Assessing the Risk of Earnings Management Through the Lens of Individual Moral Philosophy: Insights from Accounting Professionals

Abstract

This study explores how individual moral philosophies influence accountants’ ethical perceptions of earnings management risk, addressing the broader question of how moral reasoning interacts with the cultural environment in shaping financial reporting decisions. Although accounting standards such as IFRS/IAS aim to harmonize reporting, cultural, and institutional factors can lead professionals to interpret and apply them differently, making ethical perceptions context-dependent. Building on positive accounting theory and Forsyth’s model of personal moral philosophy, we conducted a scenario-based survey among Polish accounting professionals, using an extended set of earnings management scenarios developed by Bruns and Merchant and modified by Jooste. Our results indicate that subjectivists demonstrate greater ethical sensitivity to earnings-altering behavior, while absolutists exhibit the least. We also examined ethical evaluations across different types of earnings management practices, including income-increasing versus income-decreasing, accrual-based versus real earnings management, and multi-year versus single-year manipulations. Understanding how different moral orientations influence the perception of managerial interventions in reported figures can help executives foster an organizational culture that promotes the provision of reliable and accurate information to stakeholders. Study limitations include sample size and scope, suggesting the need for future research incorporating broader demographics and contextual variables.

1. Introduction

The influence of social and cultural factors on accounting practices is widely recognized in contemporary research. This is particularly evident in the positive accounting theory stream, which focuses on describing and explaining actual phenomena as well as predicting behaviors and decisions of individuals involved in preparing, disseminating, and using financial reports. As Watts and Zimmerman (1986) argue, positive accounting theory encompasses a diverse range of empirical topics, including earnings management, lobbying for accounting standards, disclosure effects of price changes, and performance evaluation of entities. The accounting system itself depends on institutions that create, support, and oversee it; these institutions, in turn, are influenced by culture, which is embodied by the people working within them (Hofstede 1980; Neu et al. 1998). Therefore, since ethics is a component of organizational culture grounded in moral values, the role of individual moral beliefs in decision-making processes may constitute a significant factor shaping the attribution of responsibility for unethical behavior in organizations. Furthermore, such beliefs can affect individuals’ perceptions and ethical evaluations of problematic situations arising in business contexts, such as intentional earnings management (Treviño 1986; Valentine and Barnett 2003).

As widely emphasized, earnings management behavior is described as the implementation of practices that result in not reporting the level of financial performance known to management and that would be reported in the issuer’s financial statements under normal circumstances (i.e., in the absence of pre-defined financial targets achievable through creative interpretation of accounting principles, flexible approaches to the processes of recording, documenting, processing, and presenting economic events, as well as structuring business operations) (Dechow et al. 1995; Roychowdhury 2006). It should be noted that in the analysis of motivations for earnings management in an economic entity, various approaches to assessing the transparency of the relationship between financial statements and the actual performance generated by the enterprise can be distinguished (Ronen and Yaari 2008). Some consider operations that affect the reported profit (loss) as practices that not only do not undermine the reliability of financial statements but even enhance their quality by signaling the influence of certain factors on future value creation, thus taking the form of beneficial earnings management. Under certain conditions, earnings management may also act as a mechanism to mitigate contractual imperfections and reduce information asymmetry among corporate stakeholders (Arya et al. 1998). Sunder (2005) further points out that in environments with flexible accounting regulations, earnings management is more often perceived as an adaptive business strategy rather than manipulation of reported data. He also notes that social and institutional factors shape how such practices are evaluated, potentially redefining what is deemed legitimate or unethical. Other authors, however, propose equating earnings management strictly with opportunistic actions aimed at obtaining private benefits by management at the expense of deteriorating the quality of reported earnings or prioritizing short-term over long-term goals, thus becoming pernicious earnings management (Jensen 2005; Roychowdhury 2006). Despite its differing interpretations, earnings management inherently involves significant risks, including reputational damage, legal consequences, and increased uncertainty for stakeholders.

Although the influence of social and cultural factors on accounting practices has been widely acknowledged, it is often noted that the mere implementation of harmonized regulations such as IFRS or US GAAP does not necessarily ensure the comparability of financial reporting across countries. Cultural differences may lead accountants from different environments to apply the same standards in different ways (Jarne-Jarne et al. 2022), The accounting system itself depends on the institutions that create, support, and supervise it, and these institutions are embedded in culture, transmitted by the people working within them. Dimensions such as professionalism versus statutory control, uniformity versus flexibility, secrecy versus transparency, and conservatism versus optimism are likely to shape both the level and the perception of earnings management practices (Doupnik 2008; Zhang et al. 2023). However, the empirical findings in this area remain inconclusive (Grabiński and Wójtowicz 2019). The aforementioned issues raise an important and underexplored question: to what extent does the cultural environment—and, more specifically, the individual moral beliefs of accounting professionals—shape the ethical perception of earnings management? To address this gap, the present study focuses on Poland as a representative emerging economy in CEE, where the influence of cultural and moral factors on financial reporting practices has received limited scholarly attention.

This paper investigates how individual moral philosophies influence the perception and assessment of the risk associated with various earnings management practices among accounting professionals employed in Polish enterprises. Special attention is given to four moral orientations widely recognized in business ethics research: absolutism, situationism, exceptionism, and subjectivism (Chudzicka-Czupała 2013; Treviño and Nelson 2017). By contrasting the views of individuals with strong beliefs in universal moral norms and those without, the study aims to reveal differences in how these groups perceive the business risk embedded in earnings management strategies. Furthermore, the research explores how representatives of the mentioned orientations perceive various types of earnings manipulation, taking into account their classification, financial magnitude, and the impact of such behavior on reported profit or loss in the company.

This study contributes to the existing literature in several important ways. First, prior studies that have integrated a scenario-based earnings management framework with Forsyth’s (1980) model of personal moral philosophy are scarce and have either focused on developed economies (e.g., Barnett et al. 1998; Elias 2002) or relied on student samples rather than professional practitioners (Cygańska and Bartoszewicz 2024). Thus, to the best of the authors’ knowledge, this is the first study to apply such a combined approach to accounting professionals in the context of an emerging economy. Second, in addition to the Bruns and Merchant (1989) framework, the study incorporates an extended set of earnings management scenarios developed by Jooste (2010), further underscoring the originality and comprehensiveness of the research. Third, this study explicitly examines perceptions of different types of earnings management, including downward versus upward manipulations, real versus accrual-based practices, and practices affecting either multiple fiscal years or only the current fiscal year, providing a more nuanced understanding of ethical judgments across various earnings management contexts. Fourth, existing empirical studies suggest that the evaluation of risk and the perceived social acceptability of earnings manipulation remain insufficiently explored in real-world professional environments (Kaplan 2001; Geiger and Smith 2010). By addressing these gaps, this research provides novel evidence on how individual moral philosophies influence the ethical assessment of earnings management practices in an under-researched cultural and economic setting.

2. Theoretical Background

2.1. Earnings Management in the Accounting Framework

Despite the dynamic development of various communication channels with external stakeholders, the annual report produced within the accounting system remains a fundamental instrument employed by enterprises to convey financial information. Data presented in financial statements should be free from bias and not favor any party participating in the communication process. They are expected to offer an objective view of the company’s economic performance, regardless of competing interests. However, as Sunder (2005) argues, neutrality in financial reporting is to some extent socially constructed—shaped by consensus that evolves in response to changing expectations and market demands. Consequently, the adopted reporting methods reflect the needs of the socio-economic context in which the company operates. Accounting, functioning as a system of social norms, also influences which forms of earnings management are considered acceptable and which are not, often causing the distinction between permissible and unethical earnings manipulation to become blurred (Bartov et al. 2002; Galek and Čičak 2025).

Historically, the concept of earnings management was predominantly framed in pernicious terms—understood as the intentional and deliberate manipulation of the financial reporting process aimed at securing private benefits for company managers, or as actions intended to mislead shareholders and other stakeholders involved in contracts and transactions (Healy and Wahlen 1999). However, over time, scholars and practitioners began to recognize that not all earnings management is necessarily fraudulent or harmful. This evolving perspective is reflected, among others, in the study by Fields et al. (2001), who argue that earnings management occurs when managers influence financial disclosures—with or without constraints—to either enhance the informational value of reported earnings and increase firm value or to serve managerial utility. From the perspective of the revelation principle, Arya et al. (1998) contend that earnings management can play a constructive role by enhancing a firm’s credibility in the eyes of capital providers and facilitating access to external financing. Moreover, by mitigating excessive earnings volatility—which might otherwise be misinterpreted by the market as a sign of financial instability—it may reduce the likelihood of adverse market reactions. In a subsequent contribution, Arya et al. (2003) further argue that earnings management can be not only justifiable but also essential for improving the predictability of future earnings and supporting a firm’s financial liquidity. It should be noted, however, that the literature to date does not provide a universally accepted definition of earnings management. The interpretations outlined above represent only a selection of the various perspectives proposed within academic discourse.

A similar situation can be observed with regard to the classification of earnings management behaviors. The taxonomy of methods may be developed based on the consequences of manipulation (Ali and Bansal 2023), the strategic approaches employed to achieve specific objectives (Agustia et al. 2020), or the timing of such activities—distinguishing between actions taken prior to and following the balance sheet date (Roychowdhury 2006), as well as the nature of their impact on specific items within financial statements (Bojan and Lungu 2025). From the perspective of this study, the most relevant classification is based on the nature of the methods utilized, distinguishing between accrual-based earnings management and real earnings management. The first involves the use of discretion in financial reporting and interpretation of accounting standards without affecting the company’s cash flows. Common examples include changes in the estimation of bad debt allowances or the timing of revenue recognition (Cohen et al. 2008). In contrast, real earnings management is based on structuring business operations in ways that allow firms to pre-design the consequences of transactions so they are reflected in the financial statements in a desirable manner. Such practices—e.g., accelerating discretionary expenses or overproduction—may prove inefficient from a long-term perspective, yet they can generate the desired level of earnings in the short term (Piosik and Genge 2020).

2.2. Key Factors Influencing Stakeholders’ Perceptions of Earnings Management

Prior studies have shown that a wide array of factors can influence the degree of acceptance of earnings management practices in the business environment (Sunder 2005; Habbash and Alghamdi 2015). These determinants include, inter alia, institutional conditions of organizational activity, country-specific characteristics, organizational objectives, the nature of earnings management techniques applied, as well as individual personality traits.

Lewellyn and Bao (2017) emphasize that culture and corruption shape social and ethical norms, indirectly influencing how earnings management practices are perceived and accepted in different countries—for example, in cultures with high power distance or high corruption, manipulations tend to be more tolerated. Geiger et al. (2006) found that perceptions of earnings manipulation relate to a society’s Power Distance Index (acceptance of unequal power distribution) and Masculinity Index (focus on assertiveness and competition versus cooperation and modesty). Chen et al. (2021) highlight how media coverage, shaped by cultural factors, varies in intensity and tone across countries and affects earnings management perception. Their study shows that stronger, more focused media oversight leads to managers being less likely to manipulate earnings and investors perceiving such practices as less acceptable and feasible.

One of the most frequently addressed topics in the literature is the ethical assessment of earnings management from the perspective of the respondent’s profile, including their professional experience, occupational position, and individual exposure to earnings management practices. Comporek (2024) revealed that individuals representing companies engaged in accounting discretion and real activity manipulations were more likely to perceive such practices as ethically acceptable compared to those from companies not involved in earnings management. This suggests that personal exposure to earnings manipulation may shape more lenient moral judgments toward its use. At the same time, regardless of which subpopulation respondents belonged to, the majority emphasized that earnings management misleads users of financial statements. This indicates a broad consensus about the negative impact of earnings management on the transparency of company-reported data.

An ambiguous view on the issue of earnings management is also presented by Baskaran et al. (2020). These researchers found that earnings management can be viewed both as a strategic adjustment to market demands and stakeholder expectations, and as deliberate manipulation aimed at distorting the company’s financial picture. As a result, the perception of earnings management oscillates between seeing it as a tool for strategic management and as an unethical practice, influencing how different stakeholder groups evaluate and respond to earnings management practices.

Perceptions of earnings management may also vary depending on how closely such practices are aligned with an organization’s broader strategic objectives. Sugiarto and Indrajati (2020) found that the perception of earnings management can have a positive impact on the quality of accounting information, suggesting that company stakeholders may recognize its potential role in enhancing reporting quality. This finding implies that, when aligned with stakeholder interests, earnings management behavior may contribute to more useful and reliable financial disclosures. In a similar vein, Johnson et al. (2012) demonstrate that managerial ethical attitudes towards earnings management largely depend on the specific methods of earnings altering utilized within the company, as well as the broader circumstantial context in which such practices are implemented. Their study shows that certain earnings alteration behaviors are considered more ethically justifiable when perceived to align with organizational objectives. At the same time, managers remain aware of the inherent risks associated with earnings management, including the potential deterioration of stakeholder trust.

Kaplan (2001), in his study emphasizing the importance of the respondent’s role within the company, conducted a survey among two groups—managers and shareholders—focusing on their ethical views regarding three earnings management practices: postponing discretionary expenses, delaying payment of a consulting firm’s invoice, and artificially increasing losses by overstating the obsolescence reserve. The results showed that managers tended to view earnings management through real transactions as more ethically acceptable compared to accrual-based accounting manipulations, whereas shareholders regarded both approaches as unequivocally unethical. Similar observations were made by Ismail and Atik (2011), who examined differences in ethical perceptions between those preparing financial statements and users of financial information across various earnings manipulation scenarios. Their findings highlight significant variations in how these two groups of stakeholders judge specific earnings management strategies—for instance, accounting officers were more opposed to income smoothing and management buyout techniques than financial analysts, while analysts viewed misleading bank activities as more unethical than preparers did.

In the context of this paper, it is worth mentioning that an important line of ethical research on earnings management was initiated by Bruns and Merchant (1989), who developed a scenario-based questionnaire identifying morally questionable actions by managers in both real and accounting-based earnings management. By focusing on evaluating specific case studies illustrating the use of particular accounting and real activities manipulation tools within organizations, these authors enabled the assessment of the acceptability and permissibility of practices affecting a company’s financial results, such as manipulating discretionary expenses, accelerated invoicing, structuring and timing coordination of business operations, sudden adoption of aggressive short-term receivables management strategies, manipulating inventory write-downs, creating and releasing fictitious reserves, and artificially accelerating production and sales at the end of the fiscal year, among others. Their study generally showed that accounting-based earnings management is perceived as significantly less ethically acceptable than real earnings management. Later, in subsequent surveys based on Bruns and Merchant’s questionnaire, respondents included both students and finance professionals (Merchant and Rockness 1994; Fischer and Rosenzweig 1995; Vladu 2015; Cygańska and Bartoszewicz 2024; Montenegro and Rodrigues 2020). In 2010, Jooste modified the Bruns and Merchant (1989) questionnaire by adding seven new scenarios of earnings manipulation. She used this extended version to examine the ethical attitudes of accounting students in South Africa.

While the literature has fairly extensively addressed structural and contextual determinants of earnings management perception in the eyes of company stakeholders, the role of individual moral philosophy—particularly as conceptualized in ethical frameworks such as Forsyth’s (1980) taxonomy—deserves separate consideration. This perspective allows for a deeper understanding of the internalized ethical standards that may influence how individuals involved in financial reporting interpret the acceptability and perceived risk of earnings management phenomenon, beyond external organizational or cultural pressures. Moreover, such ethical reasoning can evolve over time and may vary significantly across different institutional environments and market contexts.

2.3. Forsythe’s Ethical Taxonomy: An Overwiew

One of the most widely recognized frameworks in business ethics is Forsyth’s (1980) theory of ethical positions, which classifies individuals along two dimensions: idealism and relativism. Idealism assumes that making decisions that lead to good consequences is always possible and doing no wickedness. Relativism suggests no fixed moral principles; ethical principles must be stated in the situational, cultural, and personal context (Forsyth 1980, 1992). Next, Forsyth proposed four ethical ideologies based on the combination of these two dimensions, which encourage the investigation of moral conflicts. High idealism and low relativism are typical of an absolutist approach. Such individuals believe that ethical principles are essential and that conforming always pays off. In contrast, low idealism and low relativism reflect a pragmatic belief (Ford and Richardson 1994). They recognize the value of rules but allow for exceptional circumstances when rules can be violated. The third group, subjectivists, have low idealism and high relativism. They hold that moral judgment is a matter of belief and is not answerable to universal standards. The final group, situationists, score high in both relativism and idealism; they perceive the situational context as significant but focus on causing minimum harm and maximum good (Forsyth 1980; Schlenker 2008) (Figure 1).

Forsyth developed the Ethics Position Questionnaire (EPQ) to empirically test this typology, allowing researchers to determine an individual’s ethical profile. It has been widely employed in organizational research, making it feasible to examine connections between ethical ideology and workplace behavior. Empirical studies indicate that an individual’s ethical ideology influences how they analyze and perceive moral problems in organizational and business contexts. For example, individuals high in idealism tend to exhibit prosocial attitudes, higher empathy, and less tolerance for unethical actions regardless of environmental or economic pressures (Davis et al. 2001). Conversely, high relativists are more likely to view moral norms as relative and show greater tolerance for behavior inconsistent with accepted moral standards (Barnett et al. 1994; Henle et al. 2005).

At the organizational level, differences in idealism and relativism may influence employee behavior and managerial decision-making in everyday work situations. Previous findings suggest that relativism may justify unethical behaviors such as information manipulation, rule violation, or unfair competition (Henle et al. 2005). In contrast, high idealism is linked to decision-making that conforms to ethical standards, even at economic or personal costs (Forsyth 2019). Moreover, Forsyth’s ethical taxonomy has been applied to key organizational issues including ethical decision-making, corporate culture, ESG disclosures, digitalization, crisis management, and intergenerational value differences (Chudzicka-Czupała 2013; Forsyth 2019; Henle et al. 2005). The dimensions of idealism and relativism have also been found relevant in explaining managerial responses to regulatory pressures such as the Corporate Sustainability Reporting Directive (CSRD), selective disclosure of information, and ethical attitudes toward emerging organizational challenges (Forsyth et al. 2008).

2.4. Ethical Orientation and Earnings Management

Previous research shows that ethical behavior in organizations depends on many different factors—things like a person’s morals, personality, the subject of ethical judgment, the seriousness of the situation (Valentine and Barnett 2003; Mayer et al. 2013). Because of this, what drives employees’ behavior—including professionals like accountants—has become a hot topic for experts in management, psychology, sociology, and economics. Alongside the usual focus on following professional rules, recent studies point out that behavioral ethics plays a big role in explaining why accountants sometimes do not stick to their stated values or codes of conduct (Tenbrunsel and Messick 2004; Bazerman and Gino 2012). In accounting, it is important to remember that following accounting or tax regulations does not automatically rule out ethically questionable actions. This is especially true when performance targets and organizational needs take priority (Mintz and Morris 2020).

Research on accounting ethics has been gaining momentum in recent years. This is reflected in studies aimed at identifying the most influential publications shaping the development of ethical issues in accounting (Robertson 2008), pointing out emerging trends and areas where professional ethics could be further integrated, and highlighting key research gaps in the global literature (Bampton and Cowton 2013). In the context of the CEE region, studies that combine behavioral ethics with accounting have mostly focused on unethical behavior among accounting professionals, the codification of ethical standards, ethics education in accounting, and the role of ethical values within the profession (Barnett).

However, prior studies have only to a limited extent explored the relationship between individuals’ ethical attitudes and the perception of earnings management. And yet, this area seems particularly worth exploring if we want to understand why some professionals are able to justify or accept manipulative practices—even when they are fully aware of the ethical standards in place. That is precisely the gap this study aims to address—by examining how accountants’ ethical orientations shape their perceptions of earnings management practices in a scenario-based research setting.

Referring to previous findings, Barnett et al. (1998) discovered that U.S. marketers’ ethical judgments vary significantly depending on their ethical ideology. Specifically, absolutists consistently rated questionable business practices involving earnings management as more unethical than other groups. Interestingly, while they differed significantly from subjectivists and exceptionists, there were no significant differences between absolutists and situationists, who also exhibit high idealism. As the authors noted, their study highlights idealism as a stronger predictor of ethical judgment than relativism.

Elias (2002), based on a sample from the U.S. market, presents several other findings worthy of comment. First of all, the study revealed that individuals do not perceive real earnings management practices as unethical. This view was particularly prominent among certified public accountants (CPAs). Secondly, the study shows that regardless of the respondents’ background (practitioners, faculty, or students), accrual-based earnings management behaviors are generally perceived as ethically questionable, although students were more lenient in their assessments compared to the other groups. Importantly, Elias documented that individuals with a more idealistic moral orientation perceived earnings management as less ethical, whereas relativists were more inclined to accept such practices. Similarly, individuals who valued corporate social responsibility and long-term outcomes rated such actions as more unethical, in contrast to those who prioritized short-term gains.

Greenfield et al. (2008) documented—based on a sample of undergraduate students from several U.S. universities—that the mechanisms explaining an individual’s propensity to engage in and accept earnings management behaviors should account not only for structural factors but also for psychological and ethical characteristics of the individual. They revealed that personal differences in ethical orientation significantly influence both the individual’s approach and the tendency to engage in earnings management practices. Their experiment showed that individuals with higher relativism scores were more inclined to accept earnings altering practices, especially when personal benefits were involved, whereas idealists were less likely to view such actions as acceptable, regardless of their potential gains.

Marques and Azevedo-Pereira (2009), who conducted a similar study among Portuguese chartered accountants, found that ethical judgments related to earnings management scenarios did not vary significantly with respondents’ ethical ideology, suggesting that this variable may not be a strong predictor of acceptability in these types of moral dilemmas. Instead, demographic factors played a more prominent role: older accountants were more relativistic and lenient, male respondents tended to judge scenarios more strictly than females, and a higher level of education was associated with stricter ethical assessments.

In a slightly different tone, Maruszewska (2020) demonstrated that dividing participants into four ethical ideology types did not significantly explain differences in the moral evaluation of an accounting decision or in the propensity to replicate it in an experimental study conducted among Polish accounting students. She focused on accounting policy choices and discretion and used a scenario in which participants decided whether to replicate a questionable omission of year-end foreign currency revaluation. The findings revealed that ethical ideology had limited explanatory power, while many participants showed a tendency to conform to prior unethical behavior, often justifying it through formal compliance with accounting rules rather than ethical reflection, with only absolutists showing greater resistance to unethical replication. Finally, Cygańska and Bartoszewicz (2024) found no evidence that idealism or relativism had a statistically significant impact on the ethical evaluation of earnings management behavior among accounting students in Poland. However, they highlight that individuals classified as absolutists were, as a rule, more critical of earnings-altering practices than other groups.

2.5. Hypotheses Development

Based on the theoretical framework and previous research findings, we formulated one general (H1) and three specific hypotheses (H2–H4). All relate to ethical perceptions of earnings management, considering how associated reputational and operational risks shape individuals’ judgments across the ethical ideology types they follow:

H1. 

Ethical perceptions of earnings management practices differ significantly among individuals classified into Forsyth’s four moral philosophy categories. As mentioned earlier, prior studies (e.g., Barnett et al. 1998; Cygańska and Bartoszewicz 2024) clearly indicate that differences in moral philosophy influence the ethical assessment of accounting manipulations. However, the directions of these relationships may vary depending on the socio-cultural context.

H2. 

The ethical evaluation of accrual-based and real earnings management differs significantly and is moderated by accountants’ individual ethical ideologies. The ethical evaluation of accrual-based and real earnings management differs due to their inherent nature—accrual manipulations are often seen as a technical accounting choice, while real earnings management involves operational manipulation with potential long-term harm as well as the distortion of cash flow streams (Choi et al. 2021; Nguyen et al. 2023). Individual ethical ideologies further shape these perceptions, making some accountants more tolerant of one form over the other.

H3. 

The ethical evaluation of downward and upward earnings management differs significantly across the four moral philosophy types. The ethical evaluation of upward and downward earnings management may differ because upward earnings management is typically associated with managerial opportunism, while downward earnings management can be motivated by long-term strategic or conservative goals (Andreicovici et al. 2021; Ali and Bansal 2023). Similarly to previous assumptions, we posit that individual ethical ideologies can shape these perceptions, making some accountants more tolerant of one form over the other.

H4. 

Individuals’ moral philosophy type has a statistically significant effect on the perceived ethical acceptability of earnings management practices that impact either more than one fiscal year or only the current fiscal year. Short-term and long-term earnings management differ in their consequences. Short-term practices typically target immediate goals with limited lasting impact, while long-term manipulation can distort financial performance over multiple periods, misleading stakeholders. Consequently, individuals may judge long-term earnings management as more ethically problematic than short-term practices due to its potential to obscure sustained performance and increase risks (Garel 2017; Espahbodi et al. 2022).

3. Methodology

3.1. Research Design Overview

To address the research objectives outlined in this paper, we designed and implemented a three-stage empirical procedure (Figure 2). The study followed a predominantly qualitative approach, drawing on participants’ evaluations of ethically complex scenarios, supplemented by quantitative analysis to identify statistically significant patterns across moral philosophy profiles.

In the first stage, we identified participants’ moral philosophy profiles using Forsyth’s (1980) Ethics Position Questionnaire (EPQ), which measures two bipolar dimensions—idealism and relativism—and allows for the classification of respondents into one of four ethical positions: absolutists, situationists, subjectivists, or exceptionists.

In the second stage, we asked participants to evaluate the ethical acceptability of specific earnings management behaviors using a scenario-based instrument adapted from Bruns and Merchant (1989) and Jooste (2010).

In the third stage, we applied the Kruskal–Wallis test with Bonferroni correction, followed by Dunn’s post hoc comparisons, to examine whether statistically significant differences existed in the ethical judgments of earnings management practices. For the sake of completeness, it should be highlighted that the normality of the dependent variables, i.e., ethical evaluations across the earnings management scenarios, was assessed using the Kolmogorov–Smirnov test. All results were significant at p < 0.001, indicating deviations from normality. Consequently, non-parametric methods were employed for hypothesis testing.

We conducted the survey in the first quarter of 2025 in cooperation with the research agency Fieldstat Sp. z o.o., using the CAWI method (Computer-Assisted Web Interviewing). The sample consisted of 100 accounting professionals employed in Polish enterprises across diverse sectors. We used purposive sampling to ensure that all respondents held relevant roles in accounting or financial reporting. The survey was distributed electronically via a professional research panel. Participants were given a defined time frame to complete the questionnaire, allowing us to control the data collection process. Responses were collected automatically through a dedicated online system, reducing the risk of data entry errors. All participants were informed about the study’s purpose and assured of confidentiality, which encouraged honest and reliable responses. Although no formal pre-tests or expert reviews were conducted for the adapted scenarios, their content was based on well-established frameworks and prior studies in the literature.

3.2. Moral Philosophy Profiling

To identify respondents’ moral philosophy profiles, we employed the Ethics Position Questionnaire (EPQ) developed by Forsyth (1980). As previously mentioned, this psychometric instrument measures two independent dimensions of an individual’s ethical orientation: idealism and relativism. Based on the respondents’ scores along these dimensions, individuals can be classified into one of four ethical positions: absolutists (high idealism, low relativism), situationists (high idealism, high relativism), subjectivists (low idealism, high relativism), and exceptionists (low idealism, low relativism).

The EPQ consists of 20 statements—10 measuring idealism and 10 measuring relativism (

Table 1. Items of the Ethics Position Questionnaire (EPQ) Measuring Idealism and Relativism.

Idealism dimension

A person should make certain that their actions never intentionally harm another even to a small degree. Risks to another should never be tolerated, irrespective of how small the risks might be. The existence of potential harm to others is always wrong, irrespective of the benefits to be gained. One should never psychologically or physically harm another person. One should not perform an action which might in any way threaten the dignity and welfare of another individual. If an action could harm an innocent other, then it should not be done. Deciding whether or not to perform an act by balancing the positive consequences of the act against the negative consequences is immoral. The dignity and welfare of the people should be the most important concern in any society. It is never necessary to sacrifice the welfare of others. Moral actions are those which closely match ideals of the most “perfect” action.

Relativism dimension

11. People should make moral judgments based on personal feelings about the actions rather than on whether the actions conform to ethical principles. 12. What is ethical varies from one situation and society to another. 13. Moral standards should be seen as individualistic; what one person considers to be moral may be judged to be immoral by another person. 14. Different types of morality cannot be compared as to “rightness”. 15. Questions of what is ethical for everyone can never be resolved since what is moral or immoral is up to the individual. 16. Moral standards are simply personal rules that indicate how a person should behave, and are not to be applied in making judgments of others. 17. Ethical considerations in interpersonal relations are so complex that individuals should be allowed to formulate their own individual codes. 18. Rigidly codifying an ethical position that prevents certain types of actions could stand in the way of better human relations. 19. No rule concerning lying can be formulated; whether a lie is permissible or not depends upon the situation. 20. Whether a lie is judged to be moral or immoral depends upon the circumstances surrounding the action.

). Participants were asked to indicate their level of agreement with each item using a 9-point Likert scale (1 = strongly disagree, 9 = strongly agree). From a methodological standpoint, the classification of individuals into one of the four ethical positions was based on a median split procedure applied separately to the idealism and relativism scores. Respondents scoring above the sample median on each scale were categorized as “high,” while those scoring below were categorized as “low”. We employed the original version of the EPQ, following standard translation and validation procedures—including forward translation, expert review, and back-translation—to ensure semantic accuracy and conceptual equivalence across languages.

The internal consistency of the moral philosophy scale was assessed using Cronbach’s alpha. The idealism dimension (first ten items) yielded an alpha of 0.702, indicating acceptable reliability, while the relativism dimension (last ten items) showed an alpha of 0.866, reflecting good reliability. These results suggest that the scale is suitable for capturing the intended constructs in the study sample. Considering that each dimension is measured with only 10 items, the obtained Cronbach’s alpha values can be regarded as satisfactory indicators of reliability, in line with conventional thresholds for short scales (Nunnally and Bernstein 1994).

3.3. Judgment of Earnings Management Scenarios

We employed scenario-based questionnaires developed by Bruns and Merchant (1989) and Jooste (2010) to assess participants’ ethical judgments regarding earnings management practices. The first instrument comprises 13 scenarios illustrating various earnings management actions, encompassing both accrual-based and real earnings management activities (

Table 2. Earnings management scenarios used in the study.

	Bruns and Merchant (1989) Survey
No.	Scenarios
1	The division’s headquarters building was scheduled to be painted in 2025. But since profit performance was well ahead of budget in 2024, the general manager decided to have the work done in 2024. Amount: $150,000.
	This information applies to the following two questions. The general manager instructed division employees to defer all discretionary expenditures (e.g., employee travel, advertising, maintenance) into the next accounting period so that the division could meet its expected profit targets. Expected amount of deferrals: $150,000.
2	● The expenditures were postponed from February and March 2024 until April 2024 to achieve the first quarter target.
3	● The expenditures were postponed from November and December 2024 until January 2025 to achieve the annual target for the 2024 fiscal year.
4	On 15 December 2024, a clerk ordered office supplies worth $3000. The supplies were delivered on 29 December. This order was a mistake, as the general manager had previously instructed that no further discretionary expenses were to be incurred for the remainder of the fiscal year. According to the company’s accounting manual, office supplies should be recorded as an expense when delivered. When the general manager learned about the situation, he instructed the accounting department not to record the invoice until February 2025.
	This information applies to the following three questions. In September 2024, the general manager realized that the division would need strong performance in the fourth quarter to meet its budget targets. Therefore:
5	● He decided to implement a sales program offering very liberal payment terms to accelerate revenue recognition by pulling forward sales that would normally occur in the first quarter of 2025. Customers accepting delivery in the fourth quarter of 2024 would not be required to pay the invoice for 120 days.
6	● He instructed the production staff to work overtime in December 2024 so that as many orders as possible could be completed and shipped by year-end.
7	● He decided to sell certain unused assets, generating an additional income of $40,000.
	This information applies to the following two questions. At the beginning of December 2024, the general manager realized that the division would exceed its budgeted profit targets for the year. Therefore:
8	● He instructed the division controller to prepay certain expenses (e.g., hotel accommodations and exhibition costs) related to a trade fair scheduled for March 2025 and to record the payment as an expense in the 2024 fiscal year. The total amount was $60,000.
9	● He ordered the controller to increase the inventory write-down due to obsolescence. By taking a pessimistic view of future market conditions, the controller identified $700,000 worth of finished goods that, according to conservative accounting principles, should be written off. However, the GM remained confident that the inventory would still be sold at close to its full value.
	This information applies to the following two questions. In 2025, the division sold 70% of the inventory that had previously been written down due to obsolescence and a customer had indicated some interest in purchasing the remaining inventory the following year. The GM ordered the division controller to write the inventory back up to full cost. This involved a 210,000 increase in the inventory asset value and a corresponding increase in net income. The GM’s motivation for recapturing the profit was as follows:
10	● To be able to continue working on some important product development projects that might have been delayed due to budget constraints.
11	● To make budgeted profit targets.
	This information applies to the following two questions. In November 2025, the division was struggling to meet its budgeted profit targets. The GM contacted the consulting firm that had been doing work for the division and asked them not to send their invoice until the next year. The firm agreed. The estimated value of the work completed but not invoiced was as follows:
12	● $30,000.
13	● $500,000.
	Jooste (2010) survey—additional scenarios
14	In July 2024, the division manager noticed that production scrap costs were significantly exceeding the planned amounts. To avoid alarming senior management, he instructed his accountant to “hide” most of these costs in other expense accounts so that they would go “unnoticed”. During the remainder of the year, the accountant reclassified approximately $60,000 of scrap costs to other expense items. This operation had no effect on the entity’s final net income.
	The division manager came up with an idea to develop a new product. He commissioned the marketing department to conduct focus group interviews and a field study to determine the product’s market potential and asked the controller to estimate the costs of launching it on the market. The projections indicated an expected return of 22%, whereas the required profitability level based on internal assumptions was at least 35%. Still convinced of the project’s viability, the manager inflated the sales forecasts and understated the estimated costs in order to show a projected return at the required level. His superiors were occupied with other matters and did not challenge the presented projections, resulting in the new product project being approved. After some time, it turned out that
15	● After six months of sales, the product achieved a return of 22%;
16	● After six months of sales, the product achieved a return of 35%.
	At the end of December 2024, the division was struggling to meet its profit targets. The division manager instructed the production manager to expedite the shipment of customer orders so that the related revenues could be recognized within the current fiscal year. The ordered goods were in inventory, and the expected delivery time was two days. The customers’ originally agreed firm delivery dates were as follows:
17	● 3 January 2025;
18	● 14 January 2025;
19	● 28 January 2025;
20	● 12 February 2025.

). The latter instrument is a modified version of the Bruns and Merchant survey, which, in addition to the original scenarios, includes seven additional case studies.

We employed a Likert-type scale to assess participants’ evaluations of the ethical acceptability of each earnings management scenario. Respondents were asked to indicate their level of acceptance for each described action using the following scale: 1—ethically acceptable; 2—questionable (I would not report it to those involved, but it causes me discomfort); 3—minor violation (the individual should be warned to avoid repeating such conduct); 4—serious violation (the individual should be strongly reprimanded); 5—unethical (the individual should be dismissed). Similarly to the previous research stage, we used the original version of the questionnaire, adhering to rigorous standard translation and validation procedures to ensure accuracy and reliability. Certain phrases were slightly modified to ensure compliance with local accounting regulations and terminology, while preserving the ethical neutrality and intent of the original scenarios.

3.4. Statistical Analysis Based on Nonparametric Testing

In our study, we applied the Kruskal–Wallis test to assess differences in the perception of earnings management among representatives of the four ethical philosophy types. Generally, the Kruskal–Wallis H test is used to compare differences between more than two independent groups on the variables under analysis. It is a nonparametric alternative to ANOVA, utilized in cases where the assumptions of normality or homogeneity of variance cannot be met (Field 2013). This test allows for comparison of median ranks between groups without requiring assumptions about population distribution (Sheskin 2011). We performed a post hoc analysis (the Dunn test) with a Bonferroni correction (p < 0.05) to reduce the risk of Type I error resulting from multiple group comparisons (Armstrong 2014). Pairwise comparisons between groups and mean rank differences were examined post hoc.

4. Results

4.1. Basic Descriptive Statistics

The first step of the conducted empirical research involved assigning each respondent to one of the four moral philosophy types according to Forsyth’s taxonomy. We found that among the 100 participants, there were 30 situationists, 37 pragmatists, 16 absolutists, and 17 subjectivists. Thus, we revealed that among the individuals under consideration, participants who concurrently exhibit high levels of idealism and relativism (situationists) or low levels of idealism and relativism (pragmatists) dominate.

Table 3. Spearman’s Rank-Order Correlations Between Ethical Ideologies and Types of Earnings Management.

Variables	IDE	REL	BM	JOOSTE	UP	DOWN	ACC	REAL	MULTI	ONE
IDE	1
REL	0.357 **	1
BM	−0.191	0.015	1
JOOSTE	−0.236 *	0.050	0.949 **	1
UP	−0.248 *	0.078	0.802 **	0.920 **	1
DOWN	−0.179	−0.001	0.936 **	0.895 **	0.657 **	1
ACC	−0.047	−0.001	0.875 **	0.844 **	0.706 **	0.835 **	1
REAL	−0.334 **	0.059	0.822 **	0.922 **	0.909 **	0.763 **	0.589 **	1
MULTI	−0.242 *	0.040	0.889 **	0.967 **	0.906 **	0.846 **	0.809 **	0.896 **	1
ONE	−0.168	0.033	0.886 **	0.870 **	0.772 **	0.817 **	0.755 **	0.797 **	0.722 **	1

Note: IDE = Idealism; REL = Relativism; BM = Aggregate Earnings Management measure by Bruns and Merchant; JOOSTE = Aggregate Earnings Management measure by Jooste; UP = Upward Earnings Management; DOWN = Downward Earnings Management; ACC = Accrual-based Earnings Management; REAL = Real Earnings Management; MULTI = Multi-year Earnings Management; ONE = One-year Earnings Management. * p < 0.05, ** p < 0.01.

presents Spearman’s rank-order correlations between the average scores of ethical ideologies and various types of earnings management examined in the study. As we observed a very strong positive correlation between the Bruns and Merchant and Jooste measures, we decided to limit the interpretation to Jooste’s scale to avoid redundancy. Our results indicate that idealism is negatively associated with several forms of earnings management, particularly those involving real activities, multi-year strategies, and upward earnings manipulation. This suggests that individuals with higher idealistic orientations are less inclined to engage in ethically questionable financial reporting practices. On the other hand, relativism, as a rule, shows no significant associations with any type of earnings management, implying that the rejection of universal moral principles does not systematically relate to the use of such practices.

Table 4. Average Perceptions of Earnings Management Scenarios by Individuals Representing Different Ethical Ideology Types.

Scenario	Ethical Ideology Type				Pooled Sample
	Situationists (SI)	Pragmatists (PR)	Absolutists (AB)	Subjectivists (SU)
1	1.77	2.32	1.19	2.59	2.03
2	2.23	2.59	2.31	3.06	2.53
3	2.57	3.05	2.50	3.35	2.88
4	2.77	2.97	2.81	3.00	2.90
5	2.30	2.49	2.13	2.88	2.45
6	2.73	2.92	2.69	3.12	2.87
7	2.27	2.76	1.88	3.47	2.61
8	2.33	2.84	2.31	3.06	2.66
9	3.33	3.08	3.00	3.29	3.19
10	2.17	2.59	2.00	2.47	2.36
11	2.47	2.70	2.06	2.76	2.55
12	2.30	2.51	2.69	3.06	2.58
13	2.97	3.30	3.25	3.41	3.20
14	3.27	3.54	2.67	2.94	3.22
15	2.73	3.19	3.07	3.24	3.04
16	2.60	3.14	2.87	3.18	2.94
17	2.17	2.43	2.60	2.41	2.37
18	2.50	2.41	3.00	3.18	2.66
19	2.67	2.73	2.67	3.24	2.79
20	2.47	2.76	3.00	2.94	2.74

shows the mean values of responses concerning the ethicality ratings assigned to 20 earnings management scenarios by individuals representing different ethical ideology types. Using a 5-point Likert scale (1 = ethically acceptable, 5 = ethically unacceptable), we found that scenarios 1, 2, and 5 were perceived as the most ethically acceptable. In particular, scenario 1 involved shifting a planned expense forward (from 2025 to 2024) in order to reduce reported profit in an exceptionally strong financial year, receiving the lowest mean score across the pooled sample. On the other hand, scenarios 13, 14, and 15 were rated as the least acceptable, indicating that accountants perceived these earnings management behaviors as clearly unethical.

4.2. Statistical Analysis of Ethical Acceptability by Moral Philosophy

Next, we used the Kruskal–Wallis test to examine whether the ratings of ethical acceptability for individual earnings management cases differed statistically based on respondents’ endorsed moral philosophy. The null hypothesis assumed that the distribution of assigned ratings is the same across the tested subpopulations. Our analysis allowed us to reject this hypothesis, both in the context of earnings management scenarios developed by Bruns and Merchant as well as those by Jooste (

Table 5. Kruskal–Wallis Test and Pairwise Comparisons for Ethical Acceptability Ratings of General Earnings Management Practices.

All Earnings Management Practices—Bruns and Merchant scenario
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Situationists (SI)	Pragmatists (PR)	Absolutists (AB)	Subjectivists (SU)
58.60	<0.001	2.48	2.78	2.37	3.04
Pairwise comparisons
Sample 1—Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
AB—SI	37.38	31.18	1.20	0.231	1.000
AB—PR	134.66	30.14	4.47	<0.000	0.000
AB—SU	−226.60	35.09	−6.46	<0.001	0.000
SI—PR	−97.28	24.75	−3.93	<0.002	0.001
SI—SU	−189.22	30.58	−6.19	<0.003	0.000
PR—SU	−91.94	29.51	−3.12	0.002	0.011
All Earnings Management Practices—Jooste scenario
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Situationists (SI)	Pragmatists (PR)	Absolutists (AB)	Subjectivists (SU)
68.94	<0.001	2.58	2.79	2.44	3.07
Pairwise comparisons
Sample 1—Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
AB—SI	76.37	38.67	1.98	0.048	0.289
AB—PR	178.78	37.37	4.78	<0.001	0.000
AB—SU	−325.78	43.51	−7.49	<0.001	0.000
SI—PR	−102.41	30.69	−3.34	<0.001	0.005
SI—SU	−249.41	37.92	−6.58	<0.001	0.000
PR—SU	−147.00	36.60	−4.02	<0.001	0.000

). Asymptotic significances (2-sided tests) are displayed, with the significance level set at 0.050. Significance values have been adjusted by the Bonferroni correction for multiple comparisons.

In-depth empirical analyses using post hoc tests revealed that statistically significant differences appeared in five out of six pairwise comparisons in both cases. We discovered that absolutists were the most lenient in their ethical judgments of the earnings management scenarios, consistently assigning the lowest (most accepting) ratings across most cases. In contrast, subjectivists were the most critical, giving the highest ratings and thus showing the least tolerance for such practices. Situationists and pragmatists usually rated the scenarios somewhere in the middle, with pragmatists being a bit stricter than situationists.

Table 6. Kruskal–Wallis Test and Pairwise Comparisons for Ethical Acceptability Ratings of Income-Increasing and Income-Decreasing Earnings Management Practices.

Upward Earnings Management—Bruns and Merchant scenario (5+6+7+10+11)
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Situationists (SI)	Pragmatists (PR)	Absolutists (AB)	Subjectivists (SU)
29.72	<0.001	2.39	2.69	2.15	2.94
Pairwise comparisons
Sample 1—Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
AB—SI	30.94	19.33	1.60	0.109	0.656
AB—PR	69.91	18.68	3.74	<0.001	0.001
AB—SU	−104.87	21.75	−4.82	<0.001	0.000
SI—PR	−38.96	15.34	−2.54	0.011	0.067
SI—SU	−73.93	18.95	−3.90	<0.001	0.001
PR—SU	−34.96	18.29	−1.91	0.056	0.336
Upward Earnings Management—Jooste scenario (5+6+7+10+11+15+16+17+18+19+20)
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Situationists (SI)	Pragmatists (PR)	Absolutists (AB)	Subjectivists (SU)
50.28	<0.001	2.56	2.75	2.27	3.03
Pairwise comparisons
Sample 1—Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
AB—SI	82.82	28.66	2.89	0.004	0.023
AB—PR	135.17	27.70	4.88	<0.001	0.000
AB—SU	−216.25	32.24	−6.71	<0.001	0.000
SI—PR	−52.34	22.74	−2.30	0.021	0.128
SI—SU	−133.43	28.10	−4.75	<0.001	0.000
PR—SU	−81.09	27.12	−2.99	0.003	0.017
Downward Earnings Management—Bruns and Merchant scenario (1+2+3+4+8+9+12+13)
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Situationists (SI)	Pragmatists (PR)	Absolutists (AB)	Subjectivists (SU)
31.07	<0.001	2.53	2.83	2.50	3.10
Pairwise comparisons
Sample 1—Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
AB—SI	6.92	24.46	0.28	0.777	1.000
AB—PR	65.64	23.64	2.78	0.005	0.033
AB—SU	−122.78	27.53	−4.46	<0.001	0.000
SI—PR	−58.73	19.41	−3.03	0.002	0.015
SI—SU	−115.86	23.99	−4.83	<0.001	0.000
PR—SU	−60.23	24.34	−2.56	0.006	0.040
Downward Earnings Management—Jooste scenario (1+2+3+4+8+9+12+13+14)
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Situationists (SI)	Pragmatists (PR)	Absolutists (AB)	Subjectivists (SU)
23.86	<0.001	2.62	2.84	2.64	3.12
Pairwise comparisons
Sample 1—Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
AB—SI	−5.43	25.95	−0.21	0.834	1.000
AB—PR	44.28	25.08	1.77	0.078	0.465
AB—SU	−110.18	29.20	−3.77	<0.001	0.001
SI—PR	−49.70	20.60	−2.41	0.016	0.095
SI—SU	−115.61	25.45	−4.54	<0.001	0.000
PR—SU	−65.91	24.56	−2.68	0.007	0.044

shows how respondents rated the ethical acceptability of earnings management aimed at increasing or decreasing reported income. We based our analysis on two sets of scenarios—one adapted from Bruns and Merchant, the other from Jooste. In both cases, the differences between moral philosophy groups were statistically significant, suggesting that a person’s ethical outlook has a real impact on how they judge these practices. Subjectivists tended to be the most critical, while absolutists were usually the most lenient, especially when it came to income-increasing strategies. These results align with our earlier findings for the full set of earnings management examples. Interestingly, participants generally viewed income-increasing practices as more acceptable than income-decreasing ones. It is also worth noting that the number of statistically significant pairwise comparisons varied depending on which scenario set was used (Bruns and Merchant versus Jooste approach). From an instrumental point of view, it is worth highlighting that Scenario 14 (which had a neutral impact on the financial result) was ultimately classified as a case of downward earnings management.

In the next stage of our study, we carried out a similar analysis focusing on perceptions of accrual-based and real earnings management (

Table 7. Kruskal–Wallis Test and Pairwise Comparisons for Ethical Acceptability Ratings of Accrual-Based and Real Earnings Management Practices.

Accrual-Based Earnings Management—Bruns and Merchant scenario (4+8+9+10+11+12+13)
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Situationists (SI)	Pragmatists (PR)	Absolutists (AB)	Subjectivists (SU)
16.07	<0.001	2.62	2.85	2.59	3.00
Pairwise comparisons
Sample 1—Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
AB—SI	8.86	22.80	0.39	0.698	1.000
AB—PR	48.68	22.04	2.21	0.027	0.163
AB—SU	−83.62	25.66	−3.26	0.001	0.007
SI—PR	−39.82	18.10	−2.20	0.028	0.167
SI—SU	−74.76	22.36	−3.34	<0.0001	0.005
PR—SU	−34.94	21.58	−1.62	0.105	0.633
Accrual-Based Earnings Management—Jooste scenario (4+8+9+10+11+12+13+14+15+16)
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Situationists (SI)	Pragmatists (PR)	Absolutists (AB)	Subjectivists (SU)
9.24	<0.001	2.77	2.87	2.79	3.05
Pairwise comparisons
Sample 1—Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
AB—SI	−5.62	27.25	−0.21	0.837	1.000
AB—PR	18.71	26.34	0.71	0.477	1.000
AB—SU	−72.78	30.66	−2.37	0.018	0.106
SI—PR	−24.33	21.62	−1.13	0.261	1.000
SI—SU	−78.40	26.72	−2.93	0.003	0.020
PR—SU	−54.07	25.79	−2.10	0.036	0.216
Real Earnings Management—Bruns and Merchant scenario (1+2+3 5+6+7)
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Situationists (SI)	Pragmatists (PR)	Absolutists (AB)	Subjectivists (SU)
47.28	<0.001	2.31	2.69	2.11	3.08
Pairwise comparisons
Sample 1—Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
AB—SI	28.26	21.24	1.33	0.183	1.000
AB—PR	85.03	20.53	4.14	<0.001	0.000
AB—SU	−142.57	23.90	−5.97	<0.001	0.000
SI—PR	−56.77	16.86	−3.37	<0.001	0.005
SI—SU	−114.31	20.83	−5.49	<0.001	0.000
PR—SU	−57.54	20.10	−2.86	0.004	0.025
Real Earnings Management—Jooste scenario (1+2+3 5+6+7+17+18+19+20)
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Situationists (SI)	Pragmatists (PR)	Absolutists (AB)	Subjectivists (SU)
79.11	<0.001	2.40	2.71	2.08	3.09
Pairwise comparisons
Sample 1—Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
AB—SI	79.33	27.39	2.90	0.004	0.023
AB—PR	157.50	26.48	5.95	<0.001	0.000
AB—SU	−250.51	30.82	−8.13	<0.001	0.000
SI—PR	−78.17	21.74	−3.60	<0.001	0.002
SI—SU	−171.18	26.86	−6.37	<0.001	0.000
PR—SU	−93.01	25.93	−3.59	<0.001	0.002

). The results indicate that individuals’ ethical profiles influence how both accounting discretion and real activity manipulations are judged. In both types of practices, subjectivists were the most ethically restrictive, giving the lowest ethical ratings to the examined cases. However, out of six possible pairwise comparisons, we observed only two statistically significant differences in the Bruns and Merchant scenario and just one in the Jooste-based version. When it comes to real earnings management, the differences between moral philosophy types were much more pronounced. Subjectivists again stood out as the most critical, while absolutists were generally more lenient—especially in the Jooste scenario, where all six pairwise comparisons turned out to be statistically significant. This suggests that actions involving real activities may be judged more harshly—and more variably—depending on the evaluator’s ethical outlook.

In the final stage of our analysis, we examined the ethical acceptability ratings of earnings management practices that affect either multiple fiscal years or just the current fiscal year (

Table 8. Kruskal–Wallis Test and Pairwise Comparisons for Ethical Acceptability Ratings of Earnings Management Practices Affecting Multiple Fiscal Years ore the Current Fiscal Year.

Multi-Year Earnings Management—Bruns and Merchant scenario (1+3+4+5+8+10+11+12+13)
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Situationists (SI)	Pragmatists (PR)	Absolutists (AB)	Subjectivists (SU)
41.92	<0.001	2.40	2.75	2.32	2.95
Pairwise comparisons
Sample 1—Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
AB—SI	22.13	25.92	0.85	0.393	1.000
AB—PR	99.48	25.05	3.97	<0.001	0.000
AB—SU	−152.17	29.16	−5.22	<0.001	0.000
SI—PR	−77.36	20.57	−3.76	<0.001	0.001
SI—SU	−130.05	25.42	−5.12	<0.001	0.000
PR—SU	−52.69	24.53	−2.15	0.032	0.190
Multi-Year Earnings Management—Jooste scenario (1+3+4+5+8+10+11+12+13+15+16+17+18+19+20)
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Situationists (SI)	Pragmatists (PR)	Absolutists (AB)	Subjectivists (SU)
61.22	<0.001	2.52	2.77	2.35	3.01
Pairwise comparisons
Sample 1—Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
AB—SI	73.67	33.45	2.20	0.028	0.166
AB—PR	165.12	32.33	5.11	<0.001	0.000
AB—SU	−263.73	37.64	−7.01	<0.001	0.000
SI—PR	−91.45	26.55	−3.45	<0.001	0.003
SI—SU	−190.06	32.80	−5.79	<0.001	0.000
PR—SU	−98.61	31.66	−3.12	0.002	0.011
One–Year Real Earnings Management—Bruns and Merchant scenario (2+6+7+9)
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Situationists (SI)	Pragmatists (PR)	Absolutists (AB)	Subjectivists (SU)
17.78	<0.001	2.64	2.83	2.47	3.23
Pairwise comparisons
Sample 1—Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
AB—SI	16.00	17.33	0.92	0.356	1.000
AB—PR	35.54	16.75	2.12	0.034	0.203
AB—SU	−74.93	19.50	−3.84	<0.001	0.001
SI—PR	−19.55	13.75	−1.42	0.155	0.931
SI—SU	−58.93	16.99	−3.47	<0.001	0.003
PR—SU	−39.39	16.40	−2.40	0.016	0.098
One–Year Earnings Management—Jooste scenario (2+6+7+9+14)
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Situationists (SI)	Pragmatists (PR)	Absolutists (AB)	Subjectivists (SU)
11.35	0.01	2.77	2.85	2.72	3.24
Pairwise comparisons
Sample 1—Sample 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
AB—SI	3.94	19.36	0.20	0.839	1.000
AB—PR	14.11	18.71	0.75	0.451	1.000
AB—SU	−61.97	21.78	−2.85	0.004	0.027
SI—PR	−10.18	15.37	−0.66	0.508	1.000
SI—SU	−58.03	18.99	−3.06	0.002	0.013
PR—SU	−47.85	18.33	−2.61	0.009	0.054

). Although this approach can be seen as somewhat debatable, we consider it more appropriate than distinguishing between short-term and long-term earnings manipulations, which can often be difficult to assign clearly within the scenarios used. We believe this distinction is important because the annual financial report remains the primary source of information for company stakeholders.

Consistent with previous findings, we observed that individuals’ moral profiles significantly influence how these earnings management actions are perceived. The Kruskal–Wallis tests confirmed significant differences between the four moral groups across both scenario sets (Bruns and Merchant, and Jooste). Whether the managerial practices impacted the current fiscal year or other fiscal years (previous or future), subjectivists consistently judged these behaviors most strictly. On the other hand, absolutists proved to be the most permissive in their ethical evaluations of the examined earnings management practices. Moreover, we found a greater number of statistically significant between-group differences in perceptions of multi-year earnings management practices compared to one-year practices.

4.3. Robustness Check Using Cluster Analysis

To verify the robustness of the ethical profiles originally derived via median splits on idealism and relativism scores, we conducted an additional cluster analysis using the k-means clustering method. Although the four-cluster assignments were data-driven, the final cluster centers corresponded closely to Forsyth’s conceptual types (

Table 9. Comparison of Cluster Profiles with Forsyth’s Ethical Ideology Types.

Cluster	Average Idealism	Average Relativism	Closest Forsyth Type	Interpretation
1	7.62	6.53	Situationist	High idealism and high relativism
2	6.04	4.24	Pragmatist	Moderate idealism and low relativism
3	7.86	3.1	Absolutist	High idealism and low relativism
4	6.25	6.01	Subjectivist	Moderate idealism and high relativism

). We acknowledge that in Cluster 4, the subjectivist profile appears relatively close to the situationist one in terms of average idealism scores. However, the clearly higher relativism values observed in this cluster justify its interpretation as predominantly subjectivist in nature.

We decided to limit the robustness check only to the overall indices of earnings management practices measured using the Bruns and Merchant and Jooste methodologies, in order to avoid redundancy. We assume that the strong convergence of patterns observed across specific earnings management subtypes in our main analyses would likely be replicated at the disaggregated level as well. Thus, the robustness check focused on the general trend rather than repeating all previous comparisons.

Next, we ran a Kruskal–Wallis test to see if accountants’ views on the ethical acceptability of earnings management differ between the clusters we identified with k-means. The obtained results (

Table 10. Kruskal–Wallis Test and Pairwise Comparisons for Ethical Acceptability Ratings of General Earnings Management Practices Across Cluster-Based Ethical Profiles.

All Earnings Management Practices—Bruns and Merchant scenario
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Cluster 1	Cluster 2	Cluster 3	Cluster 4
26.65	<0.001	2.51	2.74	2.50	2.89
Pairwise comparisons
Cluster 1—Cluster 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
3—1	10.83	44.71	0.24	0.809	1.000
3—2	91.27	42.76	2.13	0.033	0.197
3—4	−146.92	44.35	−3.31	<0.001	0.006
1—2	−80.43	26.53	−3.03	0.002	0.015
1—4	−136.08	29.02	−4.68	<0.001	0.000
2—4	−55.65	25.91	−2.14	0.032	0.191
All Earnings Management Practices—Jooste scenario
Kruskal–Wallis test		Mean values of responses
Test Statistic	Asymptotic Sig.	Cluster 1	Cluster 2	Cluster 3	Cluster 4
41.99	<0.001	2.56	2.74	2.39	2.91
Pairwise comparisons
Cluster 1—Cluster 2	Test Statistic	Std. Error	Std. Test Statistic	Sig.	Adj. Sig.
3—1	99.67	53.41	1.87	0.062	0.372
3—2	189.28	51.08	3.71	<0.001	0.001
3—4	−278.86	52.98	−5.26	<0.001	0.000
1—2	−89.61	31.69	−2.83	0.005	0.028
1—4	−179.18	34.67	−5.17	<0.001	0.000
2—4	−89.58	30.96	−2.89	0.004	0.023

) confirmed our previous observations: the distribution of earnings management ratings across the extracted clusters is significantly different. These outcomes support the robustness of our initial typological distinctions and indicate that participants’ moral philosophy—regardless of the classification method—remains a meaningful predictor of attitudes toward earnings management.

5. Discussion

Our research provides several noteworthy findings in the field of business ethics in contemporary accounting, offering both confirmatory and novel insights. First, our findings indicate that pragmatists, alongside situationists, constitute the largest ideological groups in our sample of professional accountants, which contrasts with previous insights reported by Maruszewska (2020) and Cygańska and Bartoszewicz (2024), who identified situationists as the predominant group among accounting students. Thus, the distribution of moral philosophy profiles in our study differs not only from earlier Polish student-based research but also from prior Anglo-American studies (e.g., Barnett et al. 1998). We assume that these differences may stem from factors such as varying levels of professional experience among participants, as well as shifting cultural and institutional contexts in the CEE region, where ethical reasoning often takes context into account and emphasizes compromise.

One potential explanation lies in our analytical strategy. The reliance on a median split as the cut-off point in the EPQ evaluation, as well as the choice of analytical level (respondent-level averages), may have influenced the empirical findings. Moreover, we noticed that although Forsyth’s model suggests that absolutists—those high in idealism and low in relativism—should be the strictest in their moral evaluations, our results showed the opposite: they were actually the most lenient in judging earnings management cases. We know that absolutists usually focus on avoiding harm to others, but many of the earnings management practices used in our scenarios—like deferring costs or applying accruals—may not have appeared clearly harmful. What is more, some of them might have even seemed helpful for the company’s financial stability. It is also possible that absolutists—especially in a professional setting—tend to place more trust in formal rules or organizational norms and may simply view earnings management as just a regular part of business operations. Subjectivists, on the other hand, do not follow universal moral rules or strong idealistic beliefs—but that does not mean they do not care. Thus, they may have judged the very same practices more harshly precisely because they are not bound by formal systems—instead, they rely on their own sense of right and wrong. Eventually, the previous literature has documented that the direction of links between broadly understood social and cultural factors and the earnings management phenomenon can vary across markets, regions, and societies (Paredes and Wheatley 2017; Ayub Khan and Smith Law 2018). Perhaps this also partly explains why the observed differences in our findings occurred.

In comparison with prior studies on U.S. accountants (Fischer and Rosenzweig 1995), our respondents tended to judge the same earnings management scenarios (drawn from the approach by Burns and Merchant) more critically. Specifically, managerial interventions involving accrual manipulation received higher ethical severity ratings in our sample compared to earlier studies, indicating that Polish professionals may be more sensitive to such practices. Similarly, some real earnings management scenarios were judged more harshly by our respondents than by their U.S. counterparts.

In our study, we positively confirmed the first research hypothesis, stating that ethical perceptions of earnings management practices differ significantly among individuals classified into Forsyth’s four moral philosophy categories. However, contrary to Barnett et al. (1998), who found absolutists to be the most ethically conservative, we noted that our respondents classified as subjectivists were the most ethically critical, while absolutists were the most lenient. Our findings also differ from those of Cygańska and Bartoszewicz (2024), who reported that the absolutist group was the most restrictive, judging accounting manipulation as the least ethical behavior, while situationists were significantly more stringent regarding operational manipulation compared to other groups.

The second research hypothesis, stating that the ethical evaluation of accrual-based and real earnings management differs significantly and is moderated by accountants’ individual ethical ideologies, was partially confirmed. In other words, although Kruskal–Wallis tests showed statistically significant differences in the perception of both types of earnings management across moral philosophy groups, these differences were widespread only in the case of real activities manipulation. Regarding the accounting-type earnings management, pairwise comparisons revealed only a few statistically significant differences in this respect.

Moreover, our study shows that ethical judgments varied not only between particular earnings management scenarios but also based on whether income was increased or decreased. Downward earnings manipulation practices were perceived as less acceptable by representatives of all four subpopulations. This aligns with Greenfield et al. (2008), who noted that ethical judgments often hinge on the perceived intent and outcome of a decision—especially if it appears to cause unnecessary harm or undermine transparency. Thus, we confirm the third hypothesis that the ethical evaluation of downward and upward earnings management differs significantly across individuals’ moral philosophy types.

Finally, we believe that the issue of perceiving earnings management from the perspective of temporal scope introduces an underexplored dimension of bottom-line altering. We observed greater ethical sensitivity and between-group variation in multi-year earnings management scenarios compared to single-year ones. Thus, our results enabled us to positively verify the fourth hypothesis, claiming that individuals’ moral philosophy type has a statistically significant effect on the perceived ethical acceptability of earnings management practices that impact either more than one fiscal year or only the one (current) fiscal year.

6. Conclusions

This paper adds to the ongoing conversation around business ethics in today’s organizations. Specifically, it addresses how accounting professionals perceive the ethics of earnings management practices, aiming to examine whether their personal moral philosophy influences these perceptions. To achieve this, we conducted a scenario-based study among professional accountants in a Central and Eastern European market, combining the original Bruns and Merchant (1989) scenarios with the modified Jooste (2010) framework and integrating Forsyth’s (1980) moral philosophy dimensions. In doing so, we examine the existing gap regarding the role of cultural and moral factors in shaping financial reporting practices in the Polish market.

Our findings show that the way accounting professionals assess the ethics of earnings management practices is strongly shaped by their personal moral philosophy. In particular, we observed clear differences between subjectivists and absolutists in how they view both the risk and ethical acceptability of such behavior. What makes this especially interesting is that, according to existing literature, subjectivists are typically seen as those who reject fixed moral rules and base their judgments on personal feelings and the specific context. Absolutists, on the other hand, tend to evaluate actions through the lens of universal ethical standards, looking for outcomes that do not harm—and ideally benefit—everyone involved.

We are aware that this study is not without its limitations. The most evident constraint is the relatively small sample size, drawn from a single national context. Moreover, the use of scenario-based judgment, while helpful in standardizing responses, may not fully capture the complexity of real-life decision-making in accounting practice. Nevertheless, we believe our research offers a valuable—and in some respects, novel—contribution. It adapts an empirical framework to a sample of professional accountants from one of the CEE markets and combines both the original Bruns and Merchant scenarios with the modified Jooste version. In our view, this represents a meaningful step toward understanding how moral philosophy influences ethical decision-making in the accounting profession. From a practical perspective, our findings suggest that accounting firms and regulatory bodies should consider the moral profiles of their employees when designing internal control mechanisms. Tailored programs can help absolutists and idealists better cope with the challenges of long-term and multi-year earnings management, while subjectivists and relativists may benefit from guidance that highlights the broader impact of their decisions on stakeholders. Furthermore, understanding how different moral orientations influence the perception of managerial interventions in reported figures can assist executives in fostering an organizational culture that promotes the provision of reliable and accurate information to the company’s stakeholders.

Looking ahead, future studies could take into account additional variables related to respondents’ profiles—such as gender, age, job position, professional experience, or direct exposure to earnings management practices. It could also be useful to explore whether the financial materiality of earnings manipulation influences how individuals with different moral profiles respond to it. Finally, we suggest extending this research to include representatives of other business sectors and regulatory bodies, as applying the same tools across diverse stakeholder groups could provide a broader perspective on both the ethical perception of earnings management and the associated risk.

Finally, future studies could address factors beyond the scope of this paper, particularly the role of moral hazard in the agency context. The temptation for employees to deviate from owners’ policies or programs may disrupt the intended outcomes, influencing the ethical perception and practice of earnings management. Examining how such incentives interact with personal moral philosophy, cognitive styles, and evaluative judgments could provide a more comprehensive understanding of the mechanisms behind ethical decision-making in accounting and clarify the conditions under which agency problems may lead to higher or lower engagement in earnings management.