The influence of culture on accounting judgment – Evidence from Poland and the United Kingdom

Abstract

This study investigates whether national culture influences judgments made by Polish and British accounting professionals when interpreting selected International Financial Reporting Standards (IFRS) that contain uncertainty expressions. We provide evidence from the Central and Eastern European (CEE) region, which is under-investigated in this stream of research. The motivation for this research was provided by the inference that IFRS may not be suitable in Poland because it is a country with significant differences from the profile that best suits application of IFRS. We explicitly test Hofstede-Gray’s theoretical framework with respect to conservatism and secrecy values at the individual level. To illustrate core cultural differences (especially in patterns of thinking) between the examined countries, we do not solely rely on the Hofstede theory, but also draw on sociological and historical literature. We do not confirm that national culture, at least based on Hofstede—Gray framework, influences interpretation of uncertainty expressions that require accounting judgment. We find no support neither for Gray’s conservatism nor for secrecy hypotheses. This suggests that, in the era of IFRS convergence in the European Union, Hofstede-Gray’s framework may have lost its explanatory power in cross-cultural research or is simply not suitable for explaining cultural differences. Additionally, the data analysis reveals that apart from work experience and statutory auditor or chartered accountant qualification, the other personal and professional characteristics of accountants controlled for in our study do not influence their judgments.

Keywords:

• accounting judgments
• culture
• conservatism
• secrecy
• IFRS
• interpretations

JEL classification:

• M41

PUBLIC INTEREST STATEMENT

It is generally accepted that comparability of financial reports may be best achieved through compliance with IFRS. However, the application of IFRS alone does not automatically unify the solutions adopted by various countries. The IFRS provide accounting professionals with flexibility in applying the standards due to discretion in their interpretation which might be different across nations. Prior studies have demonstrated that these differences might be influenced by culture. Our study analyzes the effect of national culture on judgments made by Polish and British accounting professionals when interpreting selected IFRS.

We compare two distinct European cultures with the view to understand their similarities and differences based on sociological and historical literature and the popular Hofstede-Gray cultural accounting model. However, our study suggests that national culture, at least based on the Hofstede-Gray framework, may have lost its explanatory power in the context of discretionary decisions under IFRS.

1. Introduction

Comparability of financial information is the major goal for both the EU Commission and the International Accounting Standards Board (IFRS, 2020). However, adopting International Financial Reporting Standards (IFRS) in a similar manner across nations does not ensure consistent understanding and application of these standards (Cieslewicz, 2014; Guermazi & Halioui, 2020; Huerta et al., 2016; Zeghal & Lahmar, 2018). IFRS provide accounting professionals with flexibility in applying the standards due to discretion in their interpretation (Ball et al., 2015; Florou et al., 2017; Ugrin et al., 2017; Wehrfritz & Haller, 2014), as they include a number of uncertainty expressions that are subjective in nature.

There are two categories of these terms: verbal uncertainty expressions (e.g., “probable”, “reasonable assurance”, “likely”) and numerical uncertainty expressions (e.g., a 50% chance) (Zhang et al., 2020). As Piercey (2009) emphasizes, verbal uncertainty expressions are more justifiable than numerical expressions. Thus, in the accounting context, the meanings of these terms are subject to professional interpretation and judgment (Doupnik & Richter, 2003; Zhang et al., 2019). Some expressions such as “probable” relate to judgments on an accounting recognition (for example, an asset in IAS 16) or disclosure (for example, contingent liability disclosures in IAS 37). Both accountants and auditors need to understand the meaning of the uncertainty expressions used in IFRS, including how they represent the level of uncertainty and, if applicable, how the numerical uncertainty is assessed (Juanchich et al., 2012). Due to the fact that they cause concerns about information quality and communication efficiency, accounting scholars have critically assessed their potential impacts in financial reporting, as well as in the context of judgment and decision-making behavior (Zhang et al., 2019, 2020). Therefore the use of uncertainty expressions creates significant challenges in achieving consistent accounting judgment (Chand et al., 2012; Erb & Pelger, 2015).

Many prior cross-cultural studies have demonstrated that applying professional judgment in accounting is a function of various factors, including culture (Chand et al., 2012; Doupnik & Riccio, 2006; Doupnik & Richter, 2003, 2004; Fearnley & Gray, 2015; Gray et al., 2015; Ying & Patel, 2016). In view of the fact that the vast majority of countries that have adopted IFRS have diverse national cultures, it is likely that IFRS adoption will not automatically change how people in a nation think. Differences in accounting judgment across countries occur because judgments “are not made by automatons” (Gupta et al., 2019, p. 608) but by accounting professionals, who “whether aware of it or not, are influenced by culture and their behavior reflects this” (Cieslewicz, 2014, p. 513). In this regard, the importance of the effect of national culture on those judgments cannot be ignored, especially in view of IFRS, which have been developed for worldwide usage (Perera et al., 2012).

In this study, our aim is to analyze the role of national culture in explaining the consequences of IFRS adoption. More specifically, we are interested in analyzing whether national culture influences the numerical probability assigned by professional accountants to uncertainty expressions contained in IFRS. We focus on uncertainty expressions such as “probable” and “remote” that are used as recognition or disclosure thresholds in selected IFRS standards.

Several studies have considered the impact of national culture on the professional judgments of individuals (Chand et al., 2012; Doupnik & Riccio, 2006; Doupnik & Richter, 2004; Eljammi Ayadi et al., 2020; Gupta et al., 2019; Hu et al., 2013; Schultz & Lopez, 2001; Tsakumis, 2007; Wehrfritz & Haller, 2014). These studies have almost exclusively adopted Gray’s framework, which relies on Hofstede’s cultural values. While these studies demonstrate the importance of national culture, their insights may be limited because “cultural values of individuals can change” (Hu et al., 2013, p. 52), thus culture cannot be considered as a static, stable phenomenon (Kirkman et al., 2017). Moreover, the results of these studies are mixed, indicating the issue needs further investigation. While some studies have found evidence that national culture affects the interpretation of uncertainty expressions in the context of accounting disclosures (Chand et al., 2012; Doupnik & Riccio, 2006; Eljammi Ayadi et al., 2020; Tsakumis, 2007), others were unable to find support for this hypothesis (Wehrfritz & Haller, 2014). For judgments on an accounting recognition, the researchers found a partial (Chand et al., 2012; Doupnik & Riccio, 2006; Doupnik & Richter, 2004; Eljammi Ayadi et al., 2020) or no link between national culture and the interpretation of uncertainty terms (Tsakumis, 2007).

We use questionnaire surveys to empirically test the extent and cause of differences between the judgments of Polish and British accountants and auditors. There is a clear distinction between the two countries chosen for our study. In contrast to the UK, Poland is considered as an example of the Continental European model of accounting (C. N Albu et al., 2014) and, similar to other Central and Eastern European (CEE) countries, faces several problems with applying IFRS, such as their traditional rule-oriented approach and the importance of taxation (Irvine, 2008; Jermakowicz & Gornik-Tomaszewski, 2006; Larson & Street, 2004). Firms in Continental European countries are generally expected to report more conservatively than firms in the Anglo-Saxon countries (Hartmann et al., 2020). Moreover, Poland, as a representative of the CEE countries, has a similar history, including its transition to a market economy.

Our study contributes to the existing literature in several important ways. First, we provide evidence from the CEE region, which is under-investigated in this stream of research (Albu et al., 2017, 2020). Studies that include CEE countries focus mainly on the difficulties of adopting IFRS from an institutional perspective (Albu et al., 2017; Vellam, 2004). However, this region is increasingly important from an economic and strategic perspective (Albu et al., 2017). Many leading Western companies now operate on a global basis, with subsidiaries in all important regions, including CEE countries (Ovidiu, 2011). While a number of the prior studies that consider the influence of national culture on accounting judgment have primarily been undertaken in the U.S. (i.e. Doupnik & Riccio, 2006; Doupnik & Richter, 2004; Gupta et al., 2019; Tsakumis, 2007), Australia, and China (Chand et al., 2012; Hu et al., 2013), only Wehrfritz and Haller (2014) relate this issue solely to the European context. Moreover, apart from Kosmala (2007), who focuses on the impact of culture on audit practice and on construction of auditor independence, no such research has previously been undertaken in Poland. In Poland, Gierusz et al. (2019a) provide evidence on inconsistent interpretation of the uncertainty expressions occurring in IFRSs among the Polish accounting professionals. Similarly, Gierusz et al. (2019b) find that the Polish accounting professionals interpret these expressions differently, in comparison to their British counterparts. The most significant differences were recorded for the expressions the Polish translation of which did not convey the very meaning of the phrases used in the original—English—version of the standards. Moreover, their findings indicate that Polish accounting professionals tend to be more attached to the prudence principle than their British counterparts. However both these studies have been silent on the role of culture on accountants’ judgments.¹

Given the fact that patterns of thinking in Poland often include great reluctance to take risk, implementing IFRS in Poland created “an intellectual, mental and cultural earthquake” (Kabalski, 2012, p. 17). Our study also responds to recent calls for more comparative research on countries from the CEE and other regions to investigate the determinants of various corporate accounting issues (Albu et al., 2017).

Second, our investigation determines updated Hofstede’s cultural indexes for auditors and accountants in Poland and the UK, providing a more relevant understanding of the impact of culture on accounting judgment. Thus, future studies examining the effect of national culture on any aspect of accounting issues may make use of our updated cultural scores. Furthermore, to illustrate core cultural differences (especially in patterns of thinking) between the examined countries, we do not solely rely on the Hofstede, 1980) framework, but also draw on sociological and historical literature.

Third, to advance upon prior research, we supplement the univariate test findings by performing regression analyses and controlling for certain personal and professional characteristics of accountants. Most previous judgment research follows the approach taken by Doupnik and Richter (2004), where other factors (i.e. gender, years of work experience, IFRS application at work) are not controlled.

Fourth, our data do not support the suggestion that national culture influences accounting judgments made by Polish and British accounting professionals, complementing the previous critical literature on values-based theories of culture. Therefore, our study suggests that in the era of IFRS convergence in the EU, Hofstede-Gray’s framework may have lost its explanatory power in cross-cultural research, or it is simply not suitable for explaining cultural differences. Conceivably, as Kirkman et al. (2017, p. 20) note, researchers should broaden the conceptualization and operationalization of culture and explore other potentially powerful means of exploring culture, such as beliefs and norms. In this context, our study also contributes to the question of to what extent national cultural values actually have explanatory power. Thus, the present paper provides some tentative implications for consideration by scholars in future cross-cultural research related to the application of IFRS. Moreover, our study also has implications for standard-setters to consider when using uncertainty expressions in IFRS.

The remainder of this paper is organized as follows. The second section develops the Hofstede-Gray conceptual framework, while the third section summarizes prior research in this field. In the fourth section, we describe the profiles of the countries used in our study, hypotheses formulation, methodology, and results. The final section summarizes the findings and concludes the paper.

2. Literature review

2.1. Hofstede—Gray’s theoretical framework of culture

The significance of culture in explaining individual and group behavior in social systems has been recognized and explored across a broad range of literature, including accounting literature (Aggarwal & Goodell, 2014; Khlif, 2016). There is multiple evidence that national culture has an impact on the behavior of individual actors, creating variation in IFRS implementation and practice (e.g., Prescott & Vann, 2015). However, culture is the abstraction of intangible and multifaceted attributes that are difficult to access and quantify (Bitar et al., 2020).

While examining the influence of culture on various accounting issues, including professional judgment, cross-cultural research has almost exclusively relied on the cultural values of Hofstede, 1980), Hofstede and Bond (1988), and Hofstede et al. (2010) (Heidhues & Patel, 2011). Hofstede, 1980), p. 25) defines culture as “the collective programming of the mind distinguishing the members of one group or category of people from others”. He identifies four contrasting sets of cultural dimensions that can be used to describe general similarities and differences in cultures around the world: power distance (PD), uncertainty avoidance (UA), individualism (IND), and masculinity (MS). In his later work, Hofstede adds two more cultural dimensions: long-term orientation (LTO) and indulgence (IVR) (Hofstede & Bond, 1988; Hofstede et al., 2010). These dimensions can explain the structure of institutions and the behavior of society members. Furthermore, as they are measurable, they can be used to distinguish between cultural differences of societies or nations (Hofstede, 1980).

National culture is derived from societal values, which are shaped by external factors, such as history and economic systems. Hofstede’s model of cultural patterns was extended by Gray (1988), who introduces a theoretical framework in which the differences in a country’s accounting practices are related to the national culture. According to Gray (1988), conservatism is the oldest and most dominant dimension, which, along with secrecy, is the accounting subculture value that most strongly affects the valuation and level of disclosure in financial statements. Secrecy is closely related to conservatism in that both values imply a cautious approach in financial reporting (El-Helaly et al., 2020). Thus, these accounting values are relevant to our research.

Gray (1988, p. 8) describes conservatism as “a preference for a cautious approach to measurement so as to cope with the uncertainty of future events as opposed to a more optimistic, (…) risk-taking approach”. Secrecy entails “a preference for confidentiality and the restriction of disclosure of information about the business only to those who are closely involved with its management and financing as opposed to a more transparent, open and publicly accountable approach” (Gray, 1988, p. 8). According to Gray (1988) and Radebaugh and Gray (2002), uncertainty avoidance and long-term orientation are Hofstede’s cultural dimensions that have the greatest impact on the tendency towards conservatism. In turn, these two dimensions, along with power distance and individualism, have the strongest influence on the tendency for secrecy (Gray, 1988; Radebaugh & Gray, 2002). The analysis in these studies shows that uncertainty avoidance and long-term orientation are positively related to conservatism and secrecy (a more conservative and secretive approach is expected to preserve power inequalities and security, as well as prevent conflict and competition). On the other hand, these accounting values are negatively correlated with individualism and masculinity. Individualistic and more caring societies (i.e. feminine societies) demand accountability, require extensive disclosure, and are more optimistic relative to collectivist and masculine societies. Gray’s framework does not include the sixth dimension—indulgence (IVR). This relationship has been expanded by Borker (2013, p. 173), who states that “high IVR complements the notion of the freedom of judgment associated with (…) optimism and transparency”.

A natural consequence of this theory were attempts to empirically verify whether the relationships between Hofstede’s dimensions and accounting values (conservatism and secrecy) are the same as those defined by Gray. These studies have incorporated the Hofstede-Gray framework at country level. Most studies confirm that uncertainty avoidance is positively correlated to conservatism (Kiridaran et al., 2014; Salter & Niswander, 1995; Sudarwan & Fogarty, 1996). However, mixed results have been reported on the dimensions of masculinity, individualism, and power distance. For example, Noravesh et al. (2007) find that individualism and masculinity are negatively related to conservatism. Kiridaran et al. (2014) find the same relationship only for individualism, while Salter and Niswander (1995) find it only for masculinity. What is more, the results obtained by Sudarwan and Fogarty (1996) and Noravesh et al. (2007) show a positive relationship between power distance and conservatism, which was not identified by Gray.

Most of the research that focuses on Gray’s secrecy hypothesis provides evidence of the propounded negative relationship between individualism and secrecy (Noravesh et al., 2007; Salter & Niswander, 1995; Sudarwan & Fogarty, 1996). Contrary to Salter and Niswander (1995), Sudarwan and Fogarty (1996), and Noravesh et al. (2007) also support the hypothesis regarding power distance and masculinity. Study results regarding the dimension of uncertainty avoidance are divergent as well.

It is important to note that the relationships determined by Gray are based on a simplified assumption that Hofstede’s cultural values have been correctly identified. Meanwhile, this theory has been criticized by many researchers (including Baskerville, 2003; Baskerville-Morley, 2005; Beugelsdijk et al., 2015; Devinney & Hohberger, 2017; Kirkman et al., 2017; McSweeney, 2016; Orij, 2010). One major criticism of Hofstede’s cultural framework is that the dimension scores are outdated since they were established more than 30 years ago and there has not been any development over the years. Meanwhile, these indexes may have changed, especially in those countries where significant economic changes have taken place over the past 30 years (Orij, 2010), as exemplified by Poland. Researchers also suggest the IBM data used by Hofstede is not representative of the entire world. Furthermore, other researchers have questioned whether nations are the appropriate unit of analysis (Kirkman et al., 2017), as “cultures exist among many different social groups, including regions, generations, and socio-economic groups” (p. 21). Criticism of Hofstede’s model, however, does not seem to have caused researchers to reject it (Perera et al., 2012). Hofstede’s work has dominated the literature, partly because it was the first study that has developed a parsimonious national culture framework consisting of multiple cultural dimensions that are measurable (Beugelsdijk et al., 2017).

“Despite these criticisms, until now, there is no cultural concept which comprehensively explains the difference in characteristics between countries and presents quantitative data of Hofstede’s cultural dimensions” (Sutrisno & Dularif, 2020, p. 16). Moreover, according to Ramya et al. (2019) most of the Hofstede’s framework criticisms are based on a misunderstanding of the purpose of the knowledge and the context in which it was created. They state that “the strongest justification for Hofstede’s philosophical approach is that it adopted a pragmatic stance rather than attempting to perfectly represent reality” (p. 426). Therefore, they conclude that it is too early to consider abandonment of this significant body of knowledge and instead strongly encourage its cautious future use (p. 429).

2.2. The Hofstede-Gray cultural framework and accounting judgments

A number of studies have examined the effect of culture on accountants’ judgments when they apply principle-based standards. These studies have incorporated the Hofstede-Gray framework, but at an individual, rather than country, level. For example, Schultz and Lopez (2001) indicate that judgments vary significantly among the three countries of France, Germany, and the U.S., with completely different levels of uncertainty avoidance. However, they did not study the influence of culture alone. Thus, their research design does not allow them to separate the impact of culture from the influence of legal systems and sources of financing such estimates.

A few cross-cultural studies have also found that Gray’s accounting value of conservatism influences the interpretation of expressions contained in IFRS that require accounting judgment (e.g., Doupnik & Riccio, 2006; Doupnik & Richter, 2004). For example, Doupnik and Richter (2004) analyze the issue with respect to the impact that conservatism has on how professional accountants representing Anglo and German cultures interpret these terms, while Doupnik and Riccio (2006) extend this research to a previously unexamined Latin cultural area by additionally examining the impact of Gray’s value of secrecy. Doupnik and Richter (2004) find that German professional accountants represent a more (less) conservative approach when recognizing elements that increase (decrease) income, compared to the U.S. subjects. Thus, this finding shows that conservatism influences the interpretation of these terms in a predictable manner. Testing the secrecy hypothesis, Doupnik and Riccio (2006) suggest that Brazilian accountants require higher minimum levels of certainty than their U.S. counterparts when making disclosures in financial statements. They provide strong evidence for this hypothesis; nevertheless, their results for the conservatism hypothesis are mixed, as recognition of income-increasing items is supported, but no support is found for income-decreasing items.

This strand of research has been further advanced by Tsakumis (2007), who conducted an experiment to investigate the impact of national culture on Greek and U.S. accountants’ recognition and disclosure decisions, with particular emphasis on Gray’s accounting values of conservatism and secrecy. Specifically, participants were asked to interpret probability expressions resulting in recognition or disclosure of contingent assets and liabilities. The results of his study, however, are mixed. No significant differences were found between the recognition decisions of Greek and U.S. accountants. Thus, he suggests culture may not play any role in applying recognition rules across nations. Nonetheless, his results indicate that culture does play a role in accountants’ disclosure judgments.

In the same vein, Chand (2012), referring to accountants in Australia and Fiji, supports the view that culture influences their interpretation of the uncertainty expressions contained in IFRS. Similarly, Chand et al. (2012) and Hu et al. (2013) focus on the impact of national culture on judgments by comparing Australian and Chinese accounting students residing in Australia. While Hu et al. (2013) provide evidence for the generalizability of Gray’s conservatism hypothesis to the Confucian cultural area, Chand et al.’s (2012) results are mixed. With regard to the secrecy hypothesis, Chand et al. (2012) find strong support for the notion that Chinese accounting students are more secretive than their Australian counterparts.

Wehrfritz and Haller (2014) investigate the judgments of German and British accountants that require discretion under IFRS. They find only some support for international differences between the nationalities surveyed in terms of accounting interpretation and estimation. They provide significant evidence that German accountants adopt a more conservative approach to the recognition of provisions than UK subjects. With regard to their measurement and disclosures, however, the evidence is contrary to the prediction. Therefore, they suggest “national culture (…) may have lost a considerable degree of significance in the context of application of IFRS” (Wehrfritz & Haller, 2014, p. 204).

Recently, Prather-Kinsey et al. (2018) analyzed U.S. and Indian accountants to investigate whether work locations or core self-evaluations influence their interpretation of the term “control” when applying the rule-based (as defined in U.S. GAAP) and principle-based (as defined in IFRS) definition of the term “control”. Their findings reveal that work location and core self-evaluations affect accountants’ interpretation of this term when applying principle-based terminology. Thus they conclude principle-based standards (as defined in IFRS) are likely to result in incomparable interpretations when individuals have different personalities or work in culturally different locations (p. 62). Similarly, Gupta et al. (2019) find that U.S. and Indian practitioners indeed differ significantly in their interpretation of certain identical uncertainty expressions. Specifically, Indian accountants are more conservative in their probability assessments than their U.S. counterparts. Eljammi Ayadi et al. (2020) provide evidence for the generalizability of Gray’s conservatism hypothesis in Tunisia and Egypt, in the context of income recognition. Their results also demonstrated that Gray’s accounting value of secrecy affects the interpretation of uncertainty expressions used in the IFRS to establish disclosures.

In light of these cross-cultural studies, most of the available evidence is inconclusive, indicating the issue requires further investigation.

3. Hypotheses development

In this section, we draw on sociological and historical literature to determine differences between Polish and British cultures. These characteristics are followed by the countries’ rankings of Hofstede’s cultural dimensions and their relationships to Gray’s cultural accounting values of conservatism and secrecy. We then formulate three hypotheses based on these cultural differences between the countries.

3.1. Characteristics of Polish and British culture

The main differences between the British and Poles are evident in their mental preparation for adopting the IFRS spirit. The mentality of Poles—as in other nations—is historically rooted. Polish researchers dealing with cultural determinants of economic development believe the main thought patterns of Poles that affect modern reality were, to a large extent, shaped at the end of the 16^th century (Banach, 2014; Hryniewicz, 2004, 2014). Before the 16^th century, development in Poland and other European countries located on the eastern part of the continent was similar to that of Western Europe. These countries gradually imitated Western institutional, legal, and cultural patterns. This changed in the 16^th century when the demand for grain began to increase in Western Europe due to the increase in urban population in that part of the continent. Eastern European nobility, taking advantage of this economic situation and their dominant social and political position, became the main exporters of grain (Hryniewicz, 2004, 2014). The increase in grain production was achieved by creating and developing a grange system, the functioning of which had little to do with the order of the market emerging in Western Europe. Manors were a kind of profit-oriented feudal enterprise, but were completely independent of the labor market. They were distinguished by the fact that they created an impassable distance between nobility and serfs, who were de facto slaves. Nowhere in Europe has there been such an emphasis on the differences in the estates of the realm (Zybala, 2016). The monastic-grange culture introduced a quasi-totalitarian social organization encompassing nearly 90% of Polish society (Bendyk, 2008). This situation resulted in resigning from Western institutional solutions and moving away from developed Western European countries.

The monastic-grange culture not only shaped the management style, but also the way of responding to commands. As such, it equally shaped managers and workers. The monastic-grange mentality flourished in Poland twice as much during the times of communism (Hryniewicz, 2004, 2014), when the organization of socialist enterprises strengthened the strict subordination of workers to superiors (Zybala, 2016). The communist era has been a source of anti-market mentality and has left a strong legacy antagonistic to capitalism (Czegledi & Newland, 2018). During this period, society was reluctant to make any changes or attempt reforms. Their attitude was characterized by passivity, as well as a lack of initiative and independence. At work, they expected clear instructions from management and, outside of work, social security and assistance. The mental contamination of socialism led to the formation of the ‘Homo Sovieticus’² personality in Poland.

The Polish reflection on the issue of “Homo Sovieticus” is dominated by the Tischner’s approach (Synowiec, 2019). Tischner describes the concept of “Homo Sovieticus” as a theoretical category that refers to a particular kind of mentality resulting from the influence of communist ideology (Tischner, 2018). In the perspective of the anthropology of postsocialism he/she can be regarded as in the opposition to the “capitalist pattern of a self-steering individual, responsible for its own fate, active, making independent decisions” (Kolodziejska & Hnatiuk, 2015, p. 120).

“Homo Sovieticus” is an incapacitated, mentally enslaved, deprived of initiative and critical-thinking spirit. His/her decisions depend on indications, suggestions, and orders from other people or institutions, and he/she likes to follow beaten paths (Kabalski, 2012). When external indications are missing, he/she becomes helpless and lost, leading to unwillingness to make any individual decisions (Buchowski, 2012; Walter, 2011). In other words, he/she only feels good when everything is simple. When Poland regained freedom in 1989, “Homo Sovieticus” “did not automatically transform into a citizen of a democratic state” (Wnuk-Lipinski, 2008). As Synowiec (2019) notes, in the face of this change, he/she has lost the ability to move in the new reality. The process of transformation did not change the social consciousness of Poles enough to free them from its influence (Szostak, 2004), as confirmed by research (i.e. Korach, 2009; Tobor-Osadnik & Wyganowska, 2011). The study by Newland (2018) demonstrates that Poland is an example of a country that shows an unfavorable view of capitalism which may be rooted in the ideology spread in their communist era.

The monastic-grange tradition has left a large imprint on the mentality of Poles; the attitudes developed during this period have survived to this day and still create specific social behaviors of Poles. In many companies and organizations, it is reflected in the distance between subordinates and superiors. This is expressed, among other ways, in the lack of trust and unwillingness to undertake risky ventures where responsibility must be shared (Zybala, 2016). It is also manifested in the strongly marked position of managerial power and lack of employee initiative (Kolasa-Nowak, 2018).

When the above-described personality traits are confronted with IFRS, a number of problems can be pointed out related to their practical application among Polish accountants; these problems are the result of a lack of accountants’ mental adaptation to the IFRS requirements. Responsibility for a judgment lies with the accountants themselves, which requires a critical analysis of reality. Unfortunately, the ability to make such an analysis is not a feature of the “grange” or “Homo Sovieticus” personality. It is very difficult for such a person to question the sense of the rules imposed by authorities, not only because he/she lacks the ability to think critically, but primarily because he/she always realizes his/her will thoughtlessly (Walter, 2011). Meanwhile, research conducted on a group of accountants in Poland confirms that “Homo Sovieticus” personality traits occur frequently, although they are not dominant among Polish accountants (Kabalski et al., 2012).

The “preference for detailed and rigorous regulations” as a consequence of “strict adherence to the letter of the law” that is rooted in the mentality of Poles is not without influence on Polish accountants’ decisions regarding IFRS application. Polish accountants’ preference for prescriptive rules sharply contrasts with the principle-based nature of IFRS. The Anglo-Saxon concept of usefulness for decision-making seems to be irrelevant to most of them (Vellam, 2004) because it is contrary to the Polish accounting background. Before 1989, the Polish accounting tradition was shaped by the needs of the centrally-planned economy, where the state was the primary user of information generated by an accounting system that constituted a mere tool of control and a source of statistical data for planning purposes. After that period, greater significance was still attributed to the control function of accounting than to its information-related functions (Jaruga et al., 1996). At that time, the relationship between accounting law and tax law was very strong. In Poland, its communist history coupled with the conservative Continental accounting tradition, resulted in the fact that accountants “give little consideration to the overall truth and fairness of the financial statement”, focusing on tax issues, even when it “overshadows any other economic consideration” (Vellam, 2004). Despite that, the revised Accounting Act of 2000 moved Polish accounting away from a tax-oriented system towards a decision-making tool (Grabinski et al., 2014). Studies show that Polish accountants still consider tax authorities the primary users of published financial information. Furthermore, Gorowski and Stando-Gorowska (2015) study shows the same tendency among young people born after the Polish transition from the communist system to a market economy. Moreover, deviations from accounting regulations, as opposed to tax law, are usually not challenged, due to the weak position of the accounting and audit profession (Vellam, 2004). Isolation from Western influence and the communist era affected the status of accountants, who were perceived as bookkeepers rather than “business managers at the centre of decision making” (Vellam, 2004).

In contrast, Western European countries took a different development path, which had a significant impact on the formation of separate social attitudes. The history of Western Europe can be read as an intertwining of aspirations for individualization and political and economic freedom, primarily resulting from the expansion of capitalism. It has been stated that the factors contributing to the expansion of capitalism were the great geographical discoveries at the turn of the 15^th and 16^th centuries, as well as the development of overseas trade and the subsequent colonial exploitation of resources and the labor force (e.g., Bryer, 2006).

In turn, the 18^th century witnessed the beginning of the industrial revolution initiated in Great Britain and the progressive expansion of its colonial territories. The increase in the size of the urban population (outflow of people from the countryside) led to a change in ownership structure and increased the demand for industrial products, which could not be satisfied by production solely based on manufacture. In the 19^th century, however, the era of factory production and increasingly efficient and productive industrial civilization began; economic activity was an individual matter. The ventures undertaken depended on private financing, and the role of the state was limited to introducing corporate regulations in trade and maintaining competition. Ultimately, the feudal mode of production was replaced by the capitalist mode (Bryer, 2006). From a social perspective, this entailed a transition from a legal-and-personal relationship between the owner (a free person—a feudal lord) and a serf (a dependent person—servants, peasants) to a purely economic relationship between a capitalist and an employed person (an official).

The changes observed in social relationships were accompanied by the formation of a capitalist mentality, associated with a “business-like attitude” and a “responsiveness to the market” (Bryer, 2005, p. 27), which underlays the modern world of business. Freedom of an individual who has the right to express and fulfill his/her preferences constitutes an immanent feature of this mentality (Swadzba, 2011). British culture has always been considered the cradle of individualism and free markets and as the most enthusiastic for capitalism (Czegledi & Newland, 2018).

3.2. Conversion of Hofstede data to Gray’s conservatism and secrecy accounting values

The above description of the Polish and UK cultural characteristics finds confirmation in Hofstede et al.’s (2010) six-dimensional model of national culture.³ Specifically, the cultural dimensions labelled power distance, uncertainty avoidance, and indulgence are particularly reflective of the Polish and UK cultural characteristics identified in the sociological and other literature (Table 1).

Table 1. Cultural dimension scores by country—hofstede origin

Country	Power Distance (PDI)	Individualism (IND)	Uncertainty Avoidance (UA)	Masculinity (MAS)	Long-Term Orientation (LTO)	Indulgence (IVR)
Poland	68	60	93	64	38	29
United Kingdom	35	89	35	66	51	69

Source: Hofstede et al. (2010).

In contrast to the UK (ranked as low PDI), Poland, with a score of 68 PDI, is a hierarchical society. The relationships between Polish workers are based on the large distance between managers and workers, who expect their pace of work to be imposed and controlled, as well as to be accounted for the effects of this work. According to Gray’s (1988, p. 11) theory, Polish professional accountants are less likely to disclose information in financial statements “to preserve power inequalities”.

Poland, with a score of 60, is a less individualistic society than the UK. This means there is less preference for a loosely-knit social framework in which individuals are expected to take care of only themselves and their immediate families. Among others, these differences may be a result of the communist system. Based on Gray (1988), Polish accounting professionals should be more conservative in measurement and more secretive in disclosure, because they are closely involved with the firm they work for and exhibit less emphasis on individual achievements.

Unlike the UK, Poland is characterized by a very high preference for uncertainty avoidance. As such, it exhibits an emotional need for rules (even if the rules never seem to work), as well as low acceptance of exceptions and unusual situations. Security and risk aversion constitute an important element in individual motivation, which is characteristic of the “Homo Sovieticus” attitude. Therefore, Polish accountants are characterized as more conservative and secretive, reflecting their desire to preserve security.

Hofstede’s cultural-value findings indicate that Poland and the UK are somewhat close on the masculinity dimension. Poland ranks slightly lower than the UK on masculinity; however, this dimension is regarded by Gray as having less influence on conservatism and secrecy accounting values.

Polish culture can be characterized as restrained. As such, contrary to the indulgent British culture, people have a tendency towards cynicism and pessimism. People with this orientation perceive their actions as restrained by social norms and feel that indulgence is somewhat wrong. Although this dimension is not included in Gray’s framework, a low IVR is associated with a more conservative and secretive approach to judgment (Borker, 2013).

Poland’s low score of 38 in the LTO dimension means that it is more normative than pragmatic. Normative societies demonstrate a conventional short-term perspective, consistent with a less conservative and secretive approach, where quick results and more disclosure for investor decision-making are expected.

Except for the differences in the LTO dimension, differences in the other five cultural dimensions suggest that Polish accounting professionals will exhibit stronger conservatism and be more secretive in their judgments than their British counterparts (Table 2). These assumptions are based on the output of Gray’s framework of accounting dimensions, according to which (Borker, 2013; Gray, 1988; Radebaugh & Gray, 2002):

1. Higher national ranks of uncertainty avoidance and long-term orientation, as well as lower ranks of individualism, masculinity, and indulgence, will probably determine a higher national rank of conservatism.

2. Higher national ranks of uncertainty avoidance, long-term orientation, and power distance, as well as lower ranks of individualism, masculinity, and indulgence, will probably determine a higher national rank of secrecy.

Table 2. Conversion of hofstede data to gray’s conservatism and secrecy accounting values

Hofstede’s Dimensions	Country
	Poland	United Kingdom
Power Distance (PDI)	No relation identified
Individualism (IND)	Lower	Higher
Uncertainty Avoidance (UAI)	Higher	Lower
Long-Term Orientation (LTO)	Lower	Higher
Indulgence (IVR)	Lower	Higher
Gray’s Conservatism Hypothesis Result	CONSERVATISM	OPTIMISM
Power Distance (PDI)	Higher	Lower
Individualism (IND)	Lower	Higher
Uncertainty Avoidance (UAI)	Higher	Lower
Long-Term Orientation (LTO)	Lower	Higher
Indulgence (IVR)	Lower	Higher
Gray’s Secrecy Hypothesis	SECRECY	TRANSPARENCY

3.3. Hypotheses formulation

The primary question posed in this study is whether cultural differences (as measured by Hofstede) cause accountants from different countries to interpret the same financial reporting standards differently. To address this question, countries were selected that are likely to have significant differences in Hofstede’s cultural values and therefore likely to have significantly different Gray’s accounting values.

In the light of the previous discussion, the cultural characteristics of Poland and the UK suggest Polish professional accountants will be more conservative than their British counterparts when interpreting accounting standards. In accordance with Gray (1988) and Radebaugh and Gray (2002), it is argued that Polish accounting professionals will need greater certainty to recognize profit than to recognize loss. As such, they will be more likely to delay recognition of assets and net income increases, while accelerating recognition of liabilities and net income decreases, compared to British accounting professionals. This leads to the following conservatism hypotheses:

H1: Polish accounting professionals require higher certainty than their British counterparts when recognizing items that increase total assets and income.

H2: Polish accounting professionals require lower certainty than their British counterparts when recognizing items that increase total liabilities and reduce income.

With regard to the value of secrecy, Polish accounting professionals are regarded as being more secretive, therefore, they need greater certainty to make disclosures than their British counterparts. This leads to the following secrecy hypothesis:

H3: Polish accounting professionals require higher certainty than their British counterparts when making disclosures in financial statements.

4. Research design

4.1. Sample

The group surveyed consisted of accounting professionals (statutory auditors/chartered accountants and other accounting department employees) from Poland and the UK. The Polish accountants and auditors were selected during obligatory training courses given by the Polish Chamber of Statutory Auditors (Polska Izba Bieglych Rewidentow, PIBR) and the Accountants Association in Poland. The accountants and auditors from the UK were members of the ACCA and ICAEW who worked in accounting and auditing companies that were randomly selected from the website www.auditregister.org.uk.

Findings of previous accounting and non-accounting studies have indicated that the interpretation of uncertainty expressions may also be influenced by several other factors. As a result, we incorporated the following characteristics of the respondents: gender (Chand, 2012; Francis et al., 2015; Han et al., 2016), age (Chand, 2012; Wehrfritz & Haller, 2014), years of professional experience (Chand, 2012; Wehrfritz & Haller, 2014), current position held (Bierstaker et al., 2016; Laswad & Mak, 1994), statutory auditor/chartered accountant qualification (Libby & Frederick, 1990), experience in multinational companies (Doupnik & Riccio, 2006; Tsakumis, 2007), and frequency of applying IFRS at work (Doupnik & Richter, 2004).

Table 3 shows the profile of the research sample.

Table 3. Sample

Subject	Number of persons		No data
	Poland	THE UK	Poland	THE UK
Total respondents:	330	52	-	-
Profiles of Respondents
Gender:			7	0
Female	232	9
Male	91	43
Age (years)			7	0
20–29	6	3
30–39	64	6
40–49	55	8
50–59	66	26
60+	132	9
Professional experience (years)			27	1
1–5	8	4
6–10	34	0
11–15	56	6
16–20	44	4
21+	161	37
Actual position held:			18	0
Statutory auditor/chartered accountant	179	36
Other	133	16
Statutory auditor/Chartered accountant qualification			26	0
Yes	258	48
No	46	4
Professional experience in a multinational company			25	0
Yes	78	6
No	227	46
Application of IFRS at work:			16	0
Yes	247	40
No	67	12

The sample structure is broadly consistent in terms of some of the variables, such as the current position held, statutory auditor/chartered accountant qualification, professional experience in a multinational company, and application of IFRS at work.

The biggest differences concern the respondents’ gender. In Poland, contrary to the UK, the majority in the sample tested were women (71.8% to 17.3% respectively), which is not surprising. As Kabalski and Szwajcar (2015) indicate, most accountants in Poland are women. Moreover, most of the respondents from both countries are over 50 years old and have more than 16 years of work experience.

4.2. Data and research instrument

To verify the research hypotheses, we conducted the study in sequential stages that required different tools. First, a survey questionnaire was administered, which consisted of instructions for its completion, a set of IFRS excerpts containing verbal probability expressions (Appendix 2), a Hofstede Values Survey Module (VSM 2008), and respondents’ demographic factors. Additionally, a cover letter explaining the motivation behind the study was attached.

In Poland, official Polish translations of the IFRS excerpts were used. A total of 984 forms were mailed and 351 of the questionnaires were returned. Of these, 330 included answers that were subjected to analysis. The English version of the questionnaire was e-mailed to members of the ACCA and ICAEW. The authors intended to achieve the same, or similar, research sample size as the Polish respondents. Due to outdated email addresses, some questionnaires did not reach recipients, making it necessary to conduct another call. Eventually, 995 questionnaire forms reached the target group; 163 questionnaires were returned, but only 52 were fit for analysis. The high response rate in Poland (33.54%) compared to that in the UK (5.23%) can probably be explained by the fact that completing a paper questionnaire copy during a training course is a lower burden than completing an electronic version during work.

The study incorporated 10 IFRS fragments issued by the Accountants Association in Poland (Stowarzyszenie Ksiegowych w Polsce, SKwP) for the Polish version of the survey, and the IFRS Foundation for the English version. The Polish version of the VSM, available on the website http://www.geerthofstede.nl/vsm-08, is not the original version. To ensure the quality of its translation, it was first translated into English and then compared to Hofstede’s English original.

To test the hypotheses, we analyzed the interpretation of the following probability terms occurring in assorted IFRS excerpts: “probable”, “is probable”, “is not probable”, and “is remote”. Because these terms are used to establish the threshold for recognition of a given item, as well as to establish the threshold for disclosures in the financial statements, they have been grouped in the manner presented in Table 4.

Table 4. Probability expressions

IAS/	Probability expressions
Paragraph
H1: Resulting in measurement of items increasing the financial result
Nov-23	is probable
Dec-24	is probable
16-Jul	is probable
18/22	is probable
38/57	probable
H2: Resulting in measurement of items decreasing the financial result
Nov-36	is probable
Dec-36	is not probable
37/14	is probable
H3: Resulting in making disclosures in the financial statements
37/28	is remote
37/34	is probable

The respondents’ task was to determine the numerical probability that, in their opinion, would determine the threshold resulting in recognition of revenue (IAS 11 para 23; IAS 18 para 22), assets (IAS 12 paras.24, 36; IAS 16 para 7; IAS 38 para 57), losses (IAS 11 para 36), provisions (IAS 37 para 14), and disclosing contingent liabilities and assets (IAS 37 paras.28, 34). They were to adopt a scale from 0 to 100 percent, where 0 indicates impossibility and 100 indicates absolute certainty. A sample question regarding IAS 12 para 24 was formulated as follows:

“A deferred tax asset shall be recognized for all deductible temporary differences, to the extent it is probable that taxable profit will be available, against which the deductible temporary difference can be utilized (…)”

In this context, the expression “is probable” corresponds to a probability greater than what percentage?

… … . … … % ‘

An analogous research method was previously adopted by Doupnik and Richter (2004), Doupnik and Riccio (2006), Chand et al. (2012), and Hu et al. (2013). Because these terms are used to measure various items, the level of the Polish and UK respondents’ conservatism should influence the way they are interpreted. Moreover, due to the fact that these expressions are also used to establish the threshold resulting in the obligation to make disclosures in financial statements, their interpretation should also be affected by the level of secrecy shared by accountants from these countries.

4.3. Univariate correlation analyses and regression models

Measures of central tendency and dispersion were used to describe our variables. For this type of quantitative approach, a normality test is an important stage for deciding the measures of central tendency and the econometric methods for data analysis. In the first step, we conducted chi-squared tests to analyze the normality of the distribution of our variables. Due to the fact that our variables were not normally distributed, we used both non-parametric tests and estimation methods to compare the accounting judgments of the Polish and British respondents. The non-parametric Mann-Whitney U test was used to allow comparison of respective median point estimates, while for the model estimation, we applied the LAD (least absolute deviations) approach. The main advantage of LAD is its robustness, meaning that it is resistant to outliers in the data. Also, LAD places equal emphasis on all observations. Since the outliers in our study did not need to be given greater weight than other observations, LAD was considered an appropriate method for our empirical research (Bassett & Koenker, 1978; Buchinsky, 1998).

We used regression models to test the predictions in our hypotheses while controlling for other personal and professional characteristics: gender, years of professional experience, current position held, statutory auditor/chartered accountant qualification, experience in multinational companies, and frequency of applying IFRS at work.

The following model was estimated for each standard separately:

$\begin{array}{rl}& y_i={\beta }_0+{\beta }_{1}NATIONALIT{Y}_i+{\beta }_{2}GENDE{R}_i+{\beta }_{3}WORKEX{P}_i\\ & +{\beta }_{4}POSITIO{N}_i+{\beta }_{5}QUALIFICATIO{N}_i\\ & +{\beta }_{6}MULTCO{M}_i+{\beta }_7\mathrm{IFRSap}{p}_i+u_i\end{array}$

where:

yi—IAS/Para (for example: IAS 11 para 23);

NATIONALITY—dummy variable: 1 if the participant belongs to the Polish sample, 0 if the participant belongs to the British sample;

GENDER—dummy variable: 1 for male, 0 for female;

WORKEXP—years of work experience;

POSITION—dummy variable: 1 for statutory auditor/chartered accountant, 0 otherwise;

QUALIFICATION—dummy variable: 1 if the respondent has statutory auditor/chartered accountant qualification, 0 otherwise;

MULTCOM—dummy variable: 1 if the respondent has professional experience in a multinational company, 0 otherwise;

IFRSapp—dummy variable: 1 if the respondent regularly or seldom refers to IFRS at work, 0 if the respondent never refers to IFRS at work;

β₀, β₁, …, β₇—estimated coefficients; and

u_i—error term.

The independent variable is NATIONALITY and the dependent variable is the numerical probability determined by the respondents for each expression. The coefficients on NATIONALITY variable, therefore, indicate the tendency of answers from Polish respondents relative to British respondents in the respective case.

The standard error of the regression represents the average distance between the observed values and estimated ones. Smaller values are better and indicate that the fitted value of the dependent variable is closer to the real one. The coefficient of variation (CV) is defined as the ratio of the standard deviation to the mean of the dependent variable and thus shows the extent of variability in relation to the population mean. Smaller values of the CV are better.

The descriptive statistics are displayed in Appendix 3. Moreover, in our regression analysis, we performed listwise deletion. Table 5 presents a correlation matrix for the control variables in our study.

Table 5. Spearman correlations

Variable		1	2	3	4	5	6	7
1.	GENDER	1.000
2.	AGE	0.006	1.000
3.	WORKEXP	0.031	0.83*	1.000
4.	POSITION	−0.066	−0.38	−0.41*	1.000
5.	QUALIFICATION	−0.032	−0.27	−0.327	0.356	1.000
6.	MULTCOM	−0.052	0.27	0.255	−0.218	−0.122	1.000
7.	IFRSapp	−0.12	0.20	0.111	0.039	−0.070	0.368	1.000

* denotes significance at p < 0.05.

This analysis indicated that multicollinearity is not a concern in our study. Apart from a logical strong and significant correlation between age and work experience (r = 0.83, p < 0.05) and a moderate correlation between work experience and current position held (r = −0.41, p < 0.05), we did not find any other significant correlations between the control variables included in our regression model. As a consequence, we removed the control variable AGE from our regression analysis.

5. Empirical results

5.1. Hofstede cultural values scores

To ensure that accounting professionals from Poland differ from their UK counterparts in the direction identified by Hofstede, VSM 2008 was also attached to the research instrument. The results obtained are presented in Table 6.

Table 6. The vsm cultural value scores for the sample of polish and uk accounting professionals

Cultural value	Poland	UK	Hofstede’s direction of differences	Consistence with direction of differences
Power Distance	47	14	PL > UK	YES
Individualism	39	37	PL < UK	NO
Uncertainty Avoidance	61	21	PL > UK	YES
Masculinity	26	6	PL < UK	NO
Long-Term Orientation	13	4	PL < UK	NO
Indulgence	42	64	PL < UK	YES

The direction of the differences between Poland and the UK for power distance, uncertainty avoidance, and indulgence is consistent with the direction identified by Hofstede. In line with Hofstede’s expectations, our results confirm that Poles—in contrast to the British—have higher uncertainty avoidance and power distance, as well as a lower indulgence rate. At the same time, the direction of the differences for individualism, masculinity, and long-term orientation was opposite that assumed by Hofstede. The level of individualism was similar for both groups (39 for Poland and 37 for the UK), while masculinity is not considered significant when determining the tendency towards secrecy and conservatism. Our results also show that Poles are characterized by a higher level of long-term orientation than the British; therefore, according to Gray’s theory, they should also be characterized by a higher level of conservatism and secrecy. The VSM scores confirm that Polish accounting professionals should be more conservative and secretive than UK accounting professionals.

5.2. Tests of conservatism hypotheses (H1 and H2)

Hypothesis 1 is supported if Polish accounting professionals (characterized as conservative) assign a higher numerical probability to the uncertainty expressions than UK accounting professionals (characterized as optimistic) (Figure 1).

Figure 1. Conservatism vs optimism and probability threshold for items increasing financial results

CONSERVATISM VS OPTIMISM AND PROBABILITY THRESHOLD FOR ITEMS INCREASING FINANCIAL RESULTS

Table 7 present the results of non-parametric Mann-Whitney U tests.

Table 7. Results of the h1 mann-whitney u tests for median differences

IAS/ Paragraph	POLAND	UK	Expectation	Consistent with our expectations	z
	Median	Median
H 1: Resulting in measurement of items increasing the financial result
11/23	60	75.0	PL>UK	No	−2.27*
12/24	70.0	70.0		No	−0.43
16/7	70.0	60.0		Yes	1.02
18/22	70.0	68.5		Yes	−0.04
38/57	70.0	72.5		No	−0.06

* denotes significance at p < 0.05.

The Mann-Whitney U tests results in Table 7 indicate a significant difference between respondents only for IAS 11 para 23. However, the direction of this difference is not as expected. The differences in mean probabilities occur in the predicted direction in only two of the five uncertainty expressions (IAS 16 para 7 and IAS 18 para 22). For other excerpts, contrary to our expectations, the Poles assigned a lower level of probability than the British.

Table 8 provides the results of the LAD estimations for H1, which allowed testing the hypothesis by examining other factors mentioned earlier and confirming the univariate test results by controlling for respondents’ other personal and professional characteristics.

Table 8. Lad estimation results For H1

	Dependent variables (numerical probability for each IAS/Paragraph)
	Predicted sign	IAS 11/23	IAS 12/24	IAS 16/7	IAS 18/22	IAS 38/57
const		64.00*	52.50*	46.25*	36.00*	62.50*
NATIONALITY	+	−15.0*	0.0	0.0	8.3	−2.5
GENDER		0.00	−6.25	−5.00	5.00	−7.5
WORKEXP		1.00	3.75	6.25*	3.33	5.00*
POSITION		0.00	5.00	0.00	1.67	2.50
QUALIFICATION		6.00	0.00	1.25	10.67	−5.00
MULTCOM		0.00	2.50	1.25	0.00	−5.00
IFRSapp		−7.00	−3.75	−5.00	−1.67	−7.50
	Model diagnostics
standard errors		20.25	21.88	21.94	24.07	22.09
coefficient of variation		33.75	31.25	31.34	34.39	31.55
normality test		18.15*	40.99*	28.80*	70.46*	70.49*

*denotes significance at p < 0.05.

To support H1, the impact of the variable NATIONALITY should be positive and statistically significant. The results show that the NATIONALITY variable is significant only for IAS 11 para 23; however, it does not have the predicted sign. Therefore, our results, which reinforce the evidence from the univariate tests, do not provide support for H1 that Polish accounting professionals require higher minimum levels of certainty than their British counterparts when recognizing items that increase total assets and income. Excluding WORKEXP, the control variables are not statistically significant.

Hypothesis 2 is supported if Polish accounting professionals assign a lower numerical probability than UK accounting professionals to the uncertainty expressions (Figure 2).

Figure 2. Conservatism vs optimism and probability threshold for items decreasing financial results

Table 9 presents the results of non-parametric Mann-Whitney U tests.

Table 9. Results Of The H2 Mann-Whitney U Tests For Median Differences

IAS/ Paragraph	POLAND	UK	Expectation	Consistent with our expectations	z
	Median	Median
H2: Resulting in measurement of items decreasing the financial result
11/36	50.0	55.0	PL<UK	Yes	−1.57
12/36	50.0	50.0		No	−0.78
37/14	50.0	60.0		Yes	−2.17*

* denotes significance at p < 0.05.

The Mann-Whitney U test results indicate the differences in median probabilities assigned by the Polish and British respondents in two excerpts are in the expected direction, but with the exception of IAS 37 para 14, they are insignificant.

The results of the LAD estimations for H2 are presented in Table 10.

Table 10. Lad Estimation Results For H2

	Predicted sign	Dependent variables (numerical probability for each IAS/Paragraph)
		IAS 11/36	IAS 12/36	IAS 37/14
const		55.00*	20.00	60.00*
NATIONALITY	-	−5.00	0.00	−10.00
GENDER		0.00	0.00	0.00
WORKEXP		0.00	0.00	0.00
POSITION		0.00	0.00	0.00
QUALIFICATION		0.00	30.00*	0.00
MULTCOM		0.00	0.00	0.00
IFRSapp		0.00	0.00	0.00
Model diagnostics
standard errors		22.99	24.90	21.82
coefficient of variation		45.97	49.81	43.64
normality test		5.25	7.58*	5.52

*denotes significance at p < 0.05.

To support H2, the impact of the variable NATIONALITY should be negative and statistically significant. The LAD results showed that, except IAS 12 para 36, the sign was consistent with our expectations. However, this coefficient was insignificant for all analyzed excerpts. Therefore, the results from both the univariate tests and LAD regressions indicate that Polish accounting professionals require lower minimum levels of certainty than their British counterparts when recognizing items that increase total liabilities and reduce income. However, the LAD results did not confirm the significance of these differences as other than QUALIFICATION, the control variables were not significant.

Overall, the results provide no support for the notion that conservatism affects judgments made by accounting professionals from Poland and the UK when interpreting uncertainty expressions used in IFRS.

5.3. Test of the secrecy hypothesis (H3)

Hypothesis 3 is supported if Polish accounting professionals (characterized as secretive) assign higher probability than UK accounting professionals (characterized as transparent) to the uncertainty expressions (Figure 3).

Figure 3. Secrecy vs transparency and probability threshold for making of disclosures in financial statement

Table 11 presents the results of non-parametric Mann-Whitney U tests.

Table 11. Results Of The H3 Mann-Whitney U Tests For Median Differences

IAS/ Paragraph	POLAND	UK	Expectation	Consistent with our expectations	z
	Median	Median
H3: Resulting in making of disclosures in the financial statements
37/28	20.0	15.0	PL>UK	Yes	1.77
37/34	50.0	75.0		No	−2.86*

* denotes significance at p < 0.05.

The Mann-Whitney U test results demonstrate a significant difference for IAS 37 para 34; however, the direction of the difference is not in the expected direction. Thus, the results provide no support for the notion that the value of secrecy has an impact on subjective judgments made by Polish and British accounting professionals.

Table 12 provides the results of the LAD estimations for H3.

Table 12. Lad estimation results for h3

	Predicted sign	Dependent variables (numerical probability for each IAS/Paragraph)
		IAS 37/28	IAS 37/34
const		20.00	69.00*
NATIONALITY	+	10.00	−21.00*
GENDER		10.00	−6.00
WORKEXP		0.00	2.00
POSITION		0.00	−4.00
QUALIFICATION		−10.00	4.00
MULTCOM		0.00	−2.00
IFRSapp		0.00	4.00
Model diagnostics
standard error		24.64	21.77
coefficient of variation		123.2	36.29
normality test		159.1*	13.73*

*denotes significance at p < 0.05.

To support H3, the impact of the variable NATIONALITY should be positive and statistically significant. The LAD results indicate this coefficient is significant only for IAS 37 para 34; however, the sign predicted is contrary to our expectation. Apart from that, the control variables are not significant. These results, which reinforce the evidence from the univariate tests, also do not provide support for H3 that Polish accounting professionals require higher minimum levels of certainty than their British counterparts when making disclosures in financial statements.

6. Conclusion

The research question addressed in this study is whether national culture influences judgments made by Polish and British accounting professionals when interpreting selected IFRS that contain uncertainty expressions. The motivation for this research was provided by the inference that IFRS may not be suitable in Poland because it is a country with significant differences from the profile that best suits application of IFRS. The “preference for detailed and rigorous regulations” rooted in the mentality of Poles as a consequence of “strict adherence to the letter of the law” is not without influence on Polish accountants’ decisions. Polish accountants’ preference for prescriptive rules sharply contrasts with the principle-based nature of IFRS. The Anglo-Saxon concept of usefulness for decision-making is contrary to the Polish accounting background (Vellam, 2004).

Our study explicitly tests Gray’s theoretical framework with respect to conservatism and secrecy values at the individual level. We examined these hypotheses in the context of interpreting uncertainty expressions used as a threshold for recognition of items that result in increasing or decreasing financial results or making disclosures in financial statements.

Our findings did not confirm that national culture influences the interpretation of uncertainty expressions requiring accounting judgment, at least based on the Hofstede-Gray framework. We found no support for either Gray’s conservatism or secrecy hypotheses. In contrast to some prior findings, our study suggests that the influences of national culture may have lost a considerable degree of significance in the context of discretionary decisions under IFRS. With respect to the secrecy hypothesis, our findings, are consistent with those of Wehrfritz and Haller (2014), but contrary to the results obtained by Eljammi Ayadi et al. (2020), Chand et al. (2012), Tsakumis (2007), and Doupnik and Riccio (2006), who provide strong evidence that culture does play a role in accountants’ disclosure judgments.

Mixed results have been reported in previous studies for the conservatism hypothesis, including Wehrfritz and Haller (2014), Chand et al. (2012), and Doupnik and Riccio (2006), who find only some support for the notion that national culture may influence accounting judgment. While this hypothesis has been supported by Gupta et al. (2019), Hu et al. (2013), Chand (2012), and Doupnik and Richter (2004), these studies addressed non-EU countries. Thus, as Gupta et al., 2019) note, continental EU members may be less conservative after IFRS adoption than in the era of Gray’s (1988) model. Our results are in line with this concept.

To summarize, our examination of Poland and the UK does not provide evidence of the generalizability of Gray’s secrecy and conservatism hypotheses to EU members. Our study suggests that in the era of IFRS convergence in the EU, Hofstede-Gray framework may have lost its explanatory power in cross-cultural research or is simply not suitable for explaining cultural differences. Considering the results of prior research are mixed, it is not possible to rule out that studies relying solely on Hofstede-Gray generalizations to characterize individuals based on nationality are misleading and stereotyping at the individual level (McSweeney, 2016). Thus, our study complements the critical literature on cultural dimension models (Baskerville, 2003; Baskerville-Morley, 2005; Devinney & Hohberger, 2017; Kirkman et al., 2017; McSweeney, 2016). It is necessary to consider culture when examining accounting judgments and control for it in studies that include information related to firms from different countries. This issue is vital as evidence demonstrates that national patterns in the use of IFRS continued after the 2005 adoption of IFRS in the EU (Guermazi & Halioui, 2020; Kvaal & Nobes, 2012), and it also clearly shows that the benefits of IFRS adoption were unevenly distributed among countries due to differences in institutions that operate in different cultures (Andre, 2017; Gordon et al., 2019). However, alternative approaches for cross-cultural research based on non-value aspects of culture could potentially explain the differences in accounting judgments better (Baskerville, 2003; Baskerville-Morley, 2005; Devinney & Hohberger, 2017; Kirkman et al., 2017; McSweeney, 2016). As researchers (Kirkman et al., 2017) have noted, “there is a long history of rich and comprehensive research in a variety of disciplines that has successfully incorporated other approaches to culture, including cultural orientations (Kluckhohn, 1954) and styles (Hall, 1976)” (p. 20), which holds potential for this strand of research. Moreover, “Homo Sovieticus” from the perspective of post-socialist anthropology can be an interesting empirical category. This issue seems to be vital as the favorable view of capitalism that was present in Poland during the 1990s had declined by 2015, which may have roots in the ideology spread during their communist era (Newland, 2018).

Additionally, our data analysis revealed that in two cases the variable NATIONALITY was significant, however, the direction of the differences was not as expected. Moreover, some professional characteristics of accountants that were controlled for in our study, such as work experience and statutory auditor or chartered accountant qualification, affected their judgments. This leads to the conclusion that national attributes prevailing in each country such as behavioral, economic, political, firm-specific, and market factors are also relevant to explain differences in accounting judgment (Reisch, 2021; Yen et al., 2017). Other professional characteristics not controlled for in our study could also form the basis of future work in this area.

This study is subject to some limitations that are worth noting. First, our results may be biased by the fact that most of the respondents were statutory auditors/chartered accountants. However, in both the British and Polish samples, the position held did not affect the accounting judgments made by respondents, which reduces the probability of such a bias in our results.Future research should extend the investigation to financial statements preparers, who may take a different view of accounting decisions than statutory auditors/chartered accountants. Another limitation of our study that needs to be recognized is the small sample size of British accounting professionals, which may have limited the generalizability of the results. Finally, the use of a number of different IFRS excerpts may have resulted in participants not being aware of the practical outcomes of their accounting judgments, especially if they do not frequently apply them at work. Hence, their decisions might have been different if they had been made in real-life work situations.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Polish National Science Centre [grant number DEC-2012/07/B/HS4/03012].

Notes on contributors

Jerzy Gierusz

Jerzy Gierusz: Professor of Accounting. Head of the Department of Accounting at the Faculty of Management of the University of Gdansk. Member of the Polish Accounting Standards Committee and the Scientific Council of the Association of Accountants in Poland. In previous years representative of Poland as a member of the authorities of the EAA.

Katarzyna Kolesnik

Katarzyna Kolesnik: Doctor of Economics (Ph.D.) in Management. Assistant Professor of Accounting at the Faculty of Management of the University of Gdansk. Researcher and financial accounting lecturer. Statutory Auditor.

Sylwia Silska-Gembka

Sylwia Silska-Gembka: Doctor of Economics (Ph.D.) in Management. Associate Professor of Accounting at the Faculty of Management of the University of Gdansk. Researcher and financial accounting lecturer.

Anna Zamojska

Anna Zamojska: Habilitatus Doctor of Philosophy (Ph.D.) in Economics and Finance. Till 1992 researcher associated with Department of Econometrics of the University of Gdansk, from 2013 Associate Professor. Member of many research projects. A specialist with extensive knowledge in the field of quantitative methods, modelling, forecasting and simulations applicable in the corporate practice.

Notes

1. Comparative analytical review for some related papers is displayed in Appendix 1.

2. The term “homo sovieticus” was introduced at the end of the 1980s by the Russian logician, philosopher, and writer, Alexander Zinoviev (Banach, 2014).

3. Over the past twenty or so years, evolution in the intensity of individual dimensions has been observed in Polish culture. These dimensions are most visible in the decreasing level of power distance in Polish society. The changes, however, are slow (Wolniak, 2012).