 1973; DeFond and Park, 1997). Thus, the highly skilled management is supposed to devise certain tactics because they are more capable of doing so. This conjecture rationale is alike to the findings of Demerjian et al. (2013), which suggest that managers possessing high abilities are wise enough to understand the ongoing operations of firms along with their associated potential results; by this, they are in a better position to reduce the income noise by using their talent.

On the contrary, highly capable managers will engage in long-term self-destructive actions to maintain and increase their short-term reputations (Demerjian et al., 2020). Highly successful executives have high stock market values, at least partly because of the earnings management effects they can have, making them highly motivated to support shareholders, thus increasing their reputations. In addition, earnings smoothing can reduce the volatility with stakeholders, for example, by maintaining a more consistent flow of earnings over time and helping to keep the company on track about its earnings. Since high-skill and highly effective management is prevalent in the high-cap firms, we expect them to execute these complex earnings manipulation techniques on average more than CEOs with limited abilities and expertise.

Moreover, all else being equal, all managers have to reach or exceed earnings benchmarks. If managers with strong credentials fail to meet earnings targets, they will face even more pressure. As a result, high-ability managers may engage in EM. In addition, managers with greater managerial skills have a better knowledge of the environment of businesses the firm engages in (Demerjian et al., 2013), which helps them match monitoring strategies with EM. All the above discussion made us hypothesize that higher ability management is negatively correlated with earnings management, and the size of audit offers further strengthens this relationship. In summary, the archival studies on managers’ influence on earnings management are ambiguous. While there is some evidence that managers with higher experience have fewer earnings management activities, Francis et al. (2008) demonstrate that renowned managers have been involved in earnings management.

Although estimation is more challenging, we expect less EM to be correlated with more competent managers based on the following justifications. To begin, highly capable managers may increase earnings for a given set of resources owned by the corporations (Demerjian et al., 2012). Therefore, it is less likely that they are indulged in earnings management activities. Second, managers’ decision-making models include opportunity cost as a crucial element. Due to limited time and effort, many talented managers would prefer to dedicate more emphasis to routine operations rather than EM. According to Francis et al. (2016), competent managers are less likely to engage in tax avoidance than less talented managers because of the opportunity cost. Further, capable managers comprehend the negative impact of EM on long-term future business performance (Roychowdhury, 2006; Cohen and Zarowin, 2010). As a result, able managers are less likely to experience EM.

Therefore, we anticipate that more capable managers will have a greater understanding of the company and industry and a greater ability to synthesize facts into credible forward-looking statements. Thus, when the business and industry factors are held constant, and the review as mentioned earlier and evidence is considered, the link between managerial competence and earning management is negative. As a result, the hypothesis of investigation for the current study is as follows:

Highly capable managers are less likely to engage in earnings management (EM) behavior due to their strong ethical leadership, focus on long-term value creation, and superior understanding of financial systems (Lu and Lin, 2014). Skilled managers prioritize integrity and transparency, recognizing that manipulative financial practices can harm the organization’s reputation and long-term performance (Habib and Hossain, 2013). They are more concerned with sustainable growth and operational excellence, relying on strategic decision-making rather than short-term financial adjustments to achieve goals. In addition, their deep awareness of regulatory and legal risks discourages involvement in EM, as they understand the potential consequences, such as fines, penalties and reputational damage (Khan et al., 2022). Highly capable managers also foster a positive organizational culture that emphasizes accountability and ethical behavior, reducing the likelihood of unethical practices at all levels. Furthermore, their confidence in achieving organizational objectives through operational efficiency and innovation diminishes the need for earnings manipulation. By aligning their decisions with stakeholder interests and maintaining strong internal controls, skilled managers create an environment where EM is both unnecessary and discouraged. These factors collectively support the hypothesis that highly capable managers are less likely to be involved in EM behavior. Recently, Anderson et al. (2025) demonstrated a linear relationship between higher-ability managers and efficient utilization of resources.

However, there are several possible explanations for our failure to discover a negative correlation between management skills and EM. To begin, all managers face the pressure of fulfilling profit targets. If a company’s high-ability managers fail to meet their profit goals, they may face additional pressure to keep their image better in the market. As a result, when confronted with profit targets, more talented managers may get involved in EM. Secondly, business operations are well understood by competent managers (Demerjian et al., 2013), which allows them to align EM with their reporting strategy. Like this notion, Koester et al., 2017 argue that abler managers’ greater understanding of their organizations and their operational environments enables them to pursue further tax evasion tactics.

The theoretical discussion presented in the manuscript delves into the intricate relationship between managerial traits and earnings management practices, drawing upon established theories such as agency theory and UET. These theories provide a framework for understanding how executives, particularly CEOs, wield their influence over financial reporting to meet shareholder expectations and enhance their own compensation. Moreover, the discussion highlights the emerging recognition of management capabilities as a critical determinant of organizational performance, emphasizing the importance of managers’ knowledge, competence and experience in shaping financial reporting outcomes.

Furthermore, the theoretical analysis underscores the nuanced interplay between managerial ability and earnings management behavior. While highly capable managers may possess the foresight to understand the long-term ramifications of earnings management and refrain from engaging in such practices, they may also face pressure to meet earnings targets and maintain their reputations in the short term. In addition, competent managers are adept at aligning earnings management with broader organizational strategies and may use sophisticated tactics to navigate the complexities of financial reporting.

Theoretical insights from prior research are synthesized to formulate hypotheses regarding the relationship between managerial competence and earnings management behavior. While the hypothesized negative correlation suggests that more capable managers are less likely to engage in earnings management activities, the analysis acknowledges potential confounding factors such as pressure to meet profit targets and the strategic alignment of earnings management with organizational goals.

The theoretical analysis provides a comprehensive framework for understanding the complex dynamics underlying earnings management behavior, integrating insights from agency theory, UET and empirical research on managerial capabilities. By elucidating the theoretical underpinnings of the research hypotheses, the manuscript lays the groundwork for empirical investigation into the relationship between managerial competence and earnings management, contributing to the broader literature on CG and financial reporting practices.

From 2010 to 2019, the data set includes 4,132 non-financial enterprises that issue A shares and are listed on the Shanghai and Shenzhen stock exchanges. The data set was downloaded from CSMAR (China Stock Market & Accounting Research) database. Firms in financial services business were excluded from the data set, and the winsorization technique was used to omit observations with a 1% tail to minimize potential outlier issues. In addition, firms with three consecutive years of missing values were removed. The final sample consists of an unbalanced panel of 3,052 firms, constituting 20,651 firm-year observations, ranging from 1,373 in 2010 to 2,649 in 2018 (lowest to highest).

We regress the following model by panel data estimation techniques to analyze the impact of managerial ability on earning management for firm i in time period t:

where

$EMit$ is the earning management (accrual-based).

$MAit$ represents managerial ability.

$Zit$ represents the firm’s related control variables such as firm age, Tobin’s Q, debt to asset ratio, return on assets, firm size, Top10Auditor, Z-Score and net operating assets, which may affect the EM–MA relationship.

$Cit$ is a vector to control governance-related effects measured by Board Independence, Herfindahl-Hirschman Index (HHI) and Board Size.

$∈it$ is the specific error term.

$i$ represents firm i, and t fiscal year t.

We analyze the coefficient (⁠$Υ1$⁠) on MA in the model to test our hypothesis. The presence of a positive coefficient indicates that MA is linked to greater EM magnitude. A negative coefficient indicates that managerial ability is linked to less EM use. The dependent variable (EM) is the magnitude of earnings management measured by the residuals obtained from Equation # 2. The decile rank of the residual estimated in Equation # 4 is the main variable of interest in managerial ability (MA). The details of the dependent, independent and control variables are given in the following section. We use the firm and year fixed effect regression model because it is believed to mitigate omitted variables bias.

Earnings management (EM) refers to the strategic manipulation of a company’s financial statements, usually its earnings, with the intention of presenting a more favorable picture of the company’s financial performance than what may actually exist. This manipulation can involve various accounting methods, estimates and judgments to smooth out fluctuations in earnings, meet market expectations or achieve specific financial goals. While some forms of earnings management may be legal and acceptable within accounting standards, others may cross ethical boundaries and deceive investors or stakeholders. The practice of earnings management can involve actions such as altering revenue recognition timing, manipulating expense accruals or engaging in aggressive accounting practices to inflate reported earnings.

Firms use EM activity to either increase or decrease reported income. This study uses the discretionary accruals EM constructs measured by the absolute value of abnormal accruals to incorporate the effects of both kinds of activities. Hence, to calculate abnormal accruals, the study uses the cash flow approach to measure total accruals and estimates the cross-sectional-modified Jones model (Dechow et al., 1996; Jones, 1991) for each firm-year combination to estimate abnormal accruals for earnings management detection. Specifically, we denote with ACC the firm’s accruals in a given year, measured by net income less cash flows from operations; with total assets represented by TA; with change in net sales represented by $ΔS$⁠; with change in receivables represented by $ΔRA$⁠; and with the value of property, plant and equipment represented by PPE. A regression is run to find the discretionary accruals variable figure of firm unit i at time t. The predicted residuals values from equation (1) are treated and labeled as abnormal accruals (Ab_ACC), which is our proxy for EM:

Managerial ability or CEO ability refers to the aptitude and proficiency of executives in effectively utilizing resources to generate revenue and enhance the overall performance of a company. The concept of managerial ability encompasses a wide range of skills and competencies that enable CEOs and managers to make strategic decisions, allocate resources efficiently, foster innovation and navigate complex business environments. Theoretically, managerial ability suggests that certain individuals possess inherent qualities or develop capabilities over time that enable them to outperform others in similar positions. This ability is often measured by various indicators, such as the firm’s financial performance relative to industry peers, the ability to adapt to changing market conditions, strategic vision, leadership qualities and the effectiveness of resource allocation. Essentially, managerial ability reflects the capacity of executives to drive organizational success through effective decision-making, leadership and resource management, ultimately contributing to sustainable competitive advantage and long-term value creation.

Researchers use a range of criteria to evaluate management skills, such as the firm size variable used by Rosen (1990), the historical stock returns used by Fee and Hadlock (2003) and the media evaluation used by Milbourn (2003) and Rajgopal et al. (2006). Because such variables are believed to measure managerial ability in average proportion and lack the testing power, these measurements cannot be very reliable (Demerjian et al., 2012). This research uses the construct of measuring managerial ability (MA) derived from (Demerjian et al., 2020). The premise is that this measure demonstrates the CEO’s ability to turn business resources efficiently into revenue compared to the competition within the industry; managers who can generate higher output from a specific input set should be better at the ability scale. The managerial skill is not measurable and must be derived from observable outcomes of their resource allocation decisions. Demerjian et al. (2012) take two steps to improve their management skill measurement. In the first step, to calculate year- and industry-wise firm efficiency, data envelopment analysis (DEA) was used. This technique necessitates the identification of input and output variables. They used seven input variables: (I) cost of goods sold (COGS), (II) selling, general and administrative expenses (SGA), (III) property, plant and equipment (PPE), (IV) operating lease (OLease), (V) R&D cost (RD), (VI) goodwill (GW) and (VII) other intangibles (OtherInt). Net sale is the output variable. They begin by solving the following optimization problem in DEA:

The efficiency measure (calculated in the above DEA model) can be zero to one. Since both the management and the overall firm can attribute the total efficiency of the firm, the output of total efficiency was split between management and the firm. After that, six such characteristics were identified due to which management’s help is either supported or hindered. Total firm efficiency was regressed against these firm characteristics, including “firm size, market share, positive cash flow, firm age (factors that support management) and complex multi-segment and international operations (factors that hinder management)”. They calculate the Tobit regression as follows:

Equation # 4 residual captures managerial ability. Demerjian et al. (2012) use this measure to create decile ranks of managerial ability (MA in this study) by year and industry so that score could be made more comparable across years and industries and outliers’ impact could be reduced. Our analysis used MA as our managerial ability measure and followed all the stated Chinese data procedures. Many previous studies (Khan et al., 2022) have used the same construct to measure the managerial ability in their researches.

To manage factors that can confuse the relationship between MA and EM, we include several firms- and management-related controls in our tests. We follow Roychowdhury (2006) and include ReturnonAsset (Return on Asset ratio); Tobin’s Q; DebttoAssetRatio (Debt to Asset Ratio); Size (log of total assets); as control variables to reduce the probability that the discretionary accruals could be associated with firm performance, growth opportunities, financial condition or firm size. The factors contributing to management’s ability are BoardSize (board size, measured as a total number of directors on boards) and BoardInde (board independence, measured as the number of independent directors divided by total directors). Following Yuan et al. (2019), we also control market competition and firm age (FirmAge). A market competition proxied by the HHI (the sum of the squared market shares of all firms in the same industry, measured at the end of the fiscal year). Extant studies have found that competition could encourage companies to engage in earnings management-related activities. Following Huang and Sun (2017), operating assets (NOA) control a firm’s financial flexibility to do EM. Following Roychowdhury (2006), we use Altman’s z-score (ZScore) to control for effect at the industry and time levels.

Following Cohen and Zarowin (2010) and Huang and Sun (2017), we have included top audit firms’ factors, as they believe that by this, auditors’ scrutiny will increase. The link between audit firm size and audit quality has long piqued researchers’ and policymakers’ interest in the auditing sector. According to DeAngelo (1981), because of the increased collateral effect, the size of an audit firm can be used as a proxy for the efficiency of the auditors they employ. Lennox (1999b) comes to the same conclusion by citing the higher brand value and litigation rate associated with larger audit firms. Theoretical predictions that big audit firms are linked with better audit outcomes than smaller audit firms have been largely validated by subsequent empirical studies (Becker et al., 1998; Krishnan and Schauer, 2000). The findings in previous empirical studies using the Big N/non-Big N grouping to proxy audit firm size are generally consistent with the theoretical prediction that audit firm size is positively associated with audit quality (Becker et al., 1998; Krishnan and Schauer, 2000; Francis et al., 2013; Zoe-Vonna, 1988). A wide body of empirical research backs the positive link between size and auditor efficiency. Larger audit firms with a global reputation are associated with more accurate and more informative financial distress signals (Petroni and Beasley, 1996; Lennox, 1999a), lower litigation activity (Zoe-Vonna, 1988) and more consistency with the conditions for GAAP communication and with their client’s financial statements (e.g. Krishnan and Schauer, 2000). According to this belief, larger audit firms are better equipped to deal with agency costs. Large auditors’ clients have higher earnings response coefficients. (Teoh and Wong, 1993), IPOs (Initial Public Offerings) are priced significantly lower (Beatty, 1989; Firth and Smith, 1992), and big audit firms are more likely to be hired by companies with high audit quality demand (Johnson and Lys, 1990). To capture the effect of auditing firms’ size, we determine the market share of each audit firm and top10 auditing firms (Top10Audit) are selected.

Table 1 reports the variables descriptive, whereas Table 2 presents a correlation matrix between variables. As seen in Table 1, the mean value of EM accrual-based is 0.042 with a variation of 0.112, implying that very few enterprises are indulging in any EM activities on an average in the Chinese market. This is compatible with the estimating model’s intuition and earlier research outcomes (i.e. Gunny, 2010; Huang and Sun, 2017). Like the real earnings management (REM) variables, except real, earning management production-based (REM_PD) has a slightly higher mean value, i.e. 0.115. The managerial ability (MA) mean values are also close to zero, with a standard deviation of 0.158. The average age of the firms in the sample is 21 years, mean total assets are 0.24 billion, board size is 8 to 9 persons, ROA is 0.24 and Tobin’s Q is 2.038. Big 10 auditors audit 56% of sample firms. HHI mean is 0.319, and the average net operating assets of sample firms are 2.214.

The Pearson and Spearman correlation coefficients for the variables are shown in Table 2. Correlations that are statistically significant at the 0.10 level or below are italicized. Correlations between the EM and MA are positive but not significant. Similarly, most control variables have a very small, insignificant correlation coefficient. So, there is no issue of multivariate collinearity in the variables of our study. We test the regression for multicollinearity by computing the variance inflation factors for each variable. Multicollinearity does not seem to be a significant issue in our investigation. For the sake of brevity, the results are not tabulated. However, further, powerful tests are required to test the hypothesis.

To test our hypothesis, Table 3 shows the results of the association between managerial ability and earnings management. Model 4, represented by equation # 4, investigates the relationship between managerial ability and EM. At a statistically significant level of 0.01, the results show that MA is positively related to EM. Results show that managerial ability is positively related to earnings management magnitude. It suggests that higher-ability managers have a higher magnitude of discretionary accruals and, as a result, do more earnings management than their lower-ability counterparts. The results are somewhat strange, as we expected that higher management ability is translated into better utilization of firm resources, thereby not masking firm performance. This finding is similar to that of Demerjian et al. (2020), who found that purposeful smoothing is beneficial for avoiding misleading technical defaults but not postponing performance-driven defaults. He argued that reducing the possibility of a false technical default by purposeful smoothing decreases contractual costs and benefits stakeholders. Thus, our findings show that the purposeful smoothing actions of high-ability managers benefit shareholders via lower contracting costs. These results corroborate previous research on AEM and managerial capability (Cohen and Zarowin, 2010; Demerjian et al., 2012; Demerjian et al., 2013; Demerjian et al., 2015).

Results fail to substantiate the hypothesis that MA and EM are negatively associated with several possible causes. First of all, everything else being equal, all managers are under pressure to meet the earnings benchmarks. Because high-capacity managers have a higher reputation cost, they may be under even more pressure if earnings targets are missed. In the case of China, a majority of firms are government-owned, and their management is under pressure to show performance to remain in the office, so they are more likely to manipulate earnings.

As expected, many of the control variables used in the regressions are related to EM. Top10 Audit has a negative relationship with EM proxies, implying that companies audited by Top 10 auditors have fewer abnormal discretionary accruals than companies audited by non-top 10 auditors. There is no evidence of a link between firm age and EM proxies. According to EM and Tobin’s Q, when there are more growth opportunities, management with high abilities does more earnings manipulations, positively and significantly related at p > 0.01.

Adoption of AEM is hindered by the financial flexibility of accrual manipulation and external monitoring, according to prior studies (Cohen and Zarowin, 2010; Francis et al., 2008). Firms may also use real-based EM to cook accounting information along with accrual-based EM activities (Khan et al., 2021a). The current study explores if Real EM (REM) is connected with managerial skill to check the robustness of baselines model. Roychowdhury (2006) explains that REM is a collection of managerial decisions that enable corporate operations to diverge from standard operating procedures to fulfill specific objectives. Some examples of these specified goals are earnings thresholds or strategies for concealing annual losses in financial statements. Managers may use their skills and expertise to manipulate the earnings level by following measures such as (i) reducing discretionary expenses, (ii) overproducing and (iii) increasing sales (Zamri et al. (2013); Roychowdhury (2006); and Cohen and Zarowin (2010). Management will benefit from the short-term impacts of REM. Even yet, it’s bad for the company’s long-term value because the steps taken in the confronting period to conceal performance can have a detrimental impact on future cash flows (Roychowdhury, 2006). Subsequently, due to REM’s successful management, abnormally low levels of cash flow, discretionary expenditures and high levels of production costs occur.

Three REM models are developed based on previous research (Cohen et al., 2008; Cohen and Zarowin, 2010; Roychowdhury, 2006): sales manipulation, reduction in discretionary expenses and overproduction. Specifically, denoting with CFO, reporting cash flow; with DE, selling, general and administrative expenses; with PD, production costs calculated as the cost of goods sold plus a change in inventory; with Sales, net sales amount; and with TA, its total assets, the following regression is run to estimate the abnormal level of CFO (Ab_CFO), discretionary expenses (Ab_DE) and production costs (Ab_PD):

Following Enomoto et al. (2018), because three types of REM described above might be implemented to decrease earnings, Ab_CFO, Ab_DE and Ab_PD are converted into their absolute values (Abs_CFO, Abs_DE and Abs_PD, respectively) to be used as our REM proxies. Table 4 summarizes the findings. At a statistically significant level of 0.01, the results show that MA is positively related to all REM proxies. This suggests that higher ability management is linked to more irregular cash flows, discretionary expenses and production costs, leading to more REM use. This suggests that higher-ability managers have higher abnormal production costs and lower abnormal discretionary expenses, implying that they use more REM. These are the similar results that we obtained in Table 3 using AEM. So by this, our results are strengthened.

We also establish a link between CEO duality and EM. In addition to the primary measures of managerial ability, we employ CEO duality as an alternative proxy to ensure the robustness of our findings. Pollock et al. (2002) stated that CEO duality improves managerial abilities. CEO duality, where the CEO also serves as the chair of the board, is a widely recognized indicator of managerial influence and decision-making authority within an organization (e.g., Finkelstein and D’Aveni, 1994; Boyd, 1995). This measure is particularly relevant in the context of managerial ability because CEOs who hold dual roles often possess greater control over strategic decisions and resource allocation, reflecting their ability to influence organizational outcomes. Prior studies have argued that CEO duality can serve as a proxy for managerial ability, as it captures the extent to which a manager has the authority and capability to implement decisions effectively (e.g., Dalton et al., 1998). By including CEO duality in our robustness checks, we aim to account for the potential influence of managerial power and control on earnings management behavior, thereby providing a more comprehensive assessment of the relationship between managerial ability and earnings management.

As a result, we use CEO duality as an alternative managerial ability measure to reexamine our two hypotheses as a robustness check. If a CEO is also the board of directors’ chair, the dummy variable is one; otherwise, it is zero. To proxy for managerial ability, use Model (5) CEO duality instead of MA. Table 5 summarizes the findings of the investigation into the relationship between Duality and EM.

Overall, we find that Duality results in an increased magnitude of EM. This supports our previously obtained results that the higher the managerial ability, the higher the proportion of earnings management activities.

We verify our findings’ robustness using the variable instrumental technique of 2-stage least square (2SLS) regression.

Endogeneity may arise due to reverse causality or omitted variable bias, where unobserved factors could simultaneously influence both managerial ability and earnings management. The 2SLS approach is particularly suitable for mitigating these issues, as it uses instrumental variables to isolate the exogenous variation in managerial ability. For this analysis, we use the year-lagged value of managerial ability as an instrument. This choice is justified by the fact that managers require time to exert their influence and use their skills effectively within the organization. By using a lagged value, we ensure that the instrument is temporally prior to the dependent variable, reducing the likelihood of reverse causality and providing a clearer causal interpretation of the relationship between managerial ability and earnings management. This approach aligns with prior studies that have used lagged variables as instruments to address endogeneity concerns (Khan et al., 2024).

We used year lagged value of managerial ability, as managers need some time to control and use their skills over the organization. The results are reported in Table 6. The significant statistics of Durbin Watson and Wu-Hausman tests direct us to rely on the results obtained by 2SLS regression. These results also support the previously obtained results, as MA and EM are positively and significantly related.

Deviating from enormously available studies on the intersection of firm-level factors and earnings management, this study explores whether individuals’ ability may influence the latter. The study finds that higher-ability managers are more likely to magnify earnings management by using their skills and talents. The study suggests that strict monitoring works better to make the managers ethical than the manager’s abilities. The study suggests hiring managers with high skills and then auditing the firms with well-reputed auditing firms. The findings of this study need to be taken practically and with caution. This study does not undermine the recruitment of abler CEOs. Rather, it stresses the recruitment of highly able CEOs and getting the firms audited by well-reputed audit firms. It means the managers may use their talent and skills to fabricate the accounting information. A quality audit can be an effective monitoring tool to restrict the managerial ethical actions. Managers with exceptional competence use their superior skills and foresight to produce an earnings stream that avoids several reporting hazards to the advantage of all shareholders.

The research enhances the theoretical landscape by challenging the assumption that higher managerial competence equates to ethical behavior. Instead, it reveals that capable managers may engage more in earnings manipulation, adding complexity to the UET by highlighting the substantial impact of individual characteristics on organizational outcomes and financial reporting.

The practical implications are significant for managerial practice and CG. While recruiting highly skilled managers is beneficial, the study underscores the necessity of stringent oversight and robust CG systems to deter earnings manipulation. Boards of directors and governance bodies should implement rigorous control systems and ensure audits by reputable firms, promoting ethical conduct alongside managerial talent. The study suggests designing incentive structures that align managerial actions with long-term shareholder interests, such as performance-based compensation linked to shareholder value creation. In addition, fostering a corporate culture that prioritizes integrity and ethical conduct can deter earnings management practices. These measures can mitigate the risk of managerial opportunism and promote sustainable value creation for shareholders

The study acknowledges limitations, such as the difficulty in quantifying management skill, which is multidimensional and unobservable. Future research could explore more accurate metrics of managerial performance and examine the impact of managerial ability on earnings management across different institutional contingencies, such as state-owned vs non-state-owned enterprises and developing vs developed economies. Recognizing the complexity of managerial decision-making, the study suggests further exploration of factors influencing executives’ decisions regarding earnings management. These factors include executive compensation structures, CG mechanisms, regulatory environments, industry dynamics and individual characteristics such as risk appetite, ethical orientation and tenure. Future research could delve deeper into these dimensions to uncover the underlying mechanisms driving executives’ decisions and identify types of executives more predisposed to engage in earnings management.