Do multinational corporations pay their "Fair Share"?

1.   Introduction and objective of the paper

Corporate Tax Avoidance (CTA) of Multinational Corporations (MNC) consists of shifting global profit into countries with a lower effective tax rate than in the corporate headquarters' home country. It encompasses "all arrangements to reduce, eliminate, or defer a tax liability" through legal means (Freedman, 2004). It takes advantage of the technicalities and incongruities between two or more legislations in the international tax system for the purpose of significantly reducing tax obligations (European Commission, 2012; Ramboll Management Consulting and Corit Advisory, 2015). In academic literature, it is commonly referred to as an aggressive tax strategy (Hardeck et al., 2019; Hardeck & Hertl, 2014; Nebus, 2019). From a corporate perspective, CTA supposedly benefits available corporate funds, improves shareholder value, and mitigates competitive disadvantages (Khurana & Moser, 2013).

From a societal perspective, CTA is criticized for exploiting legal loopholes, and is often classified as a violation of fair burden-sharing, which is needed to sustain and develop public and governmental services. The OECD estimates the loss for governments on account of Base Erosion and Profit Shifting (BEPS) of up to US $650 billion each year, which is approximately the GDP of Turkey or Switzerland (Crivelli, 2016; Crivelli & Gupta, 2016; IMF, 2015). As a result, taxpaying citizens carry a disproportionate tax burden, because they cannot reap the benefits of country-specific legislation under the international tax system. Hence, many citizens claim that "tax codes are often written [in ways] that offer 'a way out' of paying taxes for some but not all constituents" (Payne & Raiborn, 2018). Practically, this can be emphasized by MNCs like Amazon, Delivery Hero, or Netflix, who benefitted lately from consumers' reduced mobility and, consequently, their more sedentary lifestyles and lengthier screen times during the Covid-19 pandemic. On the contrary, many citizens suffered tremendously from curfews, store and production plant closures, and reduced working hours. Nevertheless, these citizens must maintain their obligation to pay taxes while various internationally active corporations continue to strategically minimize their tax payments (OECD, 2015; Hufbauer, 2021). The problem is that "people are [especially] more concerned during difficult economic times […] that everyone, particularly those with the most, are paying their fair share" (PricewaterhouseCoopers, 2013). Therefore, CTA may generally, and especially during crises, create a perception of unfairness among citizens, a feeling of inequity for many individuals, and challenges the idea of fair societal burden sharing (Foss et al., 2019; McGaughey & Raimondos, 2019; Nebus, 2019).

The awareness of corporate tax fairness, as well as the public attention to fairness in the international tax system generally, has increased significantly in recent years (Oats & Tuck, 2019), e.g. through media coverage, through non-governmental organizations and consulting firms, and by suggestions for improvements by political institutions like the OECD or the EU who released documents intending to fight global tax avoidance (OECD, 2015; PKF International Limited, 2019; IMF, 2015; Bazel & Mintz, 2020). Additionally, the BRIC countries, the UK, and the US put a large amount of emphasis on making "sure corporate [entities] just pay their fair share" (Biden, 2021). Generally, fairness in taxation was analyzed by a plethora of scholars, primarily from a corporate, institutional, and legislative perspective (Akamah et al., 2018; Bird & Davis-Nozemack, 2018; Devereux & Vella, 2014; Foss et al., 2019; Nebus, 2019; Ting & Gray, 2019). Applying CTA was found to produce several negative consequences for the respective corporation, such as decreased reputation and CSR perceptions or reduced turnover (Asay et al., 2018; Foss et al., 2019; Hanlon & Slemrod, 2009; Hardeck et al., 2019; Hardeck & Hertl, 2014; McGaughey & Raimondos, 2019; Nebus, 2019).

Nevertheless, from a citizens' income taxpayer perspective, the perceptions of fairness in the international tax system, as well as respective consequences of CTA, remain largely unexplored (first approaches by DeZoort et al., 2018; Hillenbrand et al., 2019). This is surprising since income taxpayers are the largest taxpaying entity, contributing significantly more to the state budget than corporate taxes and all other sources of taxation combined (Oxfam, 2014). Therefore, it is relevant to analyze and understand the perception of individual citizens regarding questionable business practices, and particularly what consequences may arise from a perception of (in)equity. This topic is largely unexplored and highly relevant from a social, institutional and managerial perspective

Theoretically, Andreoni et al. (1998) identify three main factors that are relevant for the individual fairness perception towards taxation: (1) the degree of satisfaction concerning the tax system as well as the provision of public goods and services, (2) the taxpayer's perception of fair burden sharing, and (3) moral rules, values and sentiments. Firstly, this means that individual citizens evaluate their own satisfaction with the prevalent (tax) system as well as what they receive for their tax payment, such as quality of public services (Dyreng et al., 2016; Frey & Torgler, 2007). Secondly, a social comparison with other entities, such as co-workers, or corporations, is relevant and needed in order to evaluate whether their own input of tax payments is fair compared to other tax-paying entities. In this context, the concept of reciprocity argues that an individual (taxpayer) is strongly influenced by his/her perception of the behavior of others, such as corporations, institutions, or other citizens (Alm & Ortega, 2006, Fehr & Falk, 2002; Rabin, 1998; Ortega et al., 2016). Thirdly, "tax arbitrage considerations can create distortions […] that are particularly salient in the context of international business" (Foss et al., 2019) since CTA is a phenomenon that is globally-spanning and influenced by local legislation, culture and the respective perceptions of legality (Bolton et al., 2010; Brockner et al., 2001; Lind & Earley, 1992). Therefore, cultural-dependent values are relevant, and research advocates for the inclusion of those into the analysis of the subjective evaluation of fairness (Jordan et al., 2021; Reinecke, 2010). Interestingly, research does not offer any conclusions which of these three dimensions is considered as predominant towards the evaluation of external entities. Evaluating a similar entitiy (e.g., a co-worker) is well established whereas the view on other entities lacks a theoretical understanding. The same applies to the consequences of such evaluations since those are only determined for intra-human, same-entity relationships (e.g., withdrawal of cooperation of a co-worker). Hence, based on these three aforementioned fairness and equity determinants, it is theoretically postulated that individuals may turn to equity restoration techniques in case they feel unfairly treated, or perceive inequity (Colquitt et al., 2006; Leventhal, 1976; Adams, 2015; Adams & Freedman, 1976). Detrimental behavioral and emotional consequences could be the results (Colquitt et al., 2006; Leventhal, 1976; Ting & Gray, 2019). For example, in the context of the international tax system, an increased citizens' tax avoidant behavior, societal turmoil, loss of trust and legitimacy towards institutions and/or MNCs, cognitive distortions, and intentions to leave the system or state could jeopardize our institutions, our democratic system, and social cohesion (Bird & Davis-Nozemack, 2018; Lenz, 2020). An empirical analysis of these possible detrimental consequences of inequity could support academia, practice, law-making, and institutions to improve fairness in the international tax system and thereby prevent, or at least, mitigate these highly society-threatening consequences. Therefore, our study aims to answer the following research questions:

RQ1: How does the citizens' satisfaction with the international tax system, social comparison with other tax-paying entities, as well as cultural values influence the perception of (in)equity in the international tax system?

RQ2: How does citizens' perception of (in)equity in the international tax system relate to possible detrimental behavioral and emotional consequences?

Our paper addresses a controversial and topical subject area in international business and business ethics research by relying upon equity theory and the concept of fairness (Akamah et al., 2018; Farrar et al., 2020; Foss et al., 2019; Nebus, 2019). Methodologically, we replicate the theoretical mediation effect postulated in equity theory by conducting a Structural Equation Modelling (SEM) technique combined with a Confirmatory Factor Analysis (CFA). Generally, we find that the perception of inequity is prevalent among the 218 participants of our survey and that fair burden-sharing is perceived to be non-existent.

We contribute to current research and theory in multiple ways. First, we classify the social comparison determinant as most relevant for the fairness perceptions among individuals towards questionable business practices. Second, we emphasize that CTA needs to be considered a possible legitimacy threat to societal and institutional functioning since detrimental consequences were confirmed. For example, CTA increases citizens' tax avoidant behavior, jeopardizes social cohesion, and undermines institutional democratic functioning. Third, the cultural values of power distance and masculinity were found to mitigate the aforementioned detrimental consequences of CTA, emphasizing that corporate tax activities seem to be influenced by cultural values and differences.

Our practical and institutional implications put great emphasis on further promoting fairness within the international tax system. We quantify the value for a fairly perceived corporate tax rate at 29% which is (a) significantly above the tax rate of tax avoidant corporations (<1%–17%), and (a) almost twice as high as the suggested global effective tax rate (15%; Biden, 2021), which challenges his recent suggestion.

2.   Theoretical framework & hypotheses

Fairness is an increasingly important criterion in international business, corporate ethics, and corporate communication, as bilateral or multilateral transactions are not merely driven by economic motives. Instead, they are significantly influenced by concerns about fairness, which ultimately affects the choice and behavior of agents (Rabin, 1998). As emphasized in the organizational fairness and equity literature, the perceptions of fairness are vital, as they affect individuals' willingness to view authorities and entities as legitimate and trustworthy. Fairness perceptions have the ability to discourage retaliatory behavior and encourage cooperative behavior (Colquitt et al., 2020; Colquitt et al., 2006; Lind & Earley, 1992). The concept of fairness has been applied to domains like price fairness, wage fairness, and tax fairness (Campbell, 1999; Falk et al., 2006; Farrar et al., 2019; Farrar et al., 2020; Frey & Torgler, 2007; Gielissen et al., 2008). Generally, tax fairness concerns the perceived balance between taxes paid and the respective public goods received, as well as the perceived procedural justice and the consequences of violations (Wenzel, 2002). This definition of tax fairness is in line with the terminology of distributional and procedural fairness (Farrar et al., 2020). The first is focused on the outcome of an event or a transaction, whereas the latter evaluates the process itself (Jordan et al., 2021; Tyler & Folger, 2010). However, when analyzing the international tax system consisting of multiple entities (individuals, corporations, and states), the interaction effects between entities are characterized as dominant emphasizing a distributional fairness understanding (Wenzel, 2002). This means that all entities are primarily comparing their own inputs to and outcomes from the system with those of others. It is argued that a perception of distributional fairness in the international tax system (1) supports and promotes a sustainable business environment and economy, (2) minimizes the social gap among people and between classes, (3) is essential to ensure sustainable government revenues, and (4) ensures social justice, social cohesion, as well as overall taxpayer compliance (Braithwaite, 2002; Hartner-Tiefenthaler et al., 2012; Lenz, 2020; OECD, 2015; Setyonugroho & Sardjono, 2012). Hence, practice and academia state the necessity that the two main income tax-payers (corporations and citizens) take part in fair burden-sharing (Akamah et al., 2018; Devereux & Vella, 2014; Foss et al., 2019; Nebus, 2019; Ting & Gray, 2019). Research on tax fairness has traditionally relied upon theories, models, and measures often based on organizational fairness construct, although organizational researchers have stressed the importance of a subjective, individual context to fairness perceptions (Doherty & Wolak, 2012; Folger, 2001; Jordan et al., 2021; Konow, 2001; Reinecke, 2010; Steiner & Gilliland, 2001; Tyler & Folger, 2010). A theoretical framework that determines whether this individual perception of fair burden-sharing and hence fairness of distribution exists, is equity theory (Adams, 2015; Folger, 1986; Leventhal, 1976). It posits that individuals evaluate their fairness/equity perception by comparing the ratio of their own inputs to the outcomes from a relationship (equity norm) with the same input/outcome ratio of others (social comparison). Value-based cognition can influence this perception of a situation since different values or culturally-embedded norms may legitimize different events, actions, or situations. Therefore, equity theory allows us to determine individually subjective fairness perceptions in the distribution of resources and obligations between relational partners or comparative entities. Ultimately, the theory posits possible consequences of perceived (in)equity (Adams, 2015; Adams & Freedman, 1976). If differences in the input-output ratio of the individual and a comparative entity are detected, the theory postulates detrimental emotional or behavioral consequences. Individuals address this through so-called equity restoration techniques, which attempt to establish an appropriately equitable level by either altering inputs or outcomes (done via behavioral changes) or by cognitively distorting them to achieve the same results (via emotional distortions). Hence, inequity serves as a motivating force to eliminate or reduce this cognitive dissonance in the form of emotional consequences, e.g., anger or guilt, or behavioral consequences, e.g., reduced transactions or opposition. A visualization of these causalities in equity theory can be found in Figure 1, which is used hereafter to derive our hypotheses.

Equity theory postulates that the determinant of equity norm influences the perception of fairness. This means that individuals may perceive their own situation as more or less satisfactory based on their inputs and outputs. In every exchange situation, an individual evaluates how much it invests and how much it receives, determining its own perception of fairness or inequity. In our tax context, individual citizens evaluate how satisfied they are with the overall societal support system provided by the government (public services, governmental support), and especially, whether they perceive their own tax obligation as justified (income tax payment). In this context, Hashimzade et al. (2013) highlight the influence of government services and publicly provided goods. If these are of poor (good) quality, taxpayers may perceive tax payments as unfair (fair). Dowling (2014) and Frey & Torgler (2007) emphasize that improved institutional quality is linked to higher tax compliance, as taxpayers may be motivated to pay their fair share of taxes to fund an efficient government. Moreover, laboratory experiments have shown that a higher return on public goods leads to increased overall satisfaction and tax compliance (Blackwell, 2007). In contrast, when these outputs are perceived as unsatisfactory or tax inputs are perceived as too high, citizens' perception of the tax system is dominated by inequity (Lee et al., 2019). This means, when citizens are (un)satisfied with their own situation of tax payments, the public services and the government, they may (not) be receptive to lower corporate tax rates since this would distort their fairness perception. Hence, CTA by MNCs will be considered as a violation of fair burden-sharing since an overall inequity perception towards the international tax system is prevalent. Therefore, our first hypothesis states:

H1: A (low) high citizens' equity norm satisfaction is positively related to a (higher) lower corporate tax rate that is perceived as fair.

Furthermore, equity theory argues that the determinant of a social comparison influences the perception of fairness. This means that the theory allows for a comparison of inputs and outputs of different entities. Fehr and Falk (2002), Rabin (1998), as well as Frey and Torgler (2007), propose the idea of reciprocity, also referred to as the conditional cooperation hypothesis (Ortega et al., 2016). It argues that individuals' willingness to cooperate, invest or to put effort into a relationship depends on other societal and organizational entities' behavior. In the context of the international tax system, an individual's willingness to pay (taxes) depends on other organizational and corporate entities' tax compliances. Hence, conditional cooperation is an important criterion for an individual taxpayer's perception who is strongly influenced by the behavior of other taxpayers, corporations, or institutions (Alm et al., 2006). If a citizen is (un)satisfied with their own input-output evaluation compared to other entities, the perception of fairness might be (less) more present. If a fair burden-sharing between the direct taxpaying entities of individuals and corporations is perceived to be (non-) existent, a lower corporate tax rate will be perceived as (un)fair. Hence, our second hypothesis postulates:

H2: A (low) high citizens' satisfaction with social comparisons is positively related to a (higher) lower corporate tax rate that is perceived as fair.

As a third determinant, equity theory argues that cognition influences the perception of fairness. Cognition is based on different thought processes that are guided by and rooted in our values and norms. The cognitive evaluation of a country-spanning system and/or behavior is especially influenced by cultural values and norms (Minkov et al., 2013). An internationally relevant behavior, action or entity may be perceived as more or less legitimate and fair among individuals with differing cultural values (Baudot et al., 2020; Kanagaretnam et al., 2018; Minkov et al., 2013; Hanlon, 2012). The understanding of a global concept like MNCs' taxation as well as the amount of a fair share can be influenced by subjectively different cultural value-based evaluations. Since cognitive evaluations should generally "consider the implications of cultural conditioning and socialization on […] fairness perceptions" (Conner, 2002), research advocates for the inclusion of cultural values into the analysis of fairness (Bolton et al., 2010; Brockner et al., 2001; Lind & Earley, 1992). In this regard, Hofstede's five dimensions facilitate the analysis of an internationally induced event, situation or behavior. We argue that fairness perceptions towards CTA are influenced by the acceptance of hierarchies, indifferences, and power-seeking behavior between individual citizens, between corporations, as well as by the interplay among these two entities. These values are emphasized in Hofstede's dimensions of power distance (PD) and masculinity (MAS). In the case that values for these two dimensions are higher, CTA may actually be accepted by citizens since "big relevant" corporations with their hierarchical dominance, their societal, institutional and economic power, as well as their profit-seeking activities are perceived as given and legitimate. In contrast, collective, society-oriented behavior, a focus on long-term prosperity, and a preference for economic certainty may significantly increase the citizens' relevance of "being equal" as a corporation by paying an appropriate amount of corporate tax, neglecting CTA. This can be explained by the importance of securing future welfare for all stakeholders and entities (and not only the corporation), as well as by reducing possible future economic uncertainties and societal shortcomings by relying upon fair burden-sharing between corporations and citizens. These perceptions can be found in Hofstede's dimensions of collectivism (COL), long-term orientation (LTO), and uncertainty avoidance (UA) (Hofstede, 1984). In summary, we argue that Hofstede's five cultural dimensions may lead to different quantifications of a fair corporate tax rate. Specifically, we argue that the acceptance of inequalities and hierarchies (PD) and a more performance-based culture (MAS) lead to a fair perception of lower corporate tax rates. In contrast, a collective view (COL), a future-oriented focus (LTO), as well as high uncertainty avoidance (UA) may result in higher values for corporate tax rates in order to be perceived as fair. Therefore, our third hypothesis postulates that:

H3a: High values of PD and MAS are positively related to a lower corporate tax rate that is perceived as fair.

H3b: High values of COL, LTO and UA are negatively related to a lower corporate tax rate that is perceived as fair.

Considering the postulated determinants of perceived fairness or inequity in H1–H3, equity theory argues that this respective perception can lead to supportive or detrimental behavioral and emotional consequences. Accepting the low corporate tax rate and CTA as given and legitimate may result in system-supportive behavior of individuals, such as tax compliance or support for these MNCs or the responsible government. Individuals may perceive the overall situation as indifferent or irrelevant and their behavior could be classified by detachment or apathy. However, when individuals only consider high effective corporate tax rates as fair, the prevailing system of CTA can cause detrimental behavioral and emotional consequences. This means if citizens perceive a lower corporate tax rate as (un)fair, they may (leave) support or emotionally (fight) legitimize the system. Hence, we argue that:

H4a: If a low corporate tax rate is perceived as (un)fair, (detrimental) supportive emotional consequences will prevail.

H4b: If a low corporate tax rate is perceived as (un)fair, (detrimental) supportive behavioral consequences will prevail.

Considering the mediation effect of fairness perceptions postulated in equity theory, our hypotheses H1 to H4 need to be included into these structural causalities. Equity theory argues that the individual fairness/inequity perception mediates the effect between equity theory determinants and possible behavioral as well as emotional consequences. In the context of fairness in the international tax system, a large difference between a corporate tax rate that is perceived as fair as well as a practically-applied corporate tax rate (CTA) can amplify detrimental consequences. Hence, we argue that the citizens' fair perception of a corporate tax rate mediates the relationship between equity norm, social comparison, and cultural values, with emotional as well as behavioral outcomes. The resulting hypothesis states that:

H5a: The perception of a low corporate tax rate mediates the relationship between citizens' equity norm satisfaction as well as emotional and behavioral consequences.

H5b: The perception of a low corporate tax rate mediates the relationship between the citizens' satisfaction with social comparisons as well as emotional and behavioral consequences.

H5c1: The perception of a low corporate tax rate mediates the relationship between PD and MAS as well as emotional and behavioral consequences.

H5c2: The perception of a low corporate tax rate mediates the relationship between COL, LTO and UA as well as emotional and behavioral consequences.

Our overall research framework, including the five hypotheses (H1–H5), can be found in Figure 2 below. Solid, continuous lines visualize direct effects, while dotted lines illustrate the mediation effects for our three H5 hypotheses.

3.   Methodology

3.1.   Method of data collection

Previous research argues that the role of the government as the tax collector, individual local tax obligations, satisfaction with the specific tax and political systems, or the role of MNCs in a specific country could be influential towards tax perceptions (Jordan et al., 2021; Payne & Raiborn, 2018; Reinecke, 2010). We conducted an anonymous survey addressing individuals globally in order to test our hypotheses. This approach allows us to control for these effects, and hence make our results more generalizable (Matute et al., 2020). Moreover, we argue that an anonymous survey is appropriate since various individuals perceive the topic of their own tax (and financial) situation as something personal and private and, hence, sensitivity and privacy have to be considered. Hence, alternative methods like personal interviews or experiments may result in a reduced willingness to participate. Generally, social desirability bias could be present in an open or personalized survey that deals primarily with fairness and ethics. With this in mind, Krumpal (2013) argues that anonymous online surveys are significantly better when trying to mitigate this bias compared to interviews, experiments, or non-anonymous surveys. Moreover, we perceive online surveys as the appropriate methodology since it is most suitable to get a glimpse of the fast-paced lifestyle of a big number of citizens, who are more motivated to respond to a short online survey compared to lengthy interviews or experiments that may also require physical attendance.

3.2.   Sample

The survey was conducted in February and March 2021 and was distributed via Social Networks like LinkedIn or Facebook, direct email invitations, and online forums. The survey was available in English as it was addressed to an international audience. An overall completion rate of 24% resulted in a sample of 246 respondents. We cleaned the data set for short answer times (below 10 minutes), a high proportion of blank answers (more than 3 in one construct or more than 5 overall) and contradicting answers for reverse-coded questions.

A total of 218 respondents, out of which 51% are female and 49% are male, were used for further analysis. 59% of the respondents are between 25–45 years old, and 72% have international experience due to one or multiple abroad stays. Despite very diverse citizenships from Asian, European, English-Speaking and Romance Countries, 75% of the respondents pay their taxes within the European Union and, hence, within a similar institutional setting. Our respondents are well educated with more than 85% holding a bachelor's degree or higher but were spread among a wide variety of jobs, industries, and sectors, with only 16.5% being students. Even though multiple studies "suggest a liberal political bias among online study participants" (DeZoort et al., 2018; Matute et al., 2020), we can mitigate this caveat since the political orientation of the respondents was equally distributed with 42.2% stating to be "independent", and 87.6% of respondents classifying themselves between "moderate left" and "moderate right". Extreme cases ("far-left", "far-right") in regard to the political orientation were only observable for less than 1.5% of the survey participants.

Moreover, regarding CTA awareness of our respondents, Asay et al. (2018) argue that "few consumers recall ever seeing a negative media article about taxes." DeZoort et al. (2018) add that "the general public's relatively limited tax knowledge […] suggest[s] the need for further education and clarity regarding corporate tax practices." However, we argue that this caveat can be mitigated since 89% of our respondents have (frequently) heard about CTA. Hence, the needed awareness for this equity topic has significantly increased in the last three years, and hence we can mitigate this boundary condition of previous studies.

Overall, we deem our sample as appropriate to answer our research questions for multiple reasons. First, the educated, and internationally-oriented respondents of our survey are often defined by a fast-paced lifestyle using online platforms like Amazon, Google Services, Facebook, or Delivery Hero that are criticized for their tax-avoidant practices. Although buying many goods and services online, using data-gathering social networks quite frequently, as well as putting a good deal of emphasis on orienting their purchasing behavior around stylish and status-enhancing brands, they also show an above-average interest in fair, sustainable, and responsible practices (of corporations). This is emphasized by their preferences for long-term (natural) prosperity, organic products, sustainable lifestyles, and strong voter support for environmentally- and socially-oriented parties. To a certain degree, this paradox between system-supportive and system-opposing behavior towards MNCs' practices makes their perception towards questionable business practices like CTA especially relevant and interesting.

3.3.   Measures

Our survey is based on established measures that were applied in fairness research. Whenever necessary, we adapted items and scales to the tax content since it is frequently emphasized that models and measures of fairness are context-dependent and need to be aligned to different situations accordingly (e.g., Colquitt et al., 2005; Colquitt et al., 2006; Farrar et al., 2020). In order to measure the determinants of equity norm and social comparison, we relied on measures and conceptual ideas of Van Ryzin (2011), Van De Walle (2018), Dittrich & Carrell (1979), Sweeney et al. (1990), Nebus (2019), and Farrar et al. (2020). The measurement of cognition corresponds to individual measures for cultural values (Yoo et al., 2011), which are based on Hofstede's five dimensions (26-item scale). The items for all three aforementioned determinants were measured on a five-point Likert scale ranging from 1 (I do not agree at all) to 5 (I fully agree). We also included the established scales of the Equity Sensitivity Instrument (EQSI) with a 10-item scale for robustness checks (Huseman et al., 1987).

CTA is often defined as either a responsible or an aggressive tax strategy (Asay et al., 2018; Hardeck et al., 2019; Hardeck & Hertl, 2014) without further quantifications. In order to overcome this limitation, our study derives measures from the OECD BEPS study such as profit rate differentials within global MNCs, tax rate differentials to non-MNCs, competitors' behavior, or tax haven activities (OECD, 2015). We rely upon the "six indicators of BEPS activity [that] highlight BEPS behaviours using different sources of data, employing different metrics, and examining different BEPS channels. When combined […], they confirm the existence of BEPS (OECD, 2015)." For the quantification of a value of a corporate tax rate that is perceived as fair, survey participants were asked to determine a percentage value on a scale from 0–100% based on the above mentioned six OECD indicators. In order to illustrate the idea that fairness is perceived in absolute and relative terms, the statements were phrased in absolute measures (e.g., profit-to-tax-rate-ratios in headquarters and subsidiaries) as well as in relative measures of fairness (e.g., tax rate differentials to non-MNCs or competitors). We tested the reliability of our CTA measure in a pre-test that provided satisfactory results. Moreover, we calculated Cronbach's α for this multi-item scale that consists of the six OECD indicators. The respective Cronbach's alpha is between 0.73 and 0.93 and, therefore, above the recommended threshold of 0.70 (Nunnally & Bernstein, 1995).

The emotional and behavioral consequences of CTA are related to equity theory and were conceptualized and operationalized with established measures of emotional spectra (Russell & Barrett, 1999; Russell & Carroll, 1999). Moreover, we included case-specific outcomes that are associated with possible supportive or detrimental emotional and behavioral consequences towards MNCs' CTA (e.g., emphasized by Farrar et al., 2020, and Nebus, 2019). All items and measures that were used in the questionnaire can be found in the Appendix.

3.4.   Reliability, validity and robustness checks

Before the survey began, we mitigated acquiescence bias by using balanced scales and reverse coded questions. After collecting the results, we checked the possibility of nonresponse bias by comparing the answers of early, average, and late respondents (Armstrong & Overton, 1977). An ANOVA revealed no significant differences. In order to avoid social desirability bias, we emphasized that there are no right or wrong answers and that tax avoidance can also have positive effects on multiple stakeholders. Moreover, the anonymity of the questionnaire was ensured, and statements—after discussions with colleagues—were written in a neutral tone that did not bias the participants towards a more or less ethically acceptable answer. We also checked for common method bias (CMB)/common method variance (CMV) to avoid statistical distortions stemming from biased instruments. We use Harman's (1976) single factor score, in which all self-reported scales (measuring latent variables) are loaded into one common factor. With all eigenvalues greater than 1.0, the 13 determined factors accounted for 87.20% of the total variance, while the largest factor only accounted for 12.15% of the variance. Hence, we deduce that CMB/CMV does not affect our data. Moreover, we argue, in line with Podsakoff et al. (1990) as well as Podsakoff et al. (2012), that the order of our questions significantly reduced the possibility for CMB since we separated the dependent and independent variables throughout the questionnaire which made it difficult for respondents to deduce the causal links. Moreover, we ran a common latent factor (CLF) analysis double-loading all relevant items on their respective latent construct as recommended by Podsakoff et al. (2003). The factor loadings remain significant after introducing the CLF with no significant improvements in the model (∆β < 0.2). Therefore, we argue that CMB/CMV is of no concern in our data set.

In order to ensure the reliability of our data, we ran several robustness tests. For example, different levels of tax affection, tax knowledge, or of the own income tax were compared. Moreover, the country of origin (specifically comparing low and high tax countries) and the international experiences of respondents were compared via an ANOVA. Moreover, demographic robustness checks were done for education, gender, age, political orientation and perception of globalization (Dittrich & Carrell, 1979; McGee, 2006; McGee et al., 2008; Van Ryzin, 2011). Overall, our sample yielded only three significant results for these robustness checks. First, politically left-wing oriented individuals had significantly higher (p = 0.05**) expectations of a fair corporate tax rate than the average—despite a low increase in effect strength of only β = 0.9 pp. Second, the perception of globalization influences our findings to the degree that advocates of international trade and business accepted significantly lower corporate tax rates (β = −1.2%; p = 0.01***). Third, with regard to the Equity Sensitivity Instrument (EQSI), we could confirm that this construct improved the individual equity norm evaluation in a way that Benevolents' evaluations were more positive than those of Sensitives and/or Entitleds, which is theoretically sound (Huseman et al., 1987). No significant effects on the mediating or the outcome factors were determined.

3.5.   Method of data analysis

We applied Structural Equation Modelling (SEM) and a Confirmatory Factor Analysis (CFA) to analyze our data. We relied upon the approach of Baron and Kenny (1986) since missing values undermined the idea of using a bootstrapping procedure (Cheung & Lau, 2008). We argue that this approach is applicable, as we intend to empirically confirm an established mediation model via hypotheses-testing. We used the Unipark software to collect the data; and conducted the data cleaning, the computation of descriptive statistics, robustness checks and reliability measures in IBM SPSS 26. Ultimately, we used IBM SPSS AMOS 26 to test our postulated mediation effects. In AMOS, we used reflective variables which were highly correlated (Hulland, 1999) and latent constructs since most of our determinants in equity theory cannot be directly measured. We analyzed the convergent validity of our constructs and all standardized factor loadings are above 0.55 (Fornell & Larcker, 1981; Hair et al., 2019; Hair et al., 2017), except of the cultural value of masculinity as well as one item in the social comparison construct. We checked for multicollinearity since correlations between social comparison and cognition as well as behavioral and emotional consequences were found (the latter is addressed in the discussion section). However, the variance inflation factors were all below 8.5 and, hence, below the suggested maximum of 10 (Flury et al., 1988; Hair et al., 2019; Hair et al., 2017). Therefore, multicollinearity concerns could be put aside.

Our structural model for AMOS consists of six latent constructs: equity norm, social comparison, cultural value-based cognition, fairness perception of CTA, emotional consequences, and behavioral consequences. All latent constructs were measured with the inclusion of an error term. The hypothesized model provides a good fit for the data. Using the chi-square (x^2) test, which is recommended for moderate-sized samples (our number of distinct sample moments are 209; Tabachnick & Fidell, 1996), the x^2 value for the model is good with 242.86. The degrees of freedom (df) are 143 (p = 0.000***). The x^2/df fit index is 1.70 and lies within the recommended interval between 1 and 3 (Hair et al., 2017). We also calculated a confirmatory fit index (CFI) of 0.925, which is acceptable (Byrne & van de Vijver, 2010; Hu & Bentler, 1999; Jöreskog & Sörbom, 1996). The Goodness of Fit Index (GFI) had a value of 0.90 and was, therefore, sufficient (Flury et al., 1988; Hair et al., 2019). The RMSEA fit indicator had a value of 0.057, which is slightly but not significantly above the recommended threshold (<0.05; Hu & Bentler, 1999). Moreover, we compared the model fit of our 6-construct model with alternative models, as recommended by Flury et al. (1988) and Hair et al. (2017). Comparing chi-square values and fit indices, our postulated and applied model was deemed superior since it was, unlike other tested models, theoretically sound.

4.   Results

4.1.   Descriptive results

Table 1 provides an overview of the descriptive statistics of our study design. The table depicts the mean values as well as the standard deviations for the equity theory determinants, the six OECD indicators that determine a fair corporate tax rate, and emotional and behavioral consequences.

Our results reveal that overall our participants are satisfied with their financial and tax situation (means of 3.57 and 3.11) as well as with the public services and the state (means of 3.23 and 3.22). However, their satisfaction with regard to the prevalent tax system is significantly lower, with a mean of 2.83 out of 5. Moreover, the difference between the salary before and after taxes is perceived as very high (m = 4.03). Individuals seem to perceive a disadvantage in the amount of taxes that need to be paid by individuals compared to other tax-paying entities. Especially MNCs are perceived to have greater possibilities to reduce their tax obligations (m = 4.43). Overall, citizens perceive themselves to have a disadvantage towards taxation (m = 4.08) and do not agree with MNCs providing for society (m = 2.02). Our sample is defined by average cultural values that tend towards a low PD (m = 1.71) and a mediocre MAS (m = 2.23), whereas COL, LTO, and UA have rather high mean values of 3.19, 3.78, and 3.36 respectively. The corporate tax rate that is perceived as fair ranges between 24.14% and 39.97% for the six different items derived from the OECD BEPS study. Generally, the average fair corporate tax rate is 29.49% when computing the multi-item scale. This is confirmed by our item "Fair Corporate Tax Rate robustness statement" that participants evaluated with a mean of 29.29% which does not significantly differ from the average computation. Generally, these values are an important descriptive finding that will be addressed in our discussion section. The emotional and behavioral consequences showed differentiated results. On the one hand, emotional outcomes are focused around means of 2.3 up to 3.0, showing on average low emotional arousal towards supportive or detrimental consequences in terms of CTA. On the other hand, behavioral outcomes show more differentiated positive and negative consequences. For example, the agreement to legally reduce one's own tax burden has a mean value of 4.44, emphasizing a high tendency towards possible citizens' income tax avoidance. Percentage-wise, 85% of our survey respondents would (fully) agree to reduce their tax burden if they have the ability to do so. Also, the behavioral consequences of "not buying anymore from a tax-avoidant corporation" (m = 2.74), "initiate change in the (tax) system" (m = 3.28), "rebel against the system" (m = 2.77), and "a rejection of legitimatization of institutions" (m = 2.74) yielded above-average results. The majority of participants only disagreed with the statement of putting "less effort into society" due to CTA (m = 2.06).

4.2.   Test of hypotheses

Our postulated structural model yielded differentiated results (see Figure 3). The effect strength between equity norm and the corporate tax rate that is perceived as fair was positive but very low with β = 0.07, and not significant. Hence, we cannot support H1, which stated that a (low) high citizens' equity norm satisfaction is positively related to a (higher) lower corporate tax rate that is perceived as fair.

H2 posits a link between the social comparison determinant of tax paying individuals with other tax-paying entities and was moderately and negatively linked to our corporate tax fairness factor with β = −0.39 on a highly significant level of p ≤ 0.01***. This means that the higher the citizens' satisfaction with social comparisons is, the lower a corporate tax rate can be to be perceived as fair. Therefore, H2 is supported.

Hypotheses 3a and 3b postulate links between Hofstede's dimensions and our corporate tax rate that is perceived as fair. The hypotheses argued that high values of PD and MAS (COL, LTO and UA) are positively (negatively) related to a lower corporate tax rate that is perceived as fair. However, the five cultural dimensions were not significantly linked to our corporate tax rate, and effect strengths were low. Hence, overall, H3 cannot be supported.

Hypotheses 4a and 4b posit direct effects between the corporate tax rate that is perceived as fair as well as emotional and behavioral consequences. The effect strength between the corporate tax rate that is perceived as fair and the emotional consequences in H4a is weak but positive with β = 0.06 on a p ≤ 0.05** significance level. This contradicts our hypothesis 4a since, despite the absence of a fair high corporate tax rate, supportive and positive emotional consequences prevail. This has interesting implications for our discussion section. In regard to our hypothesis 4b, we detect a negative effect of the corporate tax rate that is perceived as fair and behavioral outcomes with β = −0.33 on a p ≤ 0.01*** significance level. This means that the absence of a fair high corporate tax rate, and hence a prevailing low corporate tax rate (CTA), relates to detrimental behavioral consequences. Hence, we can support H4b.

Ultimately, H5 emphasizes the postulated mediation effect of equity theory via the (corporate tax rate) fairness perception construct. H5a states that the perception of a low corporate tax rate mediates the relationship between citizens' equity norm satisfaction as well as emotional and behavioral consequences. Similar to the equity norm construct earlier, our mediated effect was also insignificant, albeit with a stronger effect, for both consequences (β(emotional) = 0.08, and β(behavioral) = 0.08). However, H5a cannot be supported.

H5b posits that the perception of a low corporate tax rate mediates the relationship between the citizens' satisfaction with social comparisons as well as emotional and behavioral consequences. Effect strengths are positive, significant, and moderate (β = 0.10 for emotional outcomes, β = 0.27 for behavioral outcomes, both on a p ≤ 0.05** level). This means that a higher citizens' satisfaction with social comparisons leads to rather positive and supportive consequences, mediated by the amount of corporate tax rate that is perceived as fair in a way that the fair perception of a lower corporate tax rate strengthens the positive consequences. This also means that the opposite effect takes place when the social comparison is defined by dissatisfaction, and only a high corporate tax rate is perceived as fair. In this case, detrimental consequences for emotional and behavioral outcomes can be the result. Ultimately, we can support H5b.

Lastly, H5c1 and H5c2 postulate that the perception of a low corporate tax rate mediates the relationship between Hofstede's five cultural dimensions as well as emotional and behavioral consequences. For H5c2, effect strengths were weak, and mediated effects insignificant, and hence H5c2 cannot be supported. For H5c1, however, the cultural dimensions of power distance and masculinity have a significant positive mediated effect on emotional (β = 0.15, p ≤ 0.05**) and behavioral consequences (β = 0.20, p ≤ 0.05**). This means that PD and MAS can reduce the detrimental consequences of CTA via the mediation effect. Overall, H5c1 can be supported.

5.   Discussion

Overall, our findings partially support the postulated causalities that are provided by equity theory and that were tested in the context of fairness in the international tax system and towards questionable business practices. Addressing the equity theory determinants, we only find a direct, significant influence of the social comparison determinant (H2***) towards the citizens' perception of corporate tax fairness. A lower corporate tax rate is only perceived as fair, when citizens do not feel to be at a disadvantage compared to other individuals, SMEs or MNCs. However, in the instance that they do feel an inequity, they request a significantly higher corporate tax rate in order to classify it as fair. Since H2 was confirmed, we argue that the citizens' satisfaction with social comparisons is highly relevant for understanding the public's perception towards questionable business practices (in our case CTA).

Our non-supported H1 emphasized the influence of the equity norm determinant on a corporate tax rate that is perceived as fair. The latter had a mean value of 29% and helps explain why the equity norm did not yield significant results. The value of 29% is significantly higher than the corporate tax rate in most OECD countries and worldwide (worldwide average 24%; Europe 20%), much higher than the recently discussed global tax rate of 15%, and particularly higher than any tax rate that is applied by tax avoidant corporations (<1%–17%). We argue that equity norm is generally of particular relevance for any kind of fairness evaluation, albeit the determinant of equity norm in the context of questionable business practices (in our case CTA) may be less influential as theoretically expected. There are tremendous differences between corporate tax rates that are perceived as fair by citizens and that are actually paid by corporations. The corporate tax rate also stands in stark contrast to what percentages of income tax rates individuals are obliged to pay. This may lead to individual cognitive distortions and mental distancing since individuals may psychologically fully decouple their own (tax) satisfaction from the corporate (tax) situation. They may perceive their "little" income (tax) as well as the output they receive (quality of public and governmental services) to be not connected to these "big, international corporate players" that are composed of millions of individuals being part of different stakeholder groups. Considering that a few MNCs are bigger than nation-states (e.g., Walmart vs. Belgium), cognitive distortions may arise and hence individuals may willingly decouple the link between individual and corporate factors (Matute et al., 2020). Ultimately, they may perceive their own tax and financial situation as not anyhow correlated with those of corporations, and hence this may explain our insignificant findings for equity norm. Overall, this means that the equity theory determinant of an individual's equity norm does not serve as a reliable and indicative measure to determine the fairness perception towards questionable business practices.

Concerning the cultural value determinant, H3 was not supported. Perceptions of questionable business practices may be influenced by cultural values (Ringov & Zollo, 2007; Williams & Zinkin, 2008), but also by cultural conditioning and socialization through media, institutions or other big interest or lobbying groups. Especially when dealing with such an abstract, corporate-focused, and individual-distant topic (Conner, 2002), we argue that cultural conditioning and socialization could outweigh the influence of cultural values when trying to determine fairness perceptions towards questionable business practices. Even if our robustness checks for the institutional framework did not reveal any significant differences, citizens may be influenced by some implicit institutional values that may shape their perceptions of events, situations or phenomena (Bird & Davis-Nozemack, 2018) —regardless of their citizenship or individual cultural values. Even if the willingness of "consumers to punish firms for bad behaviour varies in ways that appear to relate closely to the cultural characteristics" (Williams & Zinkin, 2008), most findings do not evaluate the possible effects and specificities of institutional-based influences on the creation of a perception of situations or events. Research normally connects cultural values to individual behaviors and preferences, but these cultural values seem to be less explanatory towards questionable corporate (tax) behavior. Future research should shed light on this topic.

Next, our results showed that the citizens' perception of corporate tax fairness has a direct influence on behavioral consequences confirming hypothesis 4b. Our results are in line with Asay et al. (2018), stating that "consumers indicate a reduced willingness to purchase from companies with aggressive tax planning strategies." This means that if a low corporate tax rate is perceived as (un)fair, (detrimental) supportive behavioral consequences will prevail. Next to reduced purchase intentions, higher willingness to avoid own income tax, to take initiative to change or rebel against the system, and a rejection of legitimacy were also found.

However, our results contradict H4a since we found that a high corporate tax rate that is perceived as fair -and that is absent during CTA-, is linked to positive emotional consequences. Considering our results, individuals that claim for fair burden-sharing between corporations and individuals are actually positively emotionally triggered by CTA. Despite a rather weak effect size, this finding is significant. We argue that these individuals may be characterized by positive thinking, active participation in problems and feel motivated to resolve this inequity. This proactive behavior of individuals may take some cognitive burden from them. This means that they feel better and more emotionally stable, freed, and relieved since they actively did something against inequity and unfair (corporate) behavior. Moreover, our interpretation of this behavioral change is in line with the confirmation of H4b (see in the paragraph above) since a weak, significantly negative correlation between behavioral and emotional consequences was detected. Furthermore, it can be argued that a rather financially driven and individually-distant topic like CTA may not necessarily trigger any negative emotions. We argue that emotional arousal is found somewhat more so when corporate activity harms human life, deals with animal cruelty, or destroys our habitat. However, questionable business practices that are more financial, specific and, often considered dry in nature (in our case CTA), do not seem to imply any negative emotional consequences (H4a), but only negative behavioral consequences of citizens (H4b). Future studies shall analyze whether these results hold true for other questionable business practices such as the aforementioned animal cruelty.

Ultimately, our mediation analysis supports H5b stating that the fair perception of a lower corporate tax rate mediates the positive relationship between citizens' social comparison evaluation as well as emotional and behavioral outcomes. On the one hand, this means that the causality explained earlier in H2 can be supported. While supportive, positive consequences can be seen as unproblematic, the citizens' dissatisfaction with social comparisons and the absence of a high corporate tax rate (which respondents perceive as fair) can result in detrimental, negative consequences. This applies to emotional ones like cognitive distortions, mental rebellion, anger or hatred towards the system, as well as to behavioral consequences like less support for MNCs, the society and governmental institutions. This can cause severely damaging effects for social cohesion or the government's intention to collect citizens' income tax, which 89% of our survey participants would reduce. Hence, politics, institutions, and corporations need to ensure that individuals do not join the "race to the bottom" by becoming more tax avoidant themselves due to a perceived inequity in the international tax system. This could inevitably hamper the cohabitation among citizens, corporations and the state and result in an absence of free (or at least affordable) public services and governmental support, as well as further undermining the already prevalent social deterioration.

Similar to the equity norm determinant, H5a could not be supported. Similarly to the argumentation in H1, we argue that the cognitive distortion and mental distance cannot be overcome by the (non-)perception of a corporate tax rate that is perceived as fair. In our case, the equity norm seems to be fully decoupled from a fair corporate tax rate perception and also does not relate to any emotional or behavioral consequences towards CTA (Matute et al., 2020). A possible explanation could be that CTA is a topic that is not directly related to one's own financial situation and ultimately only affects an individual in a long causal chain that reaches from the corporation over the government to the citizen as a user of public and governmental services. Albeit equity norm is a non-determining factor for the citizen's fairness evaluation towards questionable business practices, future studies shall evaluate whether this also holds true for corporate activities such as legal product fraud (sugar, chemicals, reduced package sizes etc.), that may directly affect multiple citizens.

H5c1 was supported since the cultural dimensions of PD as well as MAS had a significant, positive, and mediated effect on emotional outcomes (β = 0.15 p ≤ 0.05*) and behavioral outcomes (β = 0.20, p ≤ 0.05**). This is congruent with Ringov & Zollo (2007), claiming that "power distance and masculinity are found to have a significant negative effect" on corporate behavior in a way that these two cultural dimensions reduce ethical and fairness considerations. This means that individuals that are characterized by the acceptance of hierarchical, masculine and power-seeking values rather accept questionable business practices compared to individuals that score low on PD and MAS. The acceptance of questionable business practices creates more positive emotional and behavioral arousal on their side and hence strengthens the established and prevalent mechanisms in the international tax system (Williams & Zinkin, 2008). Corporations that apply questionable business practices may be perceived as dominant, powerful, value-driven, and profit-oriented by these individuals. They may consider these traits as desirable for corporations (and maybe even for themselves).

However, our results could not confirm a mediating effect of a corporate tax rate that is perceived as fair on the relationship between UA, COL, and LTO, as well as emotional and behavioral consequences. Effect strengths were low (β(emotional) = 0.01, and β(behavioral) = −0.01), and insignificant. We argue that the (unequal) power distribution between citizens and big international corporations is primarily captured within the dimensions of PD and MAS, but less or at least captured differently within the cultural values of UA, COL and LTO. Hence, a fairness comparison of differently shaped entities may be primarily determined by power asymmetries (PD), as well as by power-seeking and profit-oriented life or business philosophies (MAS). Our non-significant findings concerning the dimensions of UA, COL and LTO could also be influenced by the fact that these cultural dimensions do not relate well to the concept of CTA as a corporate activity. Similarly to our argumentation for H3, cultural conditioning and socialization through media, institutions or other big interest or lobbying groups could be more influential. Future studies should shed light on this delineation between institutional socialization and cultural values when analyzing questionable business practices from an individual's fairness perspective.

6.   Contributions, limitations and future research

Particularly after the Covid-19 pandemic, fairness in the international tax system is an increasingly discussed topic in politics, among citizens, and within international business research (Akamah et al., 2018; Biden, 2021; Foss et al., 2019; Hufbauer, 2021; Lei et al., 2021; Nebus, 2019; Ting & Gray, 2019). Therefore, this paper analyzed the citizens' perception of a corporate tax rate that is perceived as fair and whether this mediates the relationship between equity theory determinants as well as emotional and behavioral consequences. Drawing theoretically on the meditation effect postulated via equity theory, we methodologically executed a mediation analysis (SEM-CFA) on a sample of 218 survey participants. Our study is able to determine (1) the factors that quantify a corporate tax rate that is perceived as fair, (2) which emotional and behavioral consequences could result from a perception of inequity concerning CTA, (3) and ultimately, showed the mediating effect of the fairness perception of corporate tax rates on individuals' emotional and behavioral consequences.

More specifically, we find that citizens consider it unfair to achieve a lower tax rate by shifting corporate profits out of the country and that fair burden-sharing between corporations and individuals seems to be non-existent. Our participants perceive corporate tax avoidant behavior as negative—aligning well with DeZoort et al.'s (2018) statement that most "individuals […] adopt a social responsibility approach (i.e., executives should consider the ethicality of tax avoidance and its potential impact on the public trust)." We emphasized that the perception of the international tax system fairness is primarily driven by the social comparison perspective of individuals. Citizens evaluate how different tax-paying entities (citizens, SMEs, and MNCs) interact and whether these contribute their fair share towards the system (Dowling, 2014). This determines the citizens' perception of a fair corporate tax rate and ultimately, influences emotional and specifically behavioral consequences. In the case of inequity, these can be defined as counter-corporate and counter-government, and take the form of reduced sales, reduced legitimacy, less trust in institutions, rebellion against the system, or less social cohesion. Moreover, the increased intention to reduce one's own income tax burden was confirmed which could even accelerate the "race to the bottom" in regard to tax obligations. Surprisingly, emotional consequences are rather positive towards CTA meaning that this specific corporate behavior does not primarily trigger negative emotional arousal. Therefore, CTA may be classified as a topic that is relevant and ready for a change from the citizen's perspective, but consequences are of pure behavioral nature. CTA does generally not trigger emotional arousal such as other questionable corporate activities like pollution, animal cruelty, or human harm. Moreover, unlike our postulated hypotheses, the equity norm determinant of equity theory had no effect on the perception of questionable business practices, whereas the cultural dimensions of PD and MAS rather relate to an acceptance of CTA, and thereby to more positive emotional and behavioral consequences. We contribute to the understanding of equity theory by showing that the social comparison determinant of equity theory was found to be most influential in the formulation of fairness perceptions among individuals towards questionable corporate activities. Furthermore, we confirm that cultural cognition has an influence on fairness evaluations. Hence, research needs to especially include those two aspects into the analysis of fairness perceptions towards corporate behavior. Our analysis contributes to a growing body of research (Bierstaker et al., 2012; Cohen et al., 2007; Farrar et al., 2019; Antonetti & Anesa, 2017; Antonetti & Maklan, 2016) revealing the importance of "evaluating perceived fairness when studying individuals" and their perception of specific (corporate) situations and activities (DeZoort et al., 2018). Moreover, we emphasized the applicability of equity theory towards an individual fairness perception when dealing with different (tax paying) entities. Despite our ambiguous results, we argue that from a theoretical lens, not only organizational fairness approaches are applicable, but also theories that deal with an individual's distributive fairness perception.

As a political and institutional implication, we contribute to an updated understanding that CTA is a phenomenon, which, unlike a few years ago (Asay et al., 2018), is very well understood by citizens in 2021. Also, citizens have a clear opinion what an average fairly perceived corporate tax rate would be: 29%. Hence, institutions like the OECD or the European Union, as well as nation states like the US or Germany, need to reconsider whether the recently suggested 15% global effective corporate tax rate is sufficient. Recent announced tax system changes in the United States, the European Union, and in the OECD (pillar one and pillar two) may not be able to mitigate the perception of inequity in the international tax system in order to strengthen social cohesion. In case the global community cannot agree on an effective global corporate tax rate that is significantly higher than 15%, institutions can at least address the determinant of social comparison and ensure that citizens perceive this determinant as more fair and satisfactory. Our results have shown that this could significantly decrease the requested corporate tax rate and mitigate detrimental consequences. Practical approaches could be to initiate public tax education campaigns explaining in detail why tax rates (must) differ significantly among entities. Moreover, a more questionable approach could be that also MNCs voluntarily increase their social responsibility to compensate for their low tax rates to boost the social comparison determinant. This was proven to be successful (Col & Patel, 2019). Ultimately, society, citizens and NGOs could pressure corporations to meet their full local tax obligations by applying public shaming and public pressure. Similarly to the area of occupational safety and employee rights, these techniques raise the question of legitimacy of companies, put CTA into question, and ultimately may improve the social comparison determinant in case corporations drop CTA (Freedman, 2004; Panayi, 2015; Nebus, 2019, Dyreng et al., 2016; Graham et al., 2014).

Our study's implications are not free of caveats. One limitation is that we did not clearly separate between individuals and citizens that benefit from MNCs' CTA and those that are not. Research indicates that "accountants find it highly ethical to avoid taxes [whereas the] general public is more concerned with fairness than accountants" (DeZoort et al., 2018). The same holds true for citizens that also may be shareholders of a tax avoidant corporation (Blaufus et al., 2019). However, a robustness check for those individuals in our sample, that work in the finance and auditing sector, did not reveal any significant differences. However, we believe that future studies shall further elaborate on that which would also allow to further sharpen the construct of corporate taxation which is considered a simplification in our manuscript since it is only represented by a mere percentage number that does not differentiate between the complex and multifold nature of different taxation that apply to corporations (e.g., corporate income tax, capital gain or dividends tax, trade tax, value-added tax, taxation of property etc.).

Moreover, we believe that the topic of SMEs needs more emphasis since these differ significantly in their structure compared to an MNC. However, some of them (e.g., German Hidden Champions) are extremely international and could also leverage global tax advantages. In this case, the perception of individuals as income tax payers could be of significant relevance in order to determine whether the perception of tax-avoidant MNCs can be transferred to SMEs or not.

We would like to finish this study with the famous quote of Benjamin Franklin from 1789 stating that "in this world nothing can be said to be certain, except death and taxes." We argue, however, that the certainty of taxes can be influenced by MNCs, and that citizens do not perceive this as particularly fair. Politics, institutions, and corporations need to ensure that individuals do not join the "race to the bottom" by becoming more tax avoidant themselves due to a perceived inequity in the international tax system. This could inevitable hamper the cohabitation among citizens, corporations and the state, and result in significant worsening of public services and governmental support, and hence even further undermine the already detected social deterioration. This downward spiral needs to be stopped, e.g., by MNCs that voluntarily and proactively engage and participate in a "fair burden sharing" to mitigate social distortions generally - and especially after the recent Covid-19 pandemic.

Conflict of interest

The author declares no conflicts of interest in this paper.